Visible light photoelectrochemical properties of a hydrogenated TiO₂ nanorod film and its application in the detection of chemical oxygen demand

Abstract

A series of TiO₂ nanorod array electrodes with different lengths have been successfully fabricated by a controlled hydrothermal method for photoelectrochemical (PEC) application. In order to enhance the conductivity of the TiO₂ nanorods and enable the PEC activity under visible light irradiation, the TiO₂ nanorod samples have been further hydrogenated. The influence of the length of the hydrogenated TiO₂ nanorod arrays (H-TNRs) on their visible-light-driven photoelectrocatalytic activity was investigated. With increasing the length of the H-TNRs to about 3.0 μm, the activity was close to the maximum. Subsequently, the H-TNRs photoanode was fitted into a thin-layer photoelectrochemical cell for mineralization of organic compounds. The visible light PEC performance was enhanced so much that chemical oxygen demand (COD) detection was achieved under visible light as the light source for the first time. The excellent relationship between the photoelectrochemical COD and conventional COD values within the range of 0–288 mg L⁻¹ suggests that the simple visible light driven PEC method is a promising alternative to the conventional COD method.

---

1. Introduction

In the last decade, one-dimensional (1D) TiO₂ including nanorods, nanowires, and nanotubes has been extensively investigated since it has been confirmed to have superior photoelectrochemical (PEC) properties to other dimensional TiO₂ nanostructures.^1–6 Among them, single crystal TiO₂ nanorods (TNRs) or nanowires further offer direct electrical pathways for photogenerated electrons and could increase the oriented charge separation and the electron transport rate,^7–11 and the length of the nanorods is a key factor influencing the charge collection efficiency. Yang et al. reported that, with increasing the length of the single crystalline rutile TiO₂ nanorods to about 1.8 μm, the photocurrent density was close to the maximum.¹² Bian et al. investigated the effects of the length (0.68–2.3 μm) of TiO₂ nanorod on the PEC properties of TiO₂ nanorod/C dot nanocomposites, photocatalytic activity reached a maximum with the length of 1.3 μm.¹³ However, the mainly application of single crystalline rutile TNRs was photoelectrochemical water splitting^9,12,14–16 and solar cells.^7,8,17–19 The application in determination of chemical oxygen demand (COD) was seldom reported.

In recent years, a photoelectrochemical probe for rapid determination of COD is developed using a nanostructured TiO₂ photoanode, namely PeCOD.^20–24 As compared to the standard analytical method, PeCOD method is more promising because of the superior oxidative ability of illuminated TiO₂. Furthermore, TiO₂ nanomaterials are typically non-toxic, inexpensive, photostable, and environmentally benign. However, the drawback of TiO₂-based methods in practical applications is the easy recombination of the photogenerated electron/hole pairs in discrete TiO₂ nanoparticles and coated nanofilm, which results in low photocatalytic activity. Different from the TiO₂ coated on transparent conducting oxides (TCO) glass electrode, Zhou's group^24–26 fabricated TiO₂ nanotube arrays (TNTs) on Ti sheet as the COD sensor. The TNTs well suit for the efficient separation of photogenerated holes because the TNTs electrode is highly ordered with high aspect-ratio structures and nanocrystalline walls perpendicular to electrically conductive Ti substrates. Wang et al. investigated the application of TiO₂ nanorod arrays to the determination of COD and well correlated with those obtained using the conventional (i.e., dichromate) COD determination method.²² Li et al.²⁷ prepared hydrogenated TNTs (H-TNTs) to improve the conductivity and PEC efficiency, which was also used to detect COD under UV-light irradiation. However, both the TNTs, TNRs and coated-TiO₂ nanoparticles film electrode have an inevitable drawback that is the limiting use of the light source for the wide band gap of 3.2 or 3.0 eV.²⁸ It is necessary to use the UV light source to excite the photogenerated electron/hole pairs. The instrument user inevitable suffers UV irradiation. Therefore, some catalysts with visible-light-photoactivity as COD sensor have been studied presently. For example, Wang et al.²⁹ used the Cu₂O-loaded TiO₂ nanotube arrays electrode as PeCOD sensor to detect COD using visible light, however, the lifetime of the electrode was not desirable enough as the Cu₂O particles suffer from photocorrosion. WO₃/W nanopores sensor for COD determination under visible light has been reported, but the working range (3–60 mg L⁻¹) is too narrow.³⁰ In the previous reports, we prepared hydrogenated TiO₂ nanorod arrays (H-TNRs) used as a photoelectrode, which exhibited highly sensitive and steady photocurrent to the investigated organic compounds under visible light,³¹ indicating that it could be a promising photoelectrochemical sensor for organic compounds in water solution. However, visible light PEC properties of the H-TNRs film dependence on nanorod lengths and its application in the detection of COD have not been investigated.

Herein, we further investigated the influence of the length of the H-TNRs on their visible-light-driven photoelectrocatalytic activity and designed photoelectrochemical sensor with a thin-layer cell for COD detection. Different from the conventional TiO₂ photocatalytic sensor which needs the irradiation of UV light, the prepared H-TNRs electrode as photoelectrochemical sensor can detect COD value of the real samples using visible light.

2. Experimental

2.1 Preparation of H-TNRs

TNRs with different lengths were prepared on a fluorine-doped tin oxide (FTO) glass slide using our previous work.³¹ Briefly, 20.0 mL deionized water was mixed with 20.0 mL of concentrated hydrochloric acid (36.5 wt%) in a Teflon-lined stainless steel autoclave (100 mL). The mixture was stirred at ambient conditions for 5 min before the addition 0.48 mL of titanium butoxide (97% Aldrich). After stirring for another 5 min, two pieces of FTO substrates (5 × 2 cm²) were cleaned in a sequence of acetone, absolute alcohol and deionized water and dried in air, and subsequently placed into the Teflon-liner at an angle against the wall with the conducting side facing down. The hydrothermal synthesis was conducted at 170 °C for different time, i.e., 3 h, 6 h, 9 h and 12 h, and denoted as TNRs-3, TNRs-6, TNRs-9, and TNRs-12, respectively. The obtained samples were taken out, rinsed with deionized water, and dried in air. The H-TNRs electrodes were prepared via the hydrogenation process of the TNRs in argon (80 mL min⁻¹) and hydrogen (20 mL min⁻¹) mixed atmosphere at 350 °C in a tubular furnace under ambient pressure for 1 h with a heating rate of 5 °C min⁻¹.

2.2 Characterization of samples

The surface morphologies and cross-sectional structures of the samples were characterized by scanning electron microscope (FESEM, LEO 1530VP). X-ray diffraction (XRD) analysis was carried out with an X-ray diffractometer (D/max-IIIA, Japan), Cu Kα radiation (λ = 1.5418 Å) and a fixed power source (40 kV and 40 mA) were used. UV-vis diffuse reflectance spectroscopy (DRS) measurement was conducted by a UV-1601 spectrophotometer (Shimazu, Japan). The chemical nature of Ti and O was studied using X-ray photoelectron spectroscopy (XPS) in Krato Axis Ultra DLD spectrometer with Al Kα X-ray (hν = 1486.6 eV) at 15 kV and 150 W. The binding energy was referenced to C 1s line at 284.6 eV for calibration.

2.3 Apparatus and measurements

All photoelectrochemical experiments were performed at the room temperature (25 °C) in a thin photoelectrochemical cell with a quartz window for visible light illumination (see Fig. 1). It consists of a three-electrode system with an H-TNRs film working electrode, a saturated Ag/AgCl reference electrode and a platinum mesh counter electrodes. The flow path and the photoelectrochemical reaction zone were confined by a shaped spacer with thickness of 0.1 mm. A quartz window (0.5 × 1 cm²) was embedded into the front Teflon board and used to admit illumination. All interstices were sealed with silicone gel to keep the reactor airtight. The photocurrent–time curves in this work were measured using an electrochemical workstation (CHI 660B, Shanghai) coupled with a computer. A PLS-SXE300UV Xe lamp was used as the light source, and the UV part of the light was removed by a UV-400 filter when the process was carried out. The incident light intensity (100 mW cm⁻²) was measured by a radiant power meter (400–1000 nm, FZ-A radiation meter, Beijing Normal University).

Fig. 1 Schematic depiction of the thin-layer cell photoelectrochemical cell for continuous flow analysis.

3. Results and discussion

3.1 Structure characterizations

H-TNRs with different lengths were obtained by controlling the growth time at 170 °C. Fig. 2A–D show the top-down and cross section morphologies of H-TNRs-3, H-TNRs-6, H-TNRs-9 and H-TNRs-12, respectively. The corresponding average equivalent diameters of H-TNRs were measured to about 80, 170, 200 and 220 nm and the density of the nanorods decreased by increasing the hydrothermal synthetic time. Fig. 2A shows the H-TNRs-3 with the length of 0.6 μm has not obvious vertical array structure, while the nanorods begin to grow on the FTO substrate with the rod axis perpendicular to the substrate. And after 3 h, the top surface of the nanorods appears to contain many step edges, while the side surface is smooth, which is consistent with the work by Aydil et al.⁷ Moreover, from H-TNRs-6 to H-TNRs-12, the overall lengths of the nano-layer are about 1.8, 3.0 and 3.8 μm, respectively. However, except the underneath non-array structure layer, the lengths of nanorod arrays are about 1.2, 2.0 and 2.0 μm in H-TNRs-6, H-TNRs-9 and H-TNRs-12, respectively.

Fig. 2 FESEM images of (A) H-TNRs-3, (B) H-TNRs-6, (C) H-TNRs-9 and (D) H-TNRs-12. Insets are corresponding cross-sections.

Fig. 3A shows the XRD spectra of the FTO glass and different H-TNRs samples. After subtracting the diffraction peaks of FTO glass, two diffraction peaks centered at 2θ angles of 36.5° and 63.2° were observed in every sample, corresponding to (101) and (002) crystal planes of tetragonal rutile TiO₂, respectively (JCPDS no. 88-1175). The intensity of the (002) peak increases rapidly with the increase of synthetic time, suggesting that the TiO₂ nanorods are highly oriented in the (001) direction with respect to the FTO substrate surface and grow in the (001) direction with the growth axis parallel to the substrate surface normal.^7,32 Therefore, the length of TiO₂ nanorods increases with the synthetic time, which is consistent with the results from Fig. 2. In addition, there is no significant difference among the intensities of the (101) peaks of H-TNRs-6, H-TNRs-9 and H-TNRs-12, indicating that the equivalent diameters of the H-TNRs are similar (170–220 nm). The diffraction peaks of FTO glass decrease with the increase of the synthetic time due to the increase of the density and length of TiO₂ nanorods on the FTO glass. These results from XRD spectra confirmed that the rutile TiO₂ nanorods were grown directionally on the FTO glass and the hydrothermal synthetic time had an important influence on the TiO₂ growth.

Fig. 3 (A) XRD and (B) UV-vis DR spectra of the (a) FTO, (b) H-TNRs-3, (c) H-TNRs-6, (d) H-TNRs-9 and (e) H-TNRs-12.

Fig. 3B exhibits the UV-visible light absorption of the H-TNRs increases dramatically with the hydrothermal synthetic time from 3 to 9 h due to the more and longer TiO₂ nanorods on FTO. However, the absorption decreases slightly when the synthetic time reaches to 12 h. This can be explained by the observation on Fig. 2. Compared with H-TNRs-9, H-TNRs-12 possesses larger TiO₂ nanorods, which lead to the stronger light reflection and the reduction of the light diffusion. Consequently, the H-TNRs-12 has a slightly weaker light absorption than H-TNRs-9.

High-resolution Ti 2p_3/2 and O 1s XPS spectra of the H-TNRs-9 were displayed in Fig. 4. The Ti 2p_3/2 XPS spectrum can be separated two peaks at 458.3 and 457.6 eV, which can be ascribed to Ti⁴⁺ and Ti³⁺. The result demonstrates the presence of Ti–O and oxygen vacancies in TiO₂, which is in line with the literatures^2,33 and our previous reports.^27,31 The O 1s XPS spectrum can be deconvoluted into two peaks centered at 529.9 and 531.8 eV. One at 529.9 eV corresponds to the characteristic peak of Ti–O–Ti.³⁴ The other at 531.8 eV is attributed to Ti–OH, which has been reported to be located at ca. 1.5–1.8 eV higher binding energy corresponding to the O 1s of unhydrogenated-TiO₂.³² The result confirmed the formation of hydroxyl group on H-TNRs-9 surface.

Fig. 4 XPS spectra of (A) Ti 2p_3/2 and (B) O 1s of H-TNRs-9.

3.2 Photoelectrochemical properties and COD detection

The photocurrent density was tested in a thin photoelectrochemical cell. A 0.1 M NaNO₃ solution was used as supporting electrolyte. The photocurrent of TNRs without hydrogenation under visible light irradiation is approximately zero, as our reported before.³¹ Fig. 5A shows that the dark current is approximately zero under a constant applied potential of +0.10 V (vs. Ag/AgCl). However, under visible light illumination, the photocurrent spikes and then reaches a relatively steady state after 40 s for all the HTNRs samples, which is related to the following reasons. Hydrogenation of TiO₂ crystal can create oxygen vacancy and Ti³⁺ sites, leading to shallow donor states between the conduction band and valence band, namely mid-gap energy levels. This is responsible for the absorption of longer wavelength of light (>400 nm).^32,35 Moreover, hydrogen atoms can stay in TiO₂ crystal lattice as shallow donors enhancing the charge transport capability (electronic conductivity), promoting the PEC reaction.^31,36 The average steady photocurrent density is increasingly enhanced when the length of the H-TNRs changes from 0.6 to 3.0 μm. With the increase in the length of H-TNRs, the more charge carriers are generated by photo-stimulation due to the stronger visible-light absorption. When the overall length of H-TNRs increases from 3.0 μm to 3.8 μm, a decrease in photocurrent density is observed. Although the absorption of light decreases slightly, the photocurrent density decreases obviously. It can be explained that the photogenerated charge carriers in the top part of the H-TNRs-12 should diffuse a longer distance to reach the FTO film, and the recombination probability of the charge carriers during the transport is largely increased, leading to the significantly decreased photocurrent.

Fig. 5 (A) Photocurrent responses of (a) H-TNRs-3, (b) H-TNRs-6, (c) H-TNRs-9 and (d) H-TNRs-12; (B) photocurrent responses of H-TNRs-9 as a sensor in different concentrations of glucose; (C) quantitative relationship between Q_net and the concentration of organic compounds with H-TNRs-9 as a sensor; (D) quantitative relationship Q_net and the theoretical COD.

Fig. 5B shows the photocurrent–time plots of H-TNRs-9 electrode as a sensor to different concentrations of glucose. The photocurrent (I_background) observed for the blank solution (dash line) originates from the oxidation of water by the illuminated H-TNRs-9, whereas the photocurrent (I_total) observed in the organic (glucose) solution is the summation of two components (solid line), one arising from the oxidation of glucose and the other arising from the oxidation of water.^25,37 The decay of the I_total to the I_background level implies that all the glucose have been oxidized in the thin-cell reactor. Furthermore, the quantity of net charges (Q_net) originating from the photoelectrochemical oxidation of the glucose increases with the increasing concentration of glucose. It is also notable that the I_total for the all samples is the almost same value from 150 to 200 s, which means the excellent stability and reproducibility of H-TNRs.

Other three organic compounds (glycine, succinic acid and potassium hydrogen phthalate (KHP)) solutions were investigated and exhibited similar voltammograms to those shown in Fig. 5B (not shown here). The Q_net can be calculated according to Faraday's law, and the quantitative relationship between Q_net and the concentration of organic compounds have been established, as shown in Fig. 5C. It can be seen that for all organic compounds investigated, the quantity of net charges obtained are directly proportional to the concentration.

Since one oxygen molecule corresponds to the transfer of four electrons (eqn (1)), the measured Q_net value can be converted to an equivalent O₂ concentration (or oxygen demand). The equivalent COD value can then be represented as eqn (2).³⁸

O₂ + 4H⁺ + 4e⁻ → 2H₂O (1)

Fig. 5D has been derived from Fig. 5C by plotting the Q_net against theoretical COD values according to eqn (2). The plot reveals that all the compounds investigated here can be fitted by straight line y = 0.0886x + 0.0284 with a R² value of 0.996 in the range of 0–288 mg L⁻¹. The excellent linear relationship between Q_net and the theoretical COD suggests that the H-TNRs electrode can be adopted as visible-light-driven PeCOD sensor for various organic compounds.

Fig. 6 displays the correlation between the PeCOD method and standard dichromate method for COD measurement of the real samples, including two lake water, four civic sewage, four garbage leakage, and two dyeing wastewater. The trend line has a slope of 1.000 with correlation coefficient (R²) of 0.985. A comparison of two methods shows a highly significant correlation for both PeCOD and COD measurements, indicating that the simple and rapid PeCOD method using visible light as light sources is a reliable alternative to the conventional dichromate method.

Fig. 6 The correlation between the PeCOD method and conventional dichromate method for the real sample measurements.

4. Conclusions

In this work, the influence of the length of the hydrogenated TiO₂ nanorod arrays on their visible-light-driven photoelectrocatalytic activity has been investigated. The optimal sample, i.e., H-TNRs-9, has been used as a photoelectrochemical sensor for COD detection using visible light as light source for the first time, and provided a relative wide dynamic working rang of 0–288 mg L⁻¹. The results showed a good relationship between photoelectrochemical COD and practical COD values for the real samples. Moreover, the H-TNRs photoanode exhibited good stability and reproducibility. These advantages suggest that the hydrogenated TiO₂ nanorods would be a promising and practical photoanode as a new photoelectrochemical sensor for online COD monitoring using the visible light as the excitation light.

Acknowledgements

We acknowledge the financial support from the National Natural Science Foundation of China (No. 21328301), the Guangdong Provincial Natural Science Foundation (No. 2014A030312007) and the Fundamental Research Funds for the Central Universities (No. 2014ZM0052).

Notes and references

1. H. Liu, G. Xu, J. Wang, J. Lv, Z. Zheng and Y. Wu, Electrochim. Acta, 2014, 130, 213–221.
2. C. Zhang, H. Yu, Y. Li, Y. Gao, Y. Zhao, W. Song, Z. Shao and B. Yi, ChemSusChem, 2013, 6, 659–666.
3. K. Shankar, J. I. Basham, N. K. Allam, O. K. Varghese, G. K. Mor, X. J. Feng, M. Paulose, J. A. Seabold, K. S. Choi and C. A. Grimes, J. Phys. Chem. C, 2009, 113, 6327–6359.
4. J. Chen, Z. Xia, H. Li, Q. Li and Y. Zhang, Electrochim. Acta, 2015, 166, 174–182.
5. J. Y. Liao, B. X. Lei, H. Y. Chen, D. B. Kuang and C. Y. Su, Energy Environ. Sci., 2012, 5, 5750–5757.
6. J. I. Kim, J. M. Park, S. J. Hwang, M. J. Kang and J. C. Pyun, Anal. Chim. Acta, 2014, 836, 53–60.
7. B. Liu and E. S. Aydil, J. Am. Chem. Soc., 2009, 131, 3985–3990.
8. W. Yang, Y. Wang and W. Shi, CrystEngComm, 2012, 14, 230–234.
9. Y. Yang, G. Wang, Q. Deng, S. Kang, D. H. L. Ng and H. Zhao, CrystEngComm, 2014, 16, 3091–3096.
10. X. Feng, K. Shankar, O. K. Varghese, M. Paulose, T. J. Latempa and C. A. Grimes, Nano Lett., 2008, 8, 3781–3786.
11. H. S. Kim, D. T. Nguyen, E. C. Shin, J. S. Lee, S. K. Lee, K. S. Ahn and S. H. Kang, Electrochim. Acta, 2013, 114, 159–164.
12. Y. J. Hwang, C. Hahn, B. Liu and P. Yang, ACS Nano, 2012, 6, 5060–5069.
13. J. Bian, C. Huang, L. Wang, T. Hung, W. A. Daoud and R. Zhang, ACS Appl. Mater. Interfaces, 2014, 6, 4883–4890.
14. C. Wang, Z. Chen, H. Jin, C. Cao, J. Li and Z. Mi, J. Mater. Chem. A, 2014, 2, 17820–17827.
15. H. Wang, Y. Bai, H. Zhang, Z. Zhang, J. Li and L. Guo, J. Phys. Chem. C, 2010, 114, 16451–16455.
16. F. Su, T. Wang, R. Lv, J. Zhang, P. Zhang, J. Lu and J. Gong, Nanoscale, 2013, 5, 9001–9009.
17. G. Dai, L. Zhao, J. Li, L. Wan, F. Hu, Z. Xu, B. Dong, H. Lu, S. Wang and J. Yu, J. Colloid Interface Sci., 2012, 365, 46–52.
18. Z. J. Zhou, J. Q. Fan, X. Wang, W. Z. Sun, W. H. Zhou, Z. L. Du and S. X. Wu, ACS Appl. Mater. Interfaces, 2011, 3, 2189–2194.
19. M. Li, Y. Liu, H. Wang, H. Shen, W. Zhao, H. Huang and C. Liang, J. Appl. Phys., 2010, 108, 094304.
20. S. Zhang, D. Jiang and H. Zhao, Environ. Sci. Technol., 2006, 40, 2363–2368.
21. J. Bai and B. Zhou, Chem. Rev., 2014, 114, 10131–10176.
22. C. Wang, J. Wu, P. Wang, Y. Ao, J. Hou and J. Qian, Anal. Chim. Acta, 2013, 767, 141–147.
23. X. Li, W. Yin, J. Li, J. Bai, K. Huang, J. Li and B. Zhou, Water Environ. Res., 2014, 86, 532–539.
24. S. Yuan, R. Mao, Y. Li, Q. Zhang and H. Wang, Electrochim. Acta, 2012, 60, 347–353.
25. Q. Zheng, B. Zhou, J. Bai, L. Li, Z. Jin, J. Zhang, J. Li, Y. Liu, W. Cai and X. Zhu, Adv. Mater., 2008, 20, 1044–1049.
26. J. Zhang, B. Zhou, Q. Zheng, J. Li, J. Bai, Y. Liu and W. Cai, Water Res., 2009, 43, 1986–1992.
27. S. Li, J. Qiu, M. Ling, F. Peng, B. Wood and S. Zhang, ACS Appl. Mater. Interfaces, 2013, 5, 11129–11135.
28. A. Bendavid, P. J. Martin, A. Jamting and H. Takikawa, Thin Solid Films, 1999, 355, 6–11.
29. C. Wang, J. Wu, P. Wang, Y. Ao, J. Hou and J. Qian, Sens. Actuators, B, 2013, 181, 1–8.
30. X. Li, J. Bai, Q. Liu, J. Li and B. Zhou, Sensors, 2014, 14, 10680–10690.
31. S. Zhang, S. Zhang, B. Peng, H. Wang, H. Yu, H. Wang and F. Peng, Electrochem. Commun., 2014, 40, 24–27.
32. G. Wang, H. Wang, Y. Ling, Y. Tang, X. Yang, R. C. Fitzmorris, C. Wang, J. Z. Zhang and Y. Li, Nano Lett., 2011, 11, 3026–3033.
33. Y. Yan, M. Han, A. Konkin, T. Koppe, D. Wang, T. Andreu, G. Chen, U. Vetter, J. R. Morante and P. Schaaf, J. Mater. Chem. A, 2014, 2, 12708–12716.
34. X. Chen, L. Liu, P. Y. Yu and S. S. Mao, Science, 2011, 331, 746–750.
35. A. Naldoni, M. Allieta, S. Santangelo, M. Marelli, F. Fabbri, S. Cappelli, C. L. Bianchi, R. Psaro and V. Dal Santo, J. Am. Chem. Soc., 2012, 134, 7600–7603.
36. W. D. Zhu, C. W. Wang, J. B. Chen, D. S. Li, F. Zhou and H. L. Zhang, Nanotechnology, 2012, 23, 455204.
37. H. Zhao, D. Jiang, S. Zhang, K. Catterall and R. John, Anal. Chem., 2004, 76, 155–160.
38. S. Zhang, H. Zhao, D. Jiang and R. John, Anal. Chim. Acta, 2004, 514, 89–97.

---