Photoelectrochemical activities and low content Nb-doping effects on one-dimensional self-ordered Nb₂O₅–TiO₂ nanotubes

Abstract

Self-ordered Nb₂O₅–TiO₂ nanotube arrays were synthesized by an anodization method from a Ti–Nb alloy. Compared with pure TiO₂ nanotubes, Nb₂O₅–TiO₂ nanotubes show enhanced surface hydrophilicity and low charge-transfer resistance. The beneficial Nb-doping effects also exhibit an improved separation efficiency, leading to higher photocatalytic and photoelectrocatalytic activities.

---

Introduction

To date, many research studies have been devoted to one-dimensional self-ordered TiO₂ nanotubes since the first paper was published on the subject by Zwilling in 1999.¹ As an n-type functional semiconductor, and due to their good electrical transport property and large surface area, TiO₂ nanotubes have succeeded in capturing researchers' attention and have shown superior performance in applications such as photocatalysis, dye-sensitized solar cells, electrochromic devices, batteries, supercapacitors, drug delivery and bioelectrochemical sensors.^2,3 For further modification of the optical and electrical properties of TiO₂, a very wide range of elements (such as nonmetals N, C, B, S, P, F, etc. and transition-metal ions Cu, Co, Ni, Nb, Fe, Mo, etc.) are used in doping, e.g. by introducing new band energy or localized states.^4,5 As a promising semiconductor material with a band-gap of about 3.4 eV, niobium oxide (Nb₂O₅) takes part in a wide range of catalytic and photocatalytic activities, especially in the selective oxidation of benzylamine.⁶ In previous works,^7,8 different concentrations (from 0.05–1 wt%) of Nb₂O₅-doped TiO₂ nanotubes were grown by an electrochemical anodization method. By measuring the efficiency of the dye-sensitized solar cell and the photoresponse intensity, the optimal concentration was found to be 0.1 wt%. As reported,⁹ an enhanced optical band-gap was found by heavily Nb₂O₅-doped TiO₂ (from 1% to 30%), which is a drawback for photoelectrochemical applications.

The Nb-doping effect in solar cells is ascribed suppression of the recombination of charge carriers. However, a limited amount of studies have involved photocatalytic performance and the Nb beneficial effects of Nb₂O₅–TiO₂ compounds. Cui et al. demonstrated that the submonolayer surface coverage and acidity were increased by the introduction of 1.5–3 mol% Nb₂O₅, which led to the enhanced photocatalytic degradation of 1,4 dichlorobenzene.¹⁰ As is already known, photocatalytic processes involve redox reactions at the surface/interface of the photocatalyst and solution. Besides the changes in surface properties with Nb doping, the charge separation efficiency under light illumination also plays an important role in photocatalytic activity. Therefore, in this work we fabricate 0.1 wt% Nb₂O₅-doped TiO₂ nanotube layers and focus on the Nb-doping effects on the surface wettability, charge-transfer resistance, surface energy band, behaviours of photogenerated carriers and the corresponding results on photocatalytic activity.

Experimental

Nb₂O₅-doped TiO₂ nanotube arrays were prepared by the anodization of 0.1 wt% Ti–Nb alloy in a glycerol (65 vol%)–H₂O electrolyte containing 0.27 M NH₄F at 30 V for 3 h. Tube fabrication was carried out in a two-electrode system with a Pt sheet as the counter electrode. Pure TiO₂ nanotubes were grown in the same conditions and used as the reference. All the samples were annealed at 450 °C in air for 2 h for crystallization.

The morphologies of the samples were characterized by a field-emission scanning electron microscope (FE-SEM, Supra™55, Zeiss). X-ray diffractions analyses (XRD, D8-Advance, Bruker) with graphite monochromized Cu_Kα radiation (λ = 0.15406 nm) was carried out for detecting the crystal structure of the nanotubes. The surface hydrophilic properties were investigated on a contact angle meter (SL200B, Suolun Ltd., China). All the electrochemical measurements were carried out by an electrochemical station (CHI660D, ChenHua Ltd., China) with a Pt foil as the counter electrode and a Ag/AgCl electrode as the reference electrode. The surface contact potential difference (CPD) was obtained from a scanning Kelvin probe system (KP Technology Ltd., Scotland, UK). In a transient photovoltage (TPV) system,¹¹ the samples were excited by a 50 μJ laser pulse (wavelength of 355 nm and pulse width of 5 ns) from a third-harmonic Nd:YAG laser (Polaris II, New Wave Research, Inc.). The TPV signals were recorded by a 500 MHz digital phosphor oscilloscope (TDS 5054, Tektronix). The surface photovoltage spectroscopy (SPV)¹² consisted of a 500 W xenon lamp (CHFXQ500W, Global xenon lamp power), a monochromator (SBP500, Zolix) and a lock-in amplifier (SR830-DSP, Standford) with an optical chopper (SR540, Standford) running at 23 Hz. The AC photovoltage signals were detected from a sandwich-like holder (Cu/sample/ITO) and were recorded as a function of wavelength. The photoelectrocatalytic measurements were performed in 0.1 M Na₂SO₄ with a bias of +0.3 V. As light sources, monochromatic light of 365 nm with an intensity of 5 mW cm⁻² was filtered from a 300 W Xe lamp (I300C, Perfect Light Ltd., China). For evaluation of the photocatalytic activity, the concentration of an azo-dye (acid orange 7-AO7, C₁₆H₁₁N₂O₄SNa) with 1.5 × 10⁻⁵ mol L⁻¹ was determined by periodically measuring the absorbance at 486 nm using a UV/Vis spectrophotometer (Lambda XLS+, PerkinElmer, USA).

Results and discussion

Fig. 1a and b show the morphologies of pure TiO₂ and Nb₂O₅–TiO₂ nanotube arrays, which were grown in glycerol–H₂O contained in NH₄F electrolyte. After 3 h anodization, both of their tubes are open with a length of 1 μm and diameters of around 100 nm. No significant changes were observed in the surface morphology before and after doping, as there is only a low concentration of Nb in the Ti–Nb alloy. This is consistent with the results of previously grown nanotubes in ethylene glycol contained in NH₄F electrolyte.⁸ The corresponding XRD patterns of pure TiO₂ and Nb₂O₅–TiO₂ nanotubes annealed at 450 °C are shown in Fig. 1c. The as-formed nanotubes can be converted from amorphous to the anatase phase with heat treatment. No clear peaks assigned to Nb₂O₅ were observed, due to the low Nb₂O₅ concentration. Previous reports confirmed by XPS analysis that Nb exists in the form of Nb₂O₅ after anodization.^8,13 The inset of Fig. 1c shows the magnified XRD patterns in the range of 2θ = 24–27°. It was found that after Nb⁵⁺ doping the peaks assigned to the anatase phase shifted towards higher angles, which suggests a variation in the d-spacing due to the occupation of Nb⁵⁺ ions in the TiO₂ matrix.¹⁴

Fig. 1 SEM images of TiO₂ nanotubes (a) and Nb₂O₅–TiO₂ nanotubes (b). Inset: cross-section images of nanotube layers. (c) XRD patterns of TiO₂ nanotubes and Nb₂O₅–TiO₂ nanotubes after annealing at 450 °C. Inset of (c): magnified XRD patterns in the range of 2θ = 24–27°.

Fig. 2a shows the contact angles of a water droplet on pure TiO₂ and Nb₂O₅–TiO₂ nanotube layers calcined at 450 °C. Both of the anodic films exhibit surface hydrophilicity. Compared with pure TiO₂ (contact angle of 22.35°), Nb₂O₅-doped TiO₂ nanotubes have a smaller contact angle (10.46°), which reveals the more hydrophilic surface and better wettability. Parallel experimental results showed the same trend as displayed in the inset of Fig. 2a. Nb⁵⁺ ion replaces Ti⁴⁺ in the lattice as an n-type dopant to form hybridized Nb4d–Ti3d states in the conduction band¹⁵ after anodization of the Ti–Nb alloy. This could bring high oxidation states and a higher proportion of hydroxyl groups on to the surface,¹⁶ which lead to an improved wettability of the Nb₂O₅–TiO₂ nanotube. The Nyquist impedance plots of the TiO₂ and Nb₂O₅–TiO₂ nanotube arrays are shown in Fig. 2b. The semicircle diameter represents the charge-transfer resistance (R_ct) value for the Fe(CN)₆^3−/4− probe. Clearly, with 450 °C heat-treatment a remarkable decrease in the semicircle diameter of the Nb₂O₅–TiO₂ nanotube was obtained, which indicates that the electron transfer resistance at the electrode/electrolyte interface is decreased by Nb doping. In other words, the electrochemical oxidation–reduction reaction rate of the Fe(CN)₆^3−/4− probe on the Nb₂O₅–TiO₂ nanotubes is faster than pure TiO₂. To further clarify the Nb₂O₅ doping effect on the band structure, a Mott–Schottky (M–S) analysis was performed. According to classical Mott–Schottky theory,¹⁷ for n-type semiconductors the charge carrier density can be determined by complex impedance measurements using the following expression:

here, C is the differential capacitance of the space-charge region and N_d is the carrier concentration. ε denotes the dielectric constant of the semiconductor (ε = 41.4 for antase¹⁸), ε₀ the permittivity of free space (8.854 × 10⁻¹² F m⁻¹), E the applied potential and E_FB the flat-band potential. A is the surface area of the sample (1 cm² for nanotube layers). The quantities q, k and T represent the elementary charge (1.6 × 10⁻¹⁹ C), the Boltzmann's constant, and temperature in K, respectively. In Fig. 3c, an obvious decrease in the slope in the M–S plot is indicative of the increasing carrier concentration after Nb₂O₅ doping. The carrier concentrations N_d of the Nb₂O₅–TiO₂ nanotube and pure TiO₂ nanotube can be calculated by using the extracted slope of the linear regions from Fig. 2c, which are 1.333 × 10¹⁸ cm⁻³ and 6.656 × 10¹⁷ cm⁻³, respectively.

Fig. 2 (a) Contact angle images of TiO₂ nanotubes and Nb₂O₅–TiO₂ nanotubes with deionized water droplets. (b) Nyquist plots of pure TiO₂ nanotubes and Nb₂O₅–TiO₂ nanotubes obtained from 0.1 M KCl containing 2.0 mM Fe(CN)₆^3−/4− electrolyte with an amplitude of 5 mV in the frequency range of 0.01–100 000 Hz. (c) Mott–Schottky plots of Nb₂O₅–TiO₂ and pure TiO₂ nanotube arrays in 0.5 M Na₂SO₄ electrolyte recorded at a frequency of 1 kHz.

Fig. 3 The contact potential differences (CPD) of TiO₂ and Nb₂O₅–TiO₂ nanotubes.

To further verify the Nb₂O₅ doping effects on the energy band, the surface work functions of the samples were characterized by Kelvin probe measurements.¹⁹ Fig. 3 shows the CPD 3D-distribution maps of the pure TiO₂ nanotube and the Nb₂O₅–TiO₂ nanotube after 450 °C annealing. The data were collected by a Au reference probe (surface work function of 5.1 eV) with scanning steps of 100 μm. It can be seen that due to the surface roughness of both of the anodic layers, the CPDs show a slight change in the values of the scanning area. Compared to the pure TiO₂ nanotube, a remarkable decrease of CPD is observed in the Nb₂O₅–TiO₂ nanotubes. The relationship between the surface work function and the CPD is as follows:

CPD = Φ(sample) − Φ(Au) = Φ(sample) − 5.1 eV

where Φ is the surface work function. We calculated the surface work functions of the pure TiO₂ nanotube and Nb₂O₅–TiO₂ nanotube as approx. 5.15 eV and 5.0 eV, respectively. That is with 0.1 wt% Nb doping, the apparent feature in the energy band is the Fermi level, which moves upward to the conduction band of TiO₂, because of the strong hybridization between Ti and Nb. This shift of the Fermi level leads to an increase in carrier concentration in the conduction band. The Nb-doping effects on the electric band structure, the carrier concentration, and on the resistivity were verified by both theoretical calculations and experimental results,^15,20 which were found to be in good agreement with the CPD change and our Mott–Schottky results.

The charge separation efficiency plays an important role in photocatalytic activity. Here, surface photovoltage spectroscopy (SPS) and transient photovoltage measurement (TPV) were used to investigate the behaviours of the photoinduced carriers. When the samples were excited by light with energy larger than the band-gap, massive photogenerated carriers were generated and then separated by the built-in electric field. That is, photoinduced electrons and holes transfer in opposite directions, resulting in the change of the surface band bending and the generation of surface photovoltage signals. Fig. 4a shows the SPS spectra of Nb₂O₅–TiO₂ and TiO₂ nanotube arrays as a function of wavelength. In both cases, the SPV signals give positive responses in the range of 300–400 nm, which indicates under ultraviolet illumination that band-to-band transition occurs and photoinduced holes move to the surface. The 0.1 wt% Nb₂O₅-doped TiO₂ nanotube layer shows an approx. 6 times higher intensity of surface photovoltage response than pure TiO₂ and a slight peak red-shift. Since no significant differences in the nanotube morphology and thickness are found before or after Nb doping, this increased surface photovoltage signal can be assigned to Nb's beneficial effect in the separation and transport of photogenerated carriers in TiO₂ nanotubes. The inset of Fig. 4a shows photovoltage transients of both nanotube layers under 355 nm illumination. A clear photovoltage response is generated or vanishes with the monochromatic light on or off. It is apparent that the Nb₂O₅–TiO₂ nanotube exhibits a considerably strong and stable photovoltage intensity, which is consistent with the results in Fig. 4a. In addition, according to a previous report,²¹ an about 2.5 times higher photocurrent was obtained from Nb₂O₅–TiO₂ nanotubes compared with non-doped TiO₂ nanotubes. This can further support the improved surface photovoltage intensity. To further understand the kinetics of the photo-generated charge transfer by the Nb doping effect, TPV measurements were conducted under the illumination of 355 nm laser pulse, as shown in Fig. 4b. The TPV signals were recorded in the range of 1 ns to 0.1 s. We obtained the information on the generation, separation, and recombination of photoinduced charges directly. As seen, both of the TPV signals show positive values, indicating positive charges transfer to the surface.²² Additionally, a higher photoresponse was achieved in Nb₂O₅–TiO₂ nanotube arrays, which suggests more holes were transferred to surface by the Nb-doping effect. It is well known that TiO₂ has a large Maxwell relaxation time,²³ where the time of the TPV maximum (t_max) is affected by the lifetime of the charge carriers. As expected, the t_max of the Nb₂O₅–TiO₂ nanotube is prolonged to longer timescales than that of pure TiO₂. This demonstrates that trace amounts of Nb dopant can suppress the recombination of photogenerated carriers in excited TiO₂ nanotube arrays. It is also worth noting that a shoulder peak appears in the range of 10⁻⁵ s to 10⁻² s only in the Nb₂O₅–TiO₂ nanotube TPV spectrum. As previously reported,^23,24 this slow process (timescale is longer than 10⁻⁵ s) can be assigned to the diffusion of photogenerated carriers under concentration gradients. Therefore, we can conclude that the Nb dopant promotes the further separation of electron–hole pairs.

Fig. 4 Surface photovoltage spectra (a) and transient photovoltage responses (b) of TiO₂ and Nb₂O₅–TiO₂ nanotubes. Inset: transient surface photovoltage responses under the illumination of 355 nm.

Fig. 5 shows the photocatalytic (PC) and photoelectrocatalytic (PEC) activity of TiO₂ and Nb₂O₅–TiO₂ nanotube arrays, with error analysis. In the PEC measurement, the applied bias of +0.3 V was chosen to ensure 100% of the photogenerated electrons were removed from the conduction band of TiO₂.²⁵ Clearly, a faster degradation efficiency was obtained in the PEC process, because the applied positive bias can effectively promote the photogenerated electrons to the external circuit, and avoid photogenerated electrons recombining with oxidative intermediates. Furthermore, regarding both the PC and PEC data, the Nb₂O₅–TiO₂ nanotube layer exhibits a higher degradation efficiency than pure TiO₂. This result is in line with the SPV and TPV trend, which suggests that the presence of Nb₂O₅ in TiO₂ can affect the separation of photoinduced carriers.

Fig. 5 Photocatalytic and photoelectrocatalytic degradation of Acid Orange 7 (AO7, C₁₆H₁₁N₂O₄SNa, C₀ = 15 μM in aqueous solution) on TiO₂ and Nb₂O₅–TiO₂ nanotubes under light of 365 nm.

Conclusions

In summary, Nb₂O₅–TiO₂ nanotube layers were fabricated by electrochemical anodization. The findings in this work have revealed a considerable beneficial effect of a small (0.1 wt%) Nb₂O₅ addition on the surface hydrophilic property, charge transfer resistance, and surface energy band structure. In addition, with Nb₂O₅ doping, the charge separation efficiency of photogenerated carriers is enhanced, according to the measurements of SPV and TPV, which results in higher photocatalytic and photoelectrocatalytic activities in TiO₂ nanotubes.

Acknowledgements

For financial support the authors are thankful to the National Natural Science Foundation of China (no. 21203043), China Postdoctoral Science Foundation (no. 2012M520729 and no. 2013T60361), Postdoctoral Research Foundation of Heilongjiang Province (LRB12-008) and Initial Research Funding for talents by Harbin Institute of Technology. The authors would like to acknowledge Prof. Zhonghua Li (Harbin Institute of Technology, China) and Prof. Dejun Wang (Jilin University, China) for technical supports.

Notes and references

1. V. Zwilling, E. Darque-Ceretti, A. Boutry-Forveille, D. David, M. Y. Perrin and M. Aucouturier, Surf. Interface Anal., 1999, 27, 629–637.
2. K. Lee, A. Mazare and P. Schmuki, Chem. Rev., 2014, 114, 9385–9454.
3. I. Paramasivam, H. Jha, N. Liu and P. Schmuki, Small, 2012, 8, 3073–3103.
4. Y. C. Nah, I. Paramasivam and P. Schmuki, ChemPhysChem, 2010, 11, 2698–2713.
5. J. H. Park, S. Kim and A. J. Bard, Nano Lett., 2006, 6(1), 24–28.
6. S. Furukawa, Y. Ohno, T. Shishido, K. Teramura and T. Tanaka, ACS Catal., 2011, 1, 1150–1153.
7. M. Yang, H. Jha, N. Liu and P. Schmuki, J. Mater. Chem., 2011, 21, 15205–15208.
8. M. Yang, D. Kim, H. Jha, K. Lee, J. Paul and P. Schmuki, Chem. Commun., 2011, 47, 2032–2034.
9. D. Kurita, S. Ohta, K. Sugiura, H. Ohta and K. Koumoto, J. Appl. Phys., 2006, 100, 096105.
10. H. Cui, K. Dwight, S. Soled and A. Wold, J. Solid State Chem., 1995, 115, 187–191.
11. X. Wei, T. F. Xie, D. Xu, Q. D. Zhao, S. Pang and D. J. Wang, Nanotechnology, 2008, 19, 275707.
12. Q. D. Zhao, D. J. Wang, L. L. Peng, Y. H. Lin, M. Yang and T. F. Xie, Chem. Phys. Lett., 2007, 434, 96–100.
13. M. Z. Atashbar, H. T. Sun, B. Gong, W. Wlodarski and R. Lamb, Thin Solid Films, 1998, 326, 238–244.
14. J. T. Park, W. S. Chi, H. Jeon and J. H. Kim, Nanoscale, 2014, 6, 2718–2729.
15. X. D. Liu, E. Y. Jiang, Z. Q. Li and Q. G. Song, Appl. Phys. Lett., 2008, 92, 252104.
16. I. P. Parkin and R. G. Palgrave, J. Mater. Chem., 2005, 15, 1689–1695.
17. R. O'Hayre, M. Nanu, J. Schoonman and A. Goossens, J. Phys. Chem. C, 2007, 111, 4809–4814.
18. J. Y. Kim, H. S. Jung, J. H. No, J. R. Kim and K. S. Hong, J. Electroceram., 2006, 16, 447–451.
19. Q. D. Zhao, T. F. Xie, L. L. Peng, Y. H. Lin, P. Wang, L. Peng and D. J. Wang, J. Phys. Chem. C, 2007, 111, 17136–17145.
20. Y. Furubayashi, T. Hitosugi, Y. Yamamoto, K. Inaba, G. Kinoda, Y. Hirose, T. Shimada and T. Hasegawa, Appl. Phys. Lett., 2005, 86, 252101.
21. C. Das, P. Roy, M. Yang, H. Jha and P. Schmuki, Nanoscale, 2011, 3, 3094–3096.
22. T. F. Jiang, T. F. Xie, L. P. Chen, Z. W. Fu and D. J. Wang, Nanoscale, 2013, 5, 2938–2944.
23. V. Duzhko, V. Y. Timoshenko, F. Koch and T. Dittrich, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 64, 075204.
24. S. Li, L. J. Zhang, T. F. Jiang, L. P. Chen, Y. H. Lin, D. J. Wang and T. F. Xie, Chem.–Eur. J., 2014, 20, 311–316.
25. X. L. Liu, H. M. Zhang, C. Liu, J. Y. Chen, G. Y. Li, T. C. An, P. K. Wong and H. J. Zhao, Catal. Today, 2014, 224, 77–82.

---