Cu₂O sensitized flexible 3D-TiO₂ nanotube arrays for enhancing visible photo-electrochemical performance

Abstract

Cu₂O flake and particle modified 3D-TiO₂ nanotube arrays (TiO₂ NTAs) on flexible Ti meshes were prepared by electrochemical deposition. The phase composition, microstructure and photo-electrochemical property were characterized by X-ray diffraction (XRD), field emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), UV-vis diffusion reflection spectroscopy (DRS) and an electrochemical system. The results indicate that 3D-TiO₂ NTAs are covered by a large number of Cu₂O flakes and nanoparticles, and the flakes become longer and narrower with increasing electrochemical cycles. Cu₂O modified 3D-TiO₂ NTAs expand the photo-response range from ultraviolet light to visible light. The Cu₂O modified NTAs with 400 pulses possess the highest visible photocurrent density of 0.9 mA cm⁻², however, the Cu₂O modified NTAs with 600 pulses possess the largest photocatalytic activity toward degradation of methyl orange (MO) under visible-light irradiation. A mechanism is proposed to explain the difference in photocurrent response and photocatalytic activity.

---

Introduction

Since Fujishima and Honda reported TiO₂ as a photoanode material for photo-catalyzing H₂ from water in 1972,¹ TiO₂-based nanomaterials have attracted tremendous attention due to their unique electronic,² optical^3,4 and nontoxic properties.⁵ Recently TiO₂ NTAs have been intensively investigated due to their high regulation, large surface areas, excellent controllability and superior electron transport rate,^6,7 which have been widely used in Li-ion batteries,^8,9 dye sensitized solar cells (DSCs),¹⁰ photocatalysis,¹¹ catalyst supports¹² and super capacitors.¹³ Compared to TiO₂ NTAs prepared by anodization on Ti foil, TiO₂ NTAs on Ti mesh have much more advantages such as a higher surface area,¹⁴ a higher effluent and air flow ability,^15,16 a larger Ti conversion rate, an easier access for electron to an electrolyte,^17–19 more flexibility²⁰ and independent of the direction of the sun light, which are especially important for TiO₂ photo-degradation of organic dyes and toxic metal ions in waste water. For example, Liu et al.² prepared vertically oriented TiO₂ NTAs grown on Ti meshes for flexible dye-sensitized solar cells. Liao et al.⁴ reported that 3D TiO₂ NTAs could be used to photocatalytic degradation of methyl orange.

In order to overcome the intrinsic weakness of the wide-band-gap TiO₂ semiconductor, narrow band-gap semiconductors such as Cu₂O,²¹ Fe₂O₃,^22,23 CdS,^24–26 CdTe,^27–29 CuInSe₂ (ref. 30 and 31) and WO₃ (ref. 32) were widely used to sensitize TiO₂, showing an enhanced visible light absorption and photo-electrochemical property.³³ Among narrow band-gap semiconductors cuprous oxide is an inexpensive, non-toxic and readily available semiconductor with a band-gap of 1.95–2.2 eV.³⁴ More importantly, it has a well energy-level matching up with TiO₂,³⁵ which favor the separation of photo-generated electron–hole pairs. Though Cu₂O/TiO₂ NTAs hybrid materials have been studied,^36–40 however, to the best of our knowledge, no report is found on Cu₂O modified 3D TiO₂ NTAs, which exhibit an improved photo-electrochemical property under visible light irradiation.

Experimental

Preparation of flexible 3D-TiO₂ NTAs

All reagents are analytical grade and used without further purification. A large piece of raw Ti mesh (50 meshes, 99.5% purity) with thickness of 0.12 mm was cut into small square pieces of 150 × 70 mm², which was previously ultrasonically degreased in acetone, isopropanol and methanol for 15 min in turn, and then chemically etched in a mixture of HF and HNO₃ aqueous solution (HF : HNO₃ : H₂O = 1 : 4 : 10 in volume, total 20 mL) for 20 s, afterwards rinsed with deionized water, and finally dried in air. Anodization was performed at 50 V for 3 h in ethylene glycol solution containing 0.25 wt% NH₄F and 3 vol% H₂O, using Pt plate as the counter electrode and Ti mesh as the working electrode. The as-prepared samples were ultrasonically rinsed with deionized water and dried at 60 °C for 3 h. To convert samples from amorphous phase to anatase, thermal treatment was performed in air at 450 °C for 3 h.⁴¹

Sensitization of TiO₂ NTAs with Cu₂O

All experiments were performed with a LK98BΠ electrochemical workstation using the anodized Ti mesh as working electrode, platinum plate as counter electrode and Ag/AgCl electrode as reference electrode, respectively. Before electro-deposition, the anodized Ti mesh was carried out cyclic voltammetry from −0.2 V to −1.0 V in 1 M NH₄Cl solution in order to enhance the conductivity of the bottom of TiO₂ nanotubes.⁴² Cu₂O was electrochemically deposited into 3D-TiO₂ NTAs in 20 mL fresh electrolyte containing 0.4 M cupric sulfate and 3 M lactic acid aqueous solution. The pH value of the electrolyte was adjusted to 10 by 5 M NaOH solution. Pulse electro-deposition was adopted with a pulse voltage of −0.44 V, a pulse time for 1 s and an interval time for 4 s, which permitted the depleted ions equilibrium in the interval time. Cu₂O-3D TiO₂ NTAs hybrid materials with different electrochemical cycles were taken out from the electrolyte and rinsed with de-ionized water, then dried at 60 °C for 1 h. The samples with 200, 400 and 600 pulse cycles were denoted as CT(200), CT(400) and CT(600), respectively. In the same time, Cu₂O was electrochemically deposited on blank Ti mesh by the above-mentioned processes, the samples with 200, 400 and 600 pulse cycles were denoted as T(200), T(400) and T(600), respectively.

Characterization

Phase compositions of the samples were measured by a Rigaku D/Max 2400 X-ray diffractometer (XRD) equipped with graphite monochromatized Cu Kα radiation (λ = 0.15405 nm). Microstructures of the samples were characterized by field emission scanning electron microscopy (FESEM, Quanta 200 FEG) and transmission electron microscopy (TEM, Tecnai-F30). UV-vis diffuse reflectance spectra (DRS) of the samples were recorded on UV-2550 UV-vis spectrophotometer (HITACHI, Japan) with an integrating sphere attachment. Photoelectrochemical performances of the samples were measured with LK98II electrochemical system (Lanlike, China) using a three-electrode system composed of the samples as a working electrode, a Pt foil as a counter electrode, Ag/AgCl electrode as a reference electrode in 0.1 M Na₂SO₄ solution. The working electrode was illuminated with a solar simulator equipped with a 500 W Xe lamp with a visible-light filter (<420 nm). The photocurrent dynamics of the working electrode were recorded according to the responses to sudden switching on and off at 0.25 V bias.

Photocatalytic activity test

Photocatalytic activity of the samples was evaluated by decomposing 10⁻⁵ M methyl orange (MO) solution under 500 W Xe lamp with a 420 nm cut-off filter. Before photodegradation, MO adsorption equilibrium on the sample's surface was established by mechanical stirring in the dark for 30 min. The MO concentration was recorded in 1 h at 462 nm by UV1700 UV-vis spectrophotometer.

Results and discussion

Fig. 1 gives FESEM images, EDS and XRD spectra of the anodized TiO₂ NTAs on Ti mesh.

Fig. 1 FESEM images, EDS and XRD spectra of TiO₂ NTAs on Ti mesh.

Fig. 1(A) indicates that some micro-cracks appear in the anodized layer due to the stress mismatch between the anodized layer and Ti substrate. Fig. 1(B) indicates that the anodized layer combines tightly to Ti substrate and has a thickness of about 20 μm. Fig. 1(C) indicates that the anodized layer is composed of ordered nanotube arrays with an average tube diameter of about 100 nm and an average wall thickness of about 19 nm. Fig. 1(D) indicates that the ordered nanotube arrays are composed of anatase TiO₂ (JCPDS, card no: 21-1272). The Ti peaks originate from Ti substrate.

Fig. 2 gives FESEM images and EDS spectra of Cu₂O doped 3D-TiO₂ NTAs with different pulse cycles.

Fig. 2 FESEM images and EDS spectra of CT(200) (A, D, G and J), CT(400) (B, E, H and K) and CT(600) (C, F, I and L).

Fig. 2 indicates that 3D-TiO₂ NTAs are covered by a large number of Cu₂O flakes and nanoparticles. The flakes have an average sizes of about 6 μm (L) × 5 μm (W) × 0.2 μm (T) for 200 pulses, 8 μm (L) × 3 μm (W) × 0.2 μm (T) for 400 pulses and 10 μm (L) × 2 μm (W) × 0.2 μm (T) for 600 pulses, indicating that the flakes become longer and narrower with increasing electrochemical cycles. Besides, Cu₂O particles and aggregations can be found on the mouths of TiO₂ nanotubes in samples CT(200), CT(400) and CT(600). The EDS spectra show that the Cu₂O decorated 3D-TiO₂ NTAs are composed of Ti, O and Cu elements, and Cu atomic content increases in the order of 26.19%, 28.94% and 34.07% for electrochemical pulses of 200, 400 and 600 times, respectively. Namely Cu₂O content increases with cycle times.

Fig. 3 gives Cu₂O growth mechanism on 3D TiO₂ NTAs.

Fig. 3 Schematic diagram (A–C) and SEM images of Cu₂O growth on 3D TiO₂ NTAs for samples CT(1) (A and F), CT(120) (B and D) and CT(200) (C, E and G).

Some Cu₂O nanoparticles with diameter of about 20 nm are found in sample CT(1) as shown in Fig. 3(A and F). After 120 cycles, large amounts of spherical Cu₂O nanoparticles with diameter of about 200 nm are found in the cracks and on the top of TiO₂ NTAs as shown in Fig. 3(B and D). After 200 cycles, besides Cu₂O nanoparticles, Cu₂O flakes appear on the top of TiO₂ NTAs as shown in Fig. 3(E). At the same time, Cu₂O on the peeled-off Ti substrate changes its shape from spherical nanoparticles to cubic particles with a side length of about 300 nm as shown in Fig. 3(G). The EDS spectrum inserted in Fig. 3(G) indicates that the cubic Cu₂O particles on Ti substrate are composed of 48.85 at% O, 8.63 at% Ti and 42.52 at% Cu, no C element is found. The EDS spectrum inserted in Fig. 3(E) indicates that the Cu₂O flakes are composed of 48.00 at% C, 28.78 at% O, 9.62 at% Ti and 13.60 at% Cu. Element C, coming from lactic acid in electrolyte, can selectively bind to {111} facets of the irregular Cu₂O particle and thus effectively block the growth along the vertical axis and only allow extensive growth along the lateral direction,⁴³ which induces the formation of Cu₂O flakes.

Fig. 4 gives current–time curves for electrochemical deposition of Cu₂O and the influences of pulse cycles on the peak current and stable state current.

Fig. 4 Current–time curves for electrochemical deposition of Cu₂O and the influences of pulse times on the peak current (a) and stable state current (b).

Fig. 4(A) indicates that the current increases quickly to a peak value in short time due to the electrical charging and the reduction of Cu²⁺ ions at the working electrode surface. Then it decreases slowly up to a steady value due to the formation of the diffusion layer near the electrode.⁴⁴ The peak current and stable current become smaller with pulse cycles due to the depletion of Cu²⁺ ions in the electrolyte.

Fig. 5 gives XRD patterns and absorption spectra of samples 3D-TiO₂ NTAs, CT(200), CT(400), CT(600) and CT(400)-1. CT(400)-1 is from degradation experiment of sample CT(400) under simulated sunlight illumination for 4 h.

Fig. 5 XRD patterns (A) and UV-vis absorption spectra (B) of samples 3D-TiO₂ NTAs (a), CT(200) (b), CT(400) (c), CT(600) (d) and CT(400)-1 (e).

Fig. 5(A) indicates that four new peaks appear at 36.52°, 42.42°, 61.55° and 73.73° in the samples CT(200), CT(400) and CT(600) besides anatase TiO₂ and Ti peaks in sample 3D-TiO₂ NTAs, which can be assigned to (111), (200), (220) and (311) crystal faces of cubic Cu₂O (JCPDS, card no: 77-0199). Cu₂O phase content increases with pulse cycles based on the intensity of Cu₂O peaks. According to Scherrer equation, the calculated Cu₂O grain size is 34 nm, 21 nm, and 32 nm for samples CT(200), CT(400) and CT(600), respectively. CT(400) and CT(400)-1 have the same peak position and intensity, indicating that both Cu₂O and TiO₂ are stable after the degradation experiment. No CuO diffraction peak is found in these samples, indicating that copper ions in these samples mainly exist in the form of Cu⁺. In order to further clarify the chemical states of Cu, Fig. S2† gives the XPS spectra of CT(400), indicating the existence of Cu²⁺ because the surface of Cu₂O in the sample was partially oxidized into CuO, but its concentration is very low since no CuO diffraction peak is found in the XRD diagram of CT(400).

The absorption edge of the bare TiO₂ NTAs locates at about 387 nm as shown in Fig. 5(B), corresponding to the band-gap energy of anatase TiO₂ (3.2 eV). After electro-depositing Cu₂O, the corresponding absorption edge expands to 516 nm, corresponding to the band-gap energy of cubic Cu₂O (2.4–2.5 eV) as shown in Fig. 5(B), which is larger than the theoretical value of 2.17 eV due to quantum size effect.⁴⁵ The feature absorption peak at about 420 nm originates from excitonic absorption peak of Cu₂O nanocrystals,⁴⁶ which is similar to previous report.⁴⁷

Fig. 6 gives TEM (A and B) and HRTEM (C and D) images of sample CT(400).

Fig. 6 TEM (A and B) and HRTEM (C and D) images of CT(400).

Some broke nanotubes can be clearly observed in Fig. 6(A). Fig. 6(B) shows that the nanotube has a diameter of about 80 nm and a wall thickness of about 18 nm, which is consistent with the SEM observation. Many nanoparticles with diameters of approximately 10 nm are clearly found in the interior or on the outside wall of nanotube. The inter-planar distances of 0.156 nm, 0.196 nm and 0.346 nm can be clearly observed in Fig. 6(C), which correspond to (220) plane of cubic Cu₂O crystal, and (200) and (101) planes of TiO₂ crystal, respectively. The inter-planar distances of 0.240 nm and 0.210 nm in Fig. 6(D) are fairly close to those of (004) plane of TiO₂ and (200) plane of Cu₂O, respectively. Cu₂O nanoparticles with a diameter of about 10 nm tightly adsorb on the wall of TiO₂ nanotube as shown in Fig. 6(D).

Fig. 7 gives degradation efficiencies of samples 3D-TiO₂ NTAs, CT(200), CT(400) CT(600), T(200), T(400) and T(600) under visible light or simulated sunlight illumination.

Fig. 7 Degradation efficiencies of different samples under visible light illumination (A) or simulated sunlight illumination (B) (a) 3D-TiO₂ NTAs (b) CT(200) (c) CT(400) (d) CT(600) (e) T(200) (f) T(400) (g) T(600).

Fig. 7 shows that the physical adsorption amount of MO on all samples is very small in comparison with the corresponding photocatalytic decomposition percentage. Bare TiO₂ NTAs have the lowest degradation rate, however, CT(600) has the highest degradation rate either under visible light illumination or simulated sunlight illumination. The MO decomposition rates of samples CT(200), CT(400) and CT(600) are 73.1%, 78.0% and 90.5% under visible light irradiation for 6 h, respectively. Compared to samples CT series, samples T series have much lower MO degradation efficiencies. The results indicate that the photocatalytic activity of different samples strongly depends on not only Cu₂O content and morphology but also the synergistic effect of Cu₂O with TiO₂ because of the enhanced separation of photo-generated electron–hole pairs.

Fig. 8(A) gives the transient photocurrent response of different samples under visible light illumination. Sample CT(400) has the highest photocurrent density, indicating that Cu₂O content in TiO₂ NTAs has a important influence. When Cu₂O content in TiO₂ NTAs is low, then the photo-generated carriers are small, which lead to a low photocurrent density. When Cu₂O content is over a critical value, the photo-generated electrons hardly transfer to Ti electrode through TiO₂ nanotubes due to the enhanced recombination of electron and hole in larger Cu₂O particle aggregations and flakes as show as Fig. 8(C). The highest photocurrent density is about 0.9 mA cm⁻², which is three times of the reported value of 0.3 mA cm⁻² for Cu₂O decorated TiO₂ NTAs on Ti foil⁴² due to the significantly increased surface area in 3D-TiO₂ TNAs. Compared Fig. 8(A) to Fig. 7(A), one can find that sample CT(400) has the largest photocurrent response, however, sample CT(600) has the highest photocatalytic activity. In order to clarity the difference, we proposed a photocurrent response mechanism and a photocatalytic mechanism as shown in Fig. 8(C and D), respectively. Different from the photocurrent response, the photocatalytic activity increases with pulse cycles, namely, the photocatalytic activity increases with the content of Cu₂O flakes because the photo-generated carriers in the thin Cu₂O flakes can quickly transfer to the surface and react directly with MO in solution as shown in Fig. 8(D).

Fig. 8 (A) Transient photocurrent density of different samples under visible light illumination (a) TiO₂ NTAs (b) CT(200) (c) CT(400) (d) CT(600); (B) transport mechanism of carriers in Cu₂O/TiO₂ NTA hybrids; (C) and (D) are photocurrent response mechanism and photocatalytic mechanism for Cu₂O/TiO₂ NTA hybrids, respectively.

Conclusions

Cu₂O particles and flakes are successfully deposited into 3D-TiO₂ NTAs by electrochemical method. Cu₂O nanoparticles are firstly formed upon electrochemical deposition. Afterwards Cu₂O flakes appear on 3D-TiO₂ NTAs, which become longer and narrower with electrochemical cycles. Cu₂O modified 3D-TiO₂ NTAs expand the photo-response range from ultraviolet light to visible light. The Cu₂O modified NTAs with 400 pulses possesses the highest visible photocurrent density of 0.9 mA cm⁻², however, the Cu₂O modified NTAs with 600 pulses possesses the largest photocatalytic activity toward degradation of methyl orange (MO) under visible-light irradiation. Photocurrent response mechanism and photo-catalytic mechanism are proposed, respectively, to explain the difference in the photocurrent response and photocatalytic activity of Cu₂O/TiO₂ NTA hybrids.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (No. 50672069) and the Nanotechnology Special foundation of Shanghai (No. 11 nm0500700).

References

1. A. Fujishima and K. Honda, Nature, 1972, 238, 37.
2. Z. Liu, V. Subramanian and M. Misra, J. Phys. Chem. C, 2009, 113, 14028.
3. Z. Y. Liu, Q. Q. Zhang and T. Y. Zhao, J. Mater. Chem., 2011, 21, 10354.
4. J. J. Liao, S. W. Lin and L. Zhang, ACS Appl. Mater. Interfaces, 2012, 4, 171.
5. S. N. Frank and A. J. Bard, J. Am. Chem. Soc., 1977, 99, 303.
6. I. Paramasivalm, J. M. Macak and P. Schmuki, Electrochem. Commun., 2008, 10, 71.
7. D. Vanmaekelbergh and P. E. de Jongh, J. Phys. Chem. B, 1999, 103, 747.
8. A. R. Armstrong, J. Canales and R. Garcia, Adv. Mater., 2005, 17, 862.
9. Y. Ren, Z. Liu and F. Pourpoint, Angew. Chem., Int. Ed., 2012, 51, 2164.
10. K. Pan, Y. Dong and C. Tian, Electrochim. Acta, 2009, 54, 7350.
11. A. K. Chakraborty, Z. Qi and S. Y. Chai, Appl. Catal., B, 2010, 93, 368.
12. D. R. Rolison, Science, 2003, 299, 1698.
13. Y. Ohsaki, N. Masaki and T. Kitamura, Phys. Chem. Chem. Phys., 2005, 7, 4157.
14. Z. J. Zhang, Q. Y. Zeng and S. L. Chou, Electrochim. Acta, 2014, 133, 570.
15. H. S. Kim, J. Khamwannah and C. Choi, MRS Commun., 2013, 3, 235.
16. J. J. Liao, S. W. Lin and L. Zhang, ACS Appl. Mater. Interfaces, 2012, 4, 171.
17. W. He, J. Qiu and F. Zhuge, Nanotechnology, 2012, 23, 225602.
18. Z. Liu, Q. Zhang and T. Zhao, J. Mater. Chem., 2011, 21, 10354.
19. Y. B. Xie and X. Z. Li, Mater. Chem. Phys., 2006, 95, 39.
20. T. M. Brown, F. De Rossi and F. Di Giacomo, J. Mater. Chem. A, 2014, 2, 10788.
21. J. Y. Wang, G. B. Ji and Y. S. Liu, Catal. Commun., 2014, 46, 17.
22. S. Banerjee, S. K. Mohapatra and P. P. Das, Chem. Mater., 2008, 20, 6784.
23. S. K. Mohapatra, S. Banerjee and M. Misra, Nanotechnology, 2008, 19, 315601.
24. Y. Yin, Z. Jin and F. Hou, Nanotechnology, 2007, 18, 495608.
25. W. Wilson, A. Manivannan and V. R. Subramanian, Appl. Catal., A, 2012, 441–442, 1.
26. Q. Y. Wang, X. C. Yang and D. Liu, Electrochim. Acta, 2012, 83, 140.
27. J. A. Seabold, K. Shankar and R. H. T. Wilke, Chem. Mater., 2008, 20, 5266.
28. M. Zhang, Y. N. Wang and E. Moulin, J. Mater. Chem., 2012, 22, 10441.
29. Q. Y. Wang, X. C. Yang and D. Liu, Electrochim. Acta, 2012, 91, 330.
30. K. Zhu and A. J. Frank, MRS Bull., 2011, 36, 446.
31. Q. Wang, K. Zhu and N. R. Neale, Nano Lett., 2009, 9, 806.
32. M. L. Zhong, G. Q. Zhang and X. Q. Yang, Mater. Lett., 2015, 145, 216.
33. L. Huang, S. Zhang and F. Peng, Scr. Mater., 2010, 63, 159.
34. Y. Hou, X. Y. Li and X. J. Zou, Environ. Sci. Technol., 2009, 43, 858.
35. Y. G. Zhang, L. L. Ma and J. L. Li, Environ. Sci. Technol., 2007, 41, 6264.
36. Y. Hou, X. Y. Li and Q. D. Zhao, Appl. Phys. Lett., 2009, 95, 093108.
37. L. X. Yang, S. L. Luo, Y. Li and Y. Xiao, Environ. Sci. Technol., 2010, 44, 7641.
38. L. X. Yang, Z. Y. Li and H. M. Jiang, Appl. Catal., B, 2016, 183, 75.
39. L. K. Tsui and G. Zangari, Electrochim. Acta, 2014, 128, 341.
40. A. Jagminas, J. Kovger and A. Reza, Electrochim. Acta, 2014, 125, 516.
41. Q. Y. Wang, X. C. Yang, D. Liu and J. F. Zhao, J. Alloys Compd., 2012, 527(106), 111.
42. L. Wu, L. Tsui and N. Swami, J. Phys. Chem. C, 2010, 114, 11551.
43. Q. Zhang, N. Li and J. Goebl, J. Am. Chem. Soc., 2011, 133, 18931.
44. X. C. Yang, J. W. Hou and Y. Liu, Nanoscale Res. Lett., 2013, 8, 328.
45. A. Kellersohn, E. Knozinger and W. Langel, Adv. Mater., 1995, 7, 652.
46. S. S. Zhang, B. Y. Peng and S. Y. Yang, Int. J. Hydrogen Energy, 2013, 38, 13866.
47. B. Sinha, T. Goswami and S. Paul, RSC Adv., 2014, 4, 5092.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra11341a

---