High quality sulfur-doped titanium dioxide nanocatalysts with visible light photocatalytic activity from non-hydrolytic thermolysis synthesis

Abstract

A facile non-hydrolytic thermolysis route for monodisperse sulfur doped TiO₂ nanocatalysts is demonstrated. Compared with the as-obtained undoped TiO₂ nanocatalysts, the sulfur doped TiO₂ nanocatalysts present obvious enhanced visible light activation for the degradation of rhodamine B and methylene blue dyes under the artificial visible light (λ = 420–770 nm) irradiation.

---

Titanium dioxide (TiO₂) is the most widely used photocatalyst in environmental cleaning.^1–4 However, because of its large band gap (E_g = 3.20 eV for anatase), TiO₂ can solely be activated in the ultraviolet (UV) region. The limited UV-driven activity decreases its overall efficiency under natural sunlight irradiation, thus limiting its practical applications. Engineering the band gap of TiO₂ is one of the robust strategies for improving its photocatalytic efficiency, since it can availably extend the absorption from the UV to the visible-light region. Up to now, many attempts have been made in the field of visible-light-active TiO₂ by introducing various dopants into its crystal lattice, including metal⁵ and nonmetal elements.⁶ Although metal dopants can give the desired shift of the TiO₂ absorption from the UV to the visible light region, many of them also serve as recombination centers, which notably reduce the photocatalytic efficiency.^5a,7 Band-gap narrowing by doping nonmetals (such as N, C, S and F) into TiO₂ was recently found to be a more efficient way to yield photocatalysts with high catalytic activity under visible light irradiation.⁶ As an important sort of visible light activated doped TiO₂ photocatalysts, S-doped TiO₂ photocatalysts have been receiving extensive attention due to their high photocatalytic activity, structural stability and efficient band gap manipulation.^6c,8,12b

Various doping strategies have been adopted to achieve S-doped TiO₂ photocatalysts; while most of the S-doped TiO₂ photocatalysts were synthesized by oxidative annealing of TiS₂ or by treating the titanium contained precursors in a hydrogen sulfide (H₂S) atmosphere at high temperatures,^6c,9 besides the energy consumption and intricate setup, the high temperature treatment usually leads to low surface areas and photocatalytic activities due to the aggregation of catalysts. It is widely recognized that nanosized materials have opened the doors for discovering new properties and applications with respect to their macroscopic counterparts.¹⁰ In recent years, several methods have been developed to prepare S-doped TiO₂ nanocatalysts by a number of research groups. Examples include the catalyzed hydrolysis,¹¹ hydrothermal,¹² solvothermal synthesis,¹³ sol–gel,¹⁴ co-precipitation,¹⁵ ball milling,¹⁶ and supercritical fluid-assisted method.¹⁷

However, until now, a facile method for the synthesis of high quality (monodisperse, single crystalline, well shaped, and phase pure) dispersible colloidal S-doped TiO₂ nanocatalysts has not been demonstrated, which inspires the continuous and systematic exploration. In this paper, for the first time, we report a facile and effective non-hydrolytic thermolysis method for obtaining high quality S-doped TiO₂ nanocatalysts in high boiling solvents, and for in-depth investigation of the photocatalytic properties of as-obtained nanocatalysts. In comparison with the pure TiO₂ nanocatalysts, the absorption edge shifts to the visible light region on doping S into the TiO₂ crystal lattice, thus exhibiting an obvious visible light activation for the degradation of rhodamine B and methylene blue dyes.

The powder X-ray diffraction (PXRD) patterns of as-prepared pure TiO₂ and S-doped TiO₂ nanocrystals are shown in Fig. 1a. The as-synthesized samples exhibit the well-defined diffraction peaks, indicative of the high crystallinity of the sample. All the diffraction peaks are exclusively indexed to the anatase phase of TiO₂ (tetragonal structure; JCPDS card no. 21-1272, space group I₄₁/amd) for both pure and S-doped TiO₂ nanocrystals. No diffraction peaks from any other chemical species such as the rutile and brookite phase of TiO₂ are detectable. Compared with the pure TiO₂, a systematical peak shift towards low diffraction angles is unambiguously detected in the S-doped TiO₂ nanocrystals (Fig. S1†), suggesting a lattice expansion when introducing S into the TiO₂ crystal lattice. Using the least squares refinement of cell dimensions from PXRD data, the calculated crystal lattice parameters are as follows: a = b = 3.7852 Å, c = 9.5139 Å, V = 136.3127 ų for TiO₂ nanocrystals; a = b = 3.8008 Å, c = 9.5212 Å, V = 137.5440 ų for S-doped TiO₂ nanocrystals.

Fig. 1 (a) PXRD patterns of as-synthesized TiO₂ and S-doped TiO₂ nanocrystals. (b) Raman spectra of as-obtained pure TiO₂ and S-doped TiO₂ nanocrystals excited with Ar⁺ 514.5 nm, the inset shows the magnified region of the E_g Raman band at 147 cm⁻¹.

In order to obtain more information about the microstructure of as-obtained S-doped TiO₂ nanocrystals, their room temperature Raman spectra were measured. Group theory predicts that the typical anatase TiO₂ has six Raman active modes: Γ = A_1g + 2B_1g + 3E_g.¹⁸Fig. 1b displays the Raman spectrum of as-obtained undoped TiO₂ and S-doped TiO₂ nanocrystals excited with a 514.5 nm beam of the Ar⁺ laser. The Raman spectra of as-obtained nanocrystals present four bands at 147 cm⁻¹ (E_g), 392 cm⁻¹ (B_1g), 508 cm⁻¹ (A_1g) and 631 cm⁻¹ (E_g). All samples exhibit the characteristic Raman-active modes of the anatase TiO₂ phase, and show no detectable signals corresponding to any other titanium and sulfur containing species in the Raman spectra, demonstrating that the doped nanocrystals have the same structure as the anatase TiO₂, in agreement with the phase composition determined by PXRD. The inset of Fig. 1b shows the expanded views of the E_g Raman band at 147 cm⁻¹. It is found that the E_g band in undoped TiO₂ shifts to higher wavenumber when S doped, also indicating the incorporation of S into the TiO₂ crystal lattice.^11c,12b

Transmission electron microscopy (TEM) measurements shown in Fig. S2† reveal that the TiO₂ nanocrystals obtained from the thermo-decomposition of TiF₄ in OA/OM/ODE solvents without S doping present irregular morphologies with an average size ranging from 11.8 nm to 25.7 nm, while the S-doped TiO₂ nanocrystals are nearly monodisperse in the size of (14.5 ± 1.2) nm × (15.3 ± 1.1) nm (Fig. 2), indicating that S anions can act as a growth controlling agent to produce well-shaped nanocrystals in the current reaction system. However, the sole use of ODE in our synthesis can only yield irregular products (Fig. S3†). As demonstrated from the energy dispersive X-ray analysis (EDAX) (Fig. S4†), the atomic ratio of Ti : O is calculated to be 0.49 : 1 for undoped TiO₂ and 0.53 : 1 for S-doped TiO₂, which is in agreement with the stoichiometric ratio of TiO₂. The inset of Fig. 2 shows the representative high resolution transmission electron microscopy (HRTEM) image of S-doped TiO₂ nanocrystals that display evident lattice fringes, indicating the high crystallinity of the as-obtained nanocrystals. The interplanar spacings for the S-doped TiO₂ nanocrystals are calculated to be 0.35 nm, identical to the distances between the (101) crystal planes of anatase TiO₂. It is also worth noting from Fig. 2 that the as-obtained S-doped TiO₂ nanocrystals are highly dispersed and exhibit an ordered arrangement, indicative of the retentivity of capping ligands on the surfaces of nanocrystals, as evidenced by Fourier transform infrared spectroscopy (FTIR) measurement results in Fig. S5.†

Fig. 2 TEM images of as-obtained S-doped TiO₂ nanocrystals (inset shows the corresponding HRTEM image).

The chemical states of the S-doped TiO₂ nanocrystals have been investigated with X-ray photoelectron spectra (XPS) as shown in Fig. 3. It can be seen from the XPS survey spectra in Fig. 3a that the S-doped TiO₂ contained predominantly Ti, O, C and S elements. The sharp peaks attributed to the core levels of C 1s reveal the presence of oleic acid and oleylamine ligands on the surfaces of the doped nanocrystals.¹⁹ Panels b–d of Fig. 3 show XPS spectra recorded for the Ti 2p, O 1s, and S 2p regions of the S-doped TiO₂ nanocrystals. As is seen from Fig. 3b, two intense peaks at 465.3 and 459.5 eV ascribed to the core levels of Ti 2p_1/2 and Ti 2p_3/2, respectively, demonstrate that the oxidation states of titanium ions are mainly quadrivalence for the S-doped TiO₂ nanocrystals. The fitting of the O 1s region with two peaks shown in Fig. 3c indicates that at least two kinds of oxygen species existed in the near surface domain of the S-doped TiO₂ nanocrystals. The peak located at about 530.3 eV is due to crystal lattice oxygen of S-doped TiO₂ nanocrystals, while the peak at about 532.6 eV is due to chemisorbed oxygen on the nanocrystal surfaces. Fig. 3d depicts the XPS spectra obtained for the S 2p regions of S-doped TiO₂ nanocrystals. There were two observable peaks at 162.4 and 161.4 eV attributable to the core levels of S 2p_1/2 and S 2p_3/2, respectively, which are separated by a spin–orbit splitting of 1.0 eV. No evidence of S⁶⁺ (binding energy at 169.0 eV) is detected, demonstrating that the sulfur atoms are all in the state of S²⁻.^12a It corresponds to the Ti–S bond formed when some of the oxygen atoms in the TiO₂ crystal lattice are replaced by sulfur atoms, as a result of band gap narrowing of TiO₂ calculated by Umebayashi et al.^6c It is also noted that in our XPS characterization, we employed the Al-Kα radiation as the X ray light source, and the escape depth was around 2 nm (Fig. S6 and S7†). Relative to the sizes of as-synthesized S-doped TiO₂ nanocrystals, the XPS characterization information in our study is also from the total samples.

Fig. 3 (a) XPS survey spectra of as-obtained S-doped TiO₂ nanocrystals. (b) Ti 2p, (c) O 1s, and (d) S 2p signals taken from S-doped TiO₂ nanocrystals.

Fig. S5† shows the FTIR spectra of oleic acid, oleylamine, and the prepared S-doped TiO₂ nanocrystals dispersed in toluene–hexane solutions. The strong C–H stretching vibrations at 2920 and 2849 cm⁻¹ show the coexistence of free oleic acid and oleylamine. The observable peak located at 1456 cm⁻¹ is ascribed to carboxylate (COO⁻) stretch, indicating that the COO⁻ ligands existed on the surface of TiO₂ and S-doped TiO₂ nanocrystals. In addition, the discernible peak at 1375 cm⁻¹ is attributed to the C–N stretch of oleylamine. Based on the FTIR analysis of TiO₂ and S-doped TiO₂ nanocrystals, it is validated that the as-obtained nanocrystals are possibly coated by two kinds of organic molecules, namely oleic acid and oleylamine.²⁰ Moreover, compared with the pure TiO₂, the new band located at 1045 and 860 cm⁻¹ is the characteristic of Ti–S vibration, further indicating the successful doping of S anions into TiO₂ nanocrystals.^11b,13

Fig. 4a shows the ultraviolet visible (UV-vis) absorption spectrum of as-prepared pure TiO₂ and S-doped TiO₂ nanocrystals dispersed in toluene–hexane solutions. A discernible absorption edge was detected at 385 and 415 nm, which is the lowest-energy excitonic absorption peak of TiO₂ and S-doped TiO₂ nanocrystals, respectively. In comparison with the band-gap energy of pure TiO₂, the lowest-energy exciton transition peak of S-doped TiO₂ shows a noticeable red shift, implying that the S dopants are incorporated into the lattice of TiO₂, thus altering its crystal and electronic structures, which enables the S-doped TiO₂ nanocrystals as effective visible-light photocatalysts.^6c,12b

Fig. 4 (a) UV-visible absorbance spectra of pure TiO₂ and S-doped TiO₂ nanocrystals; UV-vis spectra of rhodamine B (RhB) dye vs. photoreaction time in the presence of (b) TiO₂ and (c) S-doped TiO₂ nanocrystals. (d) Photocatalytic degradation of the RhB dye over without a catalyst, commercial P25, TiO₂ and S-doped TiO₂ nanocatalysts under visible light irradiation (λ = 420–770 nm).

The photocatalytic properties of as-prepared nanocatalysts are studied by photodegradation of the rhodamine B (RhB) dye. Fig. 4b and c show the change of absorption spectra for the photocatalytic degradation of the RhB dye as a function of irradiation time under artificial visible light (λ = 420–770 nm) in the presence of synthesized TiO₂ and S-doped TiO₂ nanocatalysts. The intensity of the absorption band of the RhB dye located at λ = 553 nm gradually decreases when prolonging the irradiation time, showing that the decoloration of the RhB dye can be achieved by TiO₂ and S-doped TiO₂ nanocatalysts under visible-light irradiation. It should be pointed out that the visible-light photocatalytic activity of undoped TiO₂ nanocatalysts may be caused by the concomitant oxygen deficiency.^6f,12b Compared with the pure TiO₂ shown in Fig. 4b, the RhB dye has almost completely degraded after 40 min under visible light irradiation for S-doped TiO₂ nanocatalysts (Fig. 4c), indicating that the S-doped TiO₂ nanocatalysts show enhanced photocatalytic activity under the same conditions (Fig. S8†).

Fig. 4d shows the photodegradation rate of the RhB dye at various time intervals during the visible light irradiation by pure TiO₂ and S-doped TiO₂ nanocatalysts. For comparison, the decomposition over commercial P25 was carried out under the same experimental conditions. As can be seen from Fig. 4d, S-doped TiO₂ nanocrystals show a superior visible light photodegradation rate of the RhB dye to undoped TiO₂ nanocatalysts. Compared with the pure TiO₂ nanocrystals (the degradation rate is 87.5%), the photocatalytic activities and photodegradation rates increase for S-doped TiO₂ nanocrystals with the maximum photocatalytic activity being 99%. In addition, both the as-obtained undoped TiO₂ and S-doped TiO₂ nanocrystals show the enhanced degradation rate of the RhB dye compared with the commercial P25 (the degradation rate is 70.1%). On the other hand, the adsorption experiments were performed on all catalysts in the absence of light for 30 min. The larger absorption ratio of S-doped TiO₂ nanocrystals may be attributed to their higher surface area (Fig. 4d).^11a The photocatalytic stability of the S-doped TiO₂ nanocrystals under visible light irradiation is shown in Fig. S9.† Experimentally, the photocatalysts were used repeatedly six times after separation through filtration, and the activity was evaluated and compared. As can be seen from Fig. S9,† when irradiated by visible light for 15 min, the photocatalytic degradation efficiency of RhB was kept 80% after the sixth time, demonstrating a highly stable photocatalytic performance for as-obtained S-doped TiO₂ nanocatalysts. In addition, we used the methylene blue (MB) dye as another model molecule to evaluate the photodegradation performance of as-synthesized S-doped TiO₂ nanocatalysts; as can be seen from Fig. S10,† the S-doped TiO₂ nanocrystals also show an enhanced visible light photodegradation rate of the MB dye compared to pure P25 and undoped TiO₂ nanocatalysts.

As indicated from the XRD, XPS, FTIR, and UV-vis results, it can be concluded that S was incorporated into the O site of the TiO₂ crystal lattice. In the undoped TiO₂ crystal, the valence band (VB) and conduction band (CB) consist of the Ti 3d and O 2p orbital. According to the crystal field theory, the Ti 3d orbital is split into two parts of the t_2g and e_g states in an octahedral field with O_h symmetry; thus the CB is separated into the lower and upper parts. When S doped into TiO₂ crystals, the S 3p states are delocalized, thus playing a significant role in the formation of the VB with the O 2p and Ti 3d states. As a consequence, the mixing of the S 3p states with VB increases the width of the VB, which accounts for the band gap narrowing due to S doping, as illustrated in Fig. S11.†^6c,21

In summary, in this paper, we have demonstrated for the first time a facile synthesis of monodisperse S-doped TiO₂ nanocatalysts in hot boiling organic solvents of oleic/oleylamine/1-octadecene. The as-obtained S-doped TiO₂ nanocatalysts are predominately composed of quadrilateral plate-like shapes, and have an average size in the range of (14.5 ± 1.2) nm × (15.3 ± 1.1) nm. Compared with the as-obtained undoped TiO₂ nanocatalysts, the S-doped TiO₂ nanocatalysts present obvious enhanced visible light activation for the degradation of rhodamine B and methylene blue dyes under the artificial visible light (λ = 420–770 nm) irradiation. We believe that our strategy could be broadly applicable for the facile production of other TiO₂ nanocatalysts doped with nonmetals with great promise for various applications.

Acknowledgements

We gratefully acknowledge the financial aid from the start-up funding from Xi'an Jiaotong University (XJTU) and the National Nature Science Foundation of China (NSFC, grant no. 21371140).

Notes and references

1. A. Fujishima and K. Honda, Nature, 1972, 238, 37.
2. D. F. Ollis and H. Al-Ekabi, Photocatalytic Purification and Treatment of Water and Air, Elsevier Science, 1993, pp. 719–725.
3. M. R. Hoffmann, S. T. Martin, W. Choi and D. W. Bahnemann, Chem. Rev., 1995, 95, 69.
4. M. S. Dresselhaus and I. L. Thomas, Nature, 2001, 414, 332.
5. (a) W. Choi, A. Termin and M. R. Hoffmann, Angew. Chem., Int. Ed. Engl., 1994, 33, 1091; (b) J. A. Wang, R. Limas-Ballesteros, T. Lopez, A. Moreno, R. Gomez, O. Novaro and X. J. Bokhimi, J. Phys. Chem. B, 2001, 105, 9692; (c) I. S. Shah, W. Li, P. C. Huang, O. Jung and C. Ni, Proc. Natl. Acad. Sci. U. S. A., 2002, 99, 6482; (d) H. Haick and Y. Paz, J. Phys. Chem. B, 2003, 107, 2319; (e) S. Kim, S. J. Hwang and W. J. Choi, Phys. Chem. B, 2005, 109, 24260.
6. (a) R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269; (b) S. U. M. Khan, M. Al-Shahry and W. B. Ingler Jr., Science, 2002, 297, 2243; (c) T. Umebayashi, T. Yamaki, H. Itoh and K. Asai, Appl. Phys. Lett., 2002, 81, 454; (d) S. Sakthivel and H. Kisch, Angew. Chem., Int. Ed., 2003, 42, 4908; (e) J. C. Yu, W. K. Ho, J. G. Yu, H. Y. Yip, P. K. Wong and J. C. Zhao, Environ. Sci. Technol., 2005, 39, 1175; (f) W. Ho, J. C. Yu and S. Lee, Chem. Commun., 2006, 1115; (g) H. Tian, J. F. Ma, K. Li and J. J. Li, Ceram. Int., 2009, 35, 1289; (h) B. Naik, K. M. Parida and C. S. Gopinath, J. Phys. Chem. C, 2010, 114, 19473.
7. C. Wang, D. W. Bahnemann and J. K. Dohrmann, Chem. Commun., 2000, 1539.
8. T. Ohno, T. Mitsui and M. Matsumura, Chem. Lett., 2003, 32, 364.
9. T. Umebayashi, T. Yamaki, S. Tanaka and K. Asai, Chem. Lett., 2003, 32, 330.
10. L. Gao, S. Zheng and Q. H. Zhang, Nanometer Titania Photocatalytic Materials and Their Applications, Chemical Industry Press, 2002, p. 104.
11. (a) S. X. Liu and X. Y. Chen, J. Hazard. Mater., 2008, 152, 48; (b) Y. M. Liu, J. Z. Liu, Y. L. Lin, Y. F. Zhang and Y. Wei, Ceram. Int., 2009, 35, 3061; (c) S. H. Nam, T. K. Kim and J. H. Boo, Catal. Today, 2012, 185, 259.
12. (a) W. Ho, J. C. Yu and S. C. Lee, J. Solid State Chem., 2006, 179, 1171; (b) G. Liu, C. H. Sun, S. C. Smith, L. Z. Wang, G. Q. Lu and H. M. Cheng, J. Colloid Interface Sci., 2010, 349, 477.
13. G. D. Yang, Z. F. Yan and T. C. Xiao, Appl. Surf. Sci., 2012, 258, 4016.
14. (a) X. W. Wu, D. J. Wu and X. J. Liu, Appl. Phys. A, 2009, 97, 243; (b) M. V. Dozzi, S. Livraghi, E. Giamellob and E. Selli, Photochem. Photobiol. Sci., 2011, 10, 343.
15. (a) E. M. Rockafellow, L. K. Stewart and W. S. Jenks, Appl. Catal., B, 2009, 91, 554; (b) L. Szatmáry, S. Bakardjieva, J. Šubrt, P. Bezdička, J. Jirkovský, Z. Bastl, V. Brezová and M. Korenko, Catal. Today, 2011, 161, 23.
16. M. Jalalah, M. Faisal, H. Bouzid, A. A. Ismai and S. A. Al-Sayari, Mater. Res. Bull., 2013, 48, 3351.
17. H. X. Li, X. Y. Zhang, Y. N. Huo and J. Zhu, Environ. Sci. Technol., 2007, 41, 4410.
18. (a) T. Ohsaka, F. Izumi and Y. Fujiki, J. Raman Spectrosc., 1978, 7, 321; (b) C. Han, M. Pelaez, V. Likodimos, A. G. Kontos, P. Falaras, K. O'Shea and D. D. Dionysiou, Appl. Catal., B, 2011, 107, 77.
19. Y. P. Du, Z. Y. Yin, J. X. Zhu, X. Huang, X. J. Wu, Z. Y. Zeng, Q. Y. Yan and H. Zhang, Nat. Commun., 2012, 3, 1177.
20. Y. P. Du, Z. Y. Yin, X. H. Rui, Z. Y. Zeng, X. J. Wu, J. Q. Liu, Y. Y. Zhu, J. X. Zhu, X. Huang, Q. Y. Yan and H. Zhang, Nanoscale, 2013, 5, 1456.
21. (a) T. Umebayashi, T. Yamaki, S. Yamamoto, A. Miyashita, S. Tanaka, T. Sumita and K. Asai, J. Appl. Phys., 2003, 93, 5156; (b) T. Ohno, M. Akiyoshi, T. Umebayashi, K. Asai, T. Mitsui and M. Matsumura, Appl. Catal., A, 2004, 265, 115.

---

Footnote

† Electronic supplementary information (ESI) available: The experimental details, result of FTIR is provided. See DOI: 10.1039/c4qi00027g

---