Fabrication of TiO₂/MS (M = Pb, Zn) core–shell coaxial nanotube arrays and their photocatalytic properties

Abstract

In this work, TiO₂/MS (M = Pb, Zn) core–shell coaxial nanotube arrays were prepared by a simple method of liquid deposition using anodic aluminium oxide templates. The mechanism of the formation of TiO₂/MS (M = Pb, Zn) coaxial nanotubes is discussed here. Compared to bare TiO₂ nanotubes, TiO₂/PbS and TiO₂/ZnS composite nanotubes showed improved photocatalytic properties and analysis of such results was also conducted.

---

With the development of economy and industry, the increase in sewage drained by factories would seriously harm the environment. The idea of using sunlight as the driving force for semiconductors to degrade pollutants into nontoxic materials is becoming more and more popular. Among various semiconductors, TiO₂ is more attractive because of its amazing characteristics such as high stability, non-toxicity, high oxidative power and low cost. It has been used widely in many fields, such as solar cells, water splitting and degradation of toxic chemicals.^1–3 However, due to the wide band gap (3.2 eV), the photoresponse of TiO₂ is restricted to only ultraviolet region with the wavelength <390 nm, which significantly diminishes the utilization of solar energy. Therefore, for TiO₂, many methods have been applied to improve its performance in optical absorption. Such designed strategies include doping TiO₂ with metal or nonmetal ions,^4,5 decorating TiO₂ with quantum dots (QDs),^6,7 and coupling TiO₂ with other semiconductors.^8–10 Among those various semiconductors, PbS and ZnS are the two considerably attractive choices. PbS as a narrow band gap material (0.37 ev) has an extended photoresponse region of sunlight and for ZnS, the potentials of the conduction band and the valence band are both charged a bit more negative than TiO₂. This unique band structure could enhance the separation of the photoproduced holes and electrons. As we all know, nanotubes have superior surface-to-volume ratios and large surface area compared with bulk materials. Until now, there have been many works concerning the fabrication of TiO₂ nanotubes, such as synthesis of TiO₂ nanotubes by anodization of high-purity Ti foils,^11,12 hydrothermal synthesis,¹³ sol–gel method¹⁴ and solvothermal process.¹⁵ Anodic aluminium oxide (AAO) membrane contains hexagonally ordered porous structure and could be a suitable template for the fabrication of one-dimensional nano-structures. Up to now, there have been some studies about fabricating nano-structures by double diffusion method with the assistance of AAO templates.^16,17

In this study we used AAO membrane (Fig. 1a) as the template to prepare TiO₂/MS (M = Pb, Zn) core–shell coaxial nanotube arrays. First, we fabricated TiO₂ shell (Fig. 1b) on the inwall of AAO template by liquid deposition. Then, precipitated MS (M = Zn, Pb) core (Fig. 1c) on the TiO₂ shell by double diffusion (Fig. 1d) during hydrothermal reaction. In brief, the AAO template with TiO₂ precursors on it was immersed into separate solutions of Na₂S₂O₃ and M(CH₃COO)₂ (M = Pb, Zn), where the template worked as a septum between the two solutions (Fig. 1d), and the reaction was carried out in teflon-lined hydrothermal synthesis reactor at 180 °C for 6 hours. Further experimental details are provided in the ESI.† Although there are some studies on the modifications of TiO₂ with MS (M = Zn, Pb), they mainly focus on fabricating the composites of disordered nanoparticles^18,19 or films.²⁰ However, there are only few reports concerning the PbS, and/or ZnS modified TiO₂ to form a core–shell coaxial composite nanotubes. One of the main purposes of our work is to propose a simple and convenient method to fabricate ordered TiO₂/MS core–shell coaxial composite nanotube arrays, which could also be used to synthesize other kinds of core–shell coaxial composites.

Fig. 1 A simple diagram of the reaction process: (a) a single nanochannel of AAO template; (b) TiO₂ shell was deposited on the inwall of AAO template; (c) MS (M = Pb, Zn) core was fabricated on the inwall of TiO₂ shell; (d) diagram of the double diffusion process during fabrication of MS core.

For SEM and TEM tests, the AAO templates were dissolved using NaOH solution and the nanotubes were all free from the templates. For the other tests, the nanotubes were all embedded in the AAO templates. Fig. 2a–f show the SEM images of the samples. Fig. 2a is the morphology of AAO template, with an average pore diameter of 250 nm. Fig. 2b shows that the TiO₂ nanotubes produced by liquid deposition had an average pore diameter of 200 nm. Fig. 2c and d are the SEM images of TiO₂/PbS nanotubes. From the two images, the pore diameter of TiO₂/PbS nanotubes was ∼180 nm, which was smaller than that of TiO₂ nanotubes. From Fig. 2e and f, the hollow structure was also determined for TiO₂/ZnS nanotubes. Similarly, the pore diameter of TiO₂/ZnS nanotubes was obviously smaller than that of TiO₂ nanotubes. Thus, it was safe to say that the MS could be deposited on the inner part of TiO₂ nanotubes by means of double diffusion through the AAO template. In order to show the structure of the prepared samples clearly, TEM test was also carried out in this work. From Fig. 2g, we could see that the pore diameter of the TiO₂/PbS nanotube was ∼180 nm, while the thickness of TiO₂ was ∼20 nm and that of PbS was ∼10 nm. From Fig. 2h, the thickness of TiO₂ shell was about 20 nm and that of ZnS core was about 30 nm. The distinct interface between the outer layer of TiO₂ and the inner layer of MS suggested the structure of core–shell coaxial nanotube were obtained by our simple method.

Fig. 2 SEM images of: (a) AAO template; (b) TiO₂ produced by liquid deposition; (c and d) TiO₂/PbS core–shell coaxial nanotube arrays; (e and f) TiO₂/ZnS core–shell coaxial nanotube arrays and TEM images of: (g) one single TiO₂/PbS core–shell coaxial nanotube; (h) one single TiO₂/ZnS core–shell coaxial nanotube.

Fig. 3 exhibited crystal structure, elemental composition and chemical state of the samples. Fig. 3a is the XRD spectra of TiO₂, TiO₂/PbS, and TiO₂/ZnS nanotube arrays. From the XRD pattern of TiO₂ nanotube arrays, it is clear that TiO₂ is anatase phase with the main peaks at 2θ = 25.3°, 38.6°, 48.0°, 53.9°, 55.0°, 62.7°, 74.0°corresponding to (101), (112), (200), (105), (211), (204), (107) planes of anatase TiO₂ (JCPD#21-1272), respectively. For TiO₂/PbS NTAs, except for the peaks (2θ = 25.3°, 48.0°) from TiO₂, the peaks at 2θ = 26.0°, 30.1°, 43.1°, 51.0°, 53.4°, 62.5°, 68.9°, 71.0°, 78.9° corresponding to (111), (200), (220), (311), (222), (400), (331), (420), (422) planes of cubic PbS (JCPDS#05-0592), revealed that the PbS existed in the product. For TiO₂/ZnS NTAs, except for the peaks of TiO₂, the other peaks located at 2θ = 28.6°, 47.6° and 56.3° was a good evidence of cubic ZnS (JCPD#05-0566).

Fig. 3 (a) XRD patterns of TiO₂, TiO₂/PbS, TiO₂/ZnS nanotube arrays; (b) XPS scanning spectrum of TiO₂, TiO₂/PbS, TiO₂/ZnS nanotube arrays; (c)–(e) high-resolution XPS spectra of S, Pb and Zn.

The elemental composition and the chemical state of the samples were further investigated by XPS (Fig. 3b–e). Fig. 3b is the entire scanning spectrum of TiO₂, TiO₂/PbS, TiO₂/ZnS nanotube arrays, showing the main elemental compositions of the samples, including Ti, O, Zn, Pb, S. Here, the XPS peaks of S (2p at 162.2 eV for ZnS and 160.9 eV for PbS) were in agreement with the binding energy of Zn–S bond and Pb–S bond, respectively (Fig. 3c).^21,22 The additional peaks of S (2p at 168.5 eV for ZnS and 168.2 eV for PbS) may arise from residual surface contaminants, for example, Na₂S₂O₃.²³ The peaks at 137.3 eV and 142.1 eV were corresponding to the previously reported binding energy of Pb 4f5/2 and 4f7/2, respectively, indicating the presence of Pb²⁺ (Fig. 3d).²⁴ The two strong peaks in Fig. 3e at 1022.4 eV and 1045.4 eV were assigned to the binding energy of Zn 2p3/2 and Zn 2p1/2, respectively, suggesting the existence of Zn²⁺.²⁵ All these results suggested that the ZnS and PbS were obtained in our work.

The mechanism of the formation of TiO₂/MS (M = Pb, Zn) core–shell coaxial nanotubes is shown in Fig. 5. The deposition of TiO₂ on the inwall of AAO template could be described by following equations:

TiF₄ + H₂O ⇆ TiO₂ + H⁺ + F⁻ (1)

12H⁺ + 12F⁻ + Al₂O₃ ⇆ 2H₃AlF₆ + 3H₂O (2)

Fig. 4 (a)–(c) Liquid deposition of TiO₂ on the inwall of AAO template; (d and e) deposition of MS (M = Pb, Zn) on the inwall of AAO/TiO₂.

Fig. 5 (a) UV-Vis diffuse-reflectance spectra of TiO₂, TiO₂/PbS, TiO₂/ZnS nanotubes; (b) absorption spectrum of methyl orange in the presence of TiO₂/PbS nanotubes; (c) photocatalytic performance of TiO₂, TiO₂/PbS, TiO₂/ZnS nanotubes for the degradation of methyl orange solution under UV light.

TiF₄ will be transformed into HF and TiO₂ in the solution (eqn (1)). As a F⁻ trapping agent, Al₂O₃ will consume F⁻ and H⁺ (reaction (2)). With the reaction (2) proceeding toward the right-hand side, the production of TiO₂ and HF (Fig. 4a) will be boosted. Because HF has a corrosion effect on AAO template (eqn (2)), the surface of the nanochannels will become rough and full of bumps and pit, which makes it easy for TiO₂ particles to adhere to (Fig. 4b). Eventually, with the proceeding of the reaction, ordered TiO₂ nanotube arrays could be formed in the AAO template (Fig. 4c). As for the deposition of MS on AAO/TiO₂, it could be explained using the following reaction (3):

S₂O₃²⁻ + M²⁺ + 2CH₃COO⁻ + H₂O = MS↓ + 2CH₃COOH + SO₄²⁻ (3)

Due to the concentration gradient between the two sides of the AAO/TiO₂ template (septum), S₂O₃²⁻ and M²⁺ will migrate to the opposite side through the nanochannels (Fig. 4d), encounter each other, and finally form the MS sediments on the inwall of AAO/TiO₂. Hereto, the fabrication of TiO₂/MS core–shell coaxial nanotube arrays was finished in the manner of double diffusion (Fig. 4e).

Fig. 5a showed the UV-Vis diffuse-reflectance spectra of the samples. From the spectra, it could be seen that the absorption onset of TiO₂ laid at 387 nm, corresponding to that of anatase TiO₂. After coupling with PbS or ZnS, the onset of TiO₂/MS showed a red-shift more or less. Especially, for TiO₂/PbS, it exhibited a significantly enhanced light absorption, covering almost the entire visible region. This characteristic may be helpful for its photocatalytic activity. The photocatalytic activity of TiO₂/PbS core–shell coaxial nanotube arrays measured by the degradation of methyl orange is shown in Fig. 5b. Experimental details about the test of photocatalytic activity are provided in the ESI.† It was clear that there were two obvious absorption peaks at about 280 nm and 465 nm, corresponding to the absorption peaks of methyl orange.²⁶ The UV-Vis absorption of methyl orange at 465 nm was chosen as the parameter for the photocatalytic activity and the peak diminished gradually as the exposure time increased. When the exposure time reached to 8 hours, the peak almost disappeared, indicating that the methyl orange was degraded almost completely. Furthermore, photocatalytic activities of the three samples were illustrated in Fig. 5c. The degradation efficiency of the samples was calculated by C/C₀, where C₀ was the initial concentration of methyl orange, and C was the concentration during the reaction. Although methyl orange showed a little degradation because of the long duration of UV light irradiation, among the four samples, the order of their photocatalytic performance was shown as follows: TiO₂/PbS > TiO₂/ZnS > TiO₂ > MO. After being coupled with MS, the photocatalytic properties of the TiO₂/MS composite core–shell nanotubes were greatly enhanced compared with that of the bare TiO₂ nanotubes.

Here, the better photocatalytic performance of TiO₂/PbS or TiO₂/ZnS than that of TiO₂ could be explained as follows (Fig. 6). As schematized in Fig. 6a, conduction band (CB) of TiO₂ lies at a more negative potential than that of PbS, therefore the excited electrons have a tendency to transfer from CB of TiO₂ to that of PbS. Because of the huge gap (2.3 eV) between valance band (VB) of PbS and that of TiO₂, it may be extremely hard for the holes to transit from VB of TiO₂ to that of PbS, though the VB of TiO₂ lies at a more positive potential. Thus, for PbS, its narrow band gap may work as electron traps capturing the excited electrons from TiO₂, thus enhancing the separation of photoproduced electrons and holes to some extent. On the other hand, after coupling with PbS, TiO₂/PbS nanotube arrays can absorb more visible-region light to generate more electrons and holes (Fig. 5a). Based on the two reasons, TiO₂/PbS nanotube arrays showed a better photocatalytic activity than the bare TiO₂ nanotube arrays.

Fig. 6 Schematic illustration of the band structure related photocatalytic mechanism of the TiO₂/PbS (a), TiO₂/ZnS (b) heterostructure.

As schematized in Fig. 6b, the CB of TiO₂ lies at a more positive potential than that of ZnS, while the VB of ZnS is more negative than that of TiO₂. For TiO₂/ZnS, excited electrons and holes will prefer to be collected by TiO₂ and ZnS, respectively. Such band structure facilitates the separation of the excited electron–hole pairs, therefore efficiently preventing the recombination of holes and electrons. Longer life-time of photogenerated electrons and holes ensures a better photocatalytic performance of TiO₂/ZnS composite nanotubes than that of bare TiO₂ nanotubes.

It was reported that the structure of the samples had an effect on the light absorption of the catalysts.^27–29 In addition, the photocatalytic activity could be influenced by the thickness of MS layer or TiO₂ layer. But, the main purpose of our work is to report a new method for preparing well-ordered arrays of TiO₂/MS (M = Pb, Zn) nanotubes and compare the difference of the photocatalytic activities with different structures. Further investigation about factors, such as the structure of the samples and the thickness of the layer will be shown in our future work.

In this study, through liquid deposition, the TiO₂/PbS and TiO₂/ZnS nanotube arrays were successfully synthesized by using AAO templates. TiO₂/PbS and TiO₂/ZnS nanotubes showed enhanced photocatalytic activities than bare TiO₂ nanotubes owing to their special band structures, which facilitated the separation of photogenerated holes and electrons. The method used to prepare the TiO₂/MS core–shell coaxial nanotube arrays was simple and convenient, and more composite materials with core–shell coaxial structure using the method could be expected in the future.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Grant no. 51272115) and the Scientific Research Encouragement Foundation for Outstanding Young and Middle Aged Scientists of Shandong Province, China (Grant no. BS2013CL025).

Notes and references

1. S. Guldin, P. Docampo, M. Stefik, G. Kamita, U. Wiesner, H. J. Snaith and U. Steiner, Small, 2012, 8(3), 432.
2. X. J. Xu, X. S. Fang, T. Y. Zhai, H. B. Zeng, B. D. Liu, X. Y. Hu, Y. S. Bando and D. Golberg, Small, 2011, 7(4), 445.
3. Z. F. Liu, K. Y. Guo, J. H. Han, Y. J. Li, T. Cui, B. Wang, J. Ya and C. L. Zhou, Small, 2014, 10(15), 3153.
4. X. L. Cui, M. Ma, W. Zhang, Y. C. Yang and Z. J. Zhang, Electrochem. Commun., 2008, 10, 367.
5. N. T. Nguyen, J. E. Yoo, M. Altomarea and P. Schmuki, Chem. Commun., 2014, 50, 9653.
6. S. Rühle, M. Shalom and A. Zaban, ChemPhysChem, 2010, 11, 2290.
7. F. Xu, X. Ma, C. R. Haughn, J. Benavides, M. F. Doty and S. G. Cloutier, ACS Nano, 2011, 5(12), 9950.
8. Q. Kang, S. H. Liu, L. X. Yang, Q. Y. Cai and C. A. Grimes, ACS Appl. Mater. Interfaces, 2011, 3, 746.
9. Y. L. Lee, C. F. Chi and S. Y. Liau, Chem. Mater., 2010, 22, 922.
10. C. Ratanatawanate, A. Chyao and K. J. Balkus Jr, J. Am. Chem. Soc., 2011, 133, 3492.
11. K. Shin, S. I. Seok, S. H. Im and J. H. Park, Chem. Commun., 2010, 46, 2385.
12. V. Galstyan, A. Vomiero, I. Concina, A. Braga, M. Brisotto, E. Bontempi, G. Faglia and G. Sberveglieri, Small, 2011, 7(17), 2437.
13. T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino and K. Niihara, Adv. Mater., 1999, 11(15), 1307.
14. J. J. Qiu, Z. G. Jin, Z. F. Liu, X. X. Liu, G. Q. Liu, W. B. Wu, X. Zhang and X. D. Gao, Thin Solid Films, 2007, 515(5), 2897.
15. Z. F. Bian, J. Zhu, F. L. Cao, Y. N. Huo, Y. F. Lu and H. X. Li, Chem. Commun., 2010, 46, 8451.
16. A. Varghese, P. Ghosh and S. Datta, J. Phys. Chem. C, 2014, 118, 21604.
17. J. H. Dai, Q. Liu, L. Wang, P. W. Wu, X. Huang, Z. B. Zhu and J. T. Tian, Mater. Lett., 2013, 107, 333.
18. K. P. Acharya, N. N. Hewa-Kasakarage, T. R. Alabi, I. Nemitz, E. Khon, B. Ullrich, P. Anzenbacher and M. Zamkov, J. Phys. Chem. C, 2010, 114, 12496.
19. X. D. Yu, Q. Y. Wu, S. C. Jiang and Y. H. Guo, Mater. Charact., 2006, 57, 333.
20. K. P. Acharya, E. Khon, T. O'Conner, I. Nemitz, A. Klinkova, R. S. Khnayzer, P. Anzenbacher and M. Zamkov, ACS Nano, 2011, 5(6), 4953.
21. B. Ding, Y. Wang, P. S. Huang, D. H. Waldeck and J. K. Lee, J. Phys. Chem. C, 2014, 118, 1474.
22. D. D. Lin, H. Wu, R. Zhang, W. Zhang and W. Pan, J. Am. Ceram. Soc., 2010, 93(10), 3384.
23. B. V. Crist, Handbook of The Elements and Native Oxides, XPS International, Inc, 3408 Emerald Drive, Ames Iowa 50010 USA, 1999.
24. W. Jang, D. Kim, J. Kim, B. K. Min, J. D. Kim and K. P. Yoo, Chem. Mater., 2010, 22, 4350.
25. X. Yang, H. T. Xue, J. Xu, X. Huang, J. Zhang, Y. B. Tang, T. W. Ng, H. L. Kwong, X. M. Meng and C. S. Lee, ACS Appl. Mater. Interfaces, 2014, 6, 9078.
26. S. H. Wang and S. Q. Zhou, J. Hazard. Mater., 2011, 185, 77.
27. C. J. Chang and S. T. Hung, Thin Solid Films, 2008, 517, 1279.
28. D. Majumdar, S. Chatterjee, M. Dhar, S. K. Dutta and H. Saha, Sol. Energy Mater. Sol. Cells, 2003, 77, 51.
29. M. H. Hsu and C. J. Chang, Int. J. Hydrogen Energy, 2014, 39, 16524.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c4ra12998a

---