Exploring indium tin oxide capped titanium dioxide nanolace arrays for plasmonic photocatalysis

Abstract

We report a convenient nanotechnique to fabricate indium tin oxide (ITO) capped titanium dioxide (TiO₂) nanolace arrays for plasmonic photocatalysis. The topography can be tuned to optimize the photocatalytic efficiencies by adjusting the anodic voltage or the time of ITO magnetron sputtering. Microstructural analysis and finite difference time domain simulations indicate the generation of ITO surface plasmon resonance in the near-infrared region and ITO–TiO₂ heterojunctions. The enhanced photocatalytic performance of such ITO–TiO₂ composite nanostructures could spur further interest in the utilization of the near-infrared region for plasmonic photocatalysis.

---

1 Introduction

Plasmonic photocatalysis is a very promising technique enhancing the photocatalytic efficiency of titanium dioxide (TiO₂).^1–5 It is found to increase the prospect of a wide range of environmental and energy applications such as wastewater treatment,^6–8 water splitting,^9,10 CO₂ reduction^11,12 and degradation of organic molecules.^13,14 Generally, plasmonic photocatalysts are mixtures of semiconductors and noble metal nanoparticles.¹ The nanoparticles improve the photocatalytic efficiency by trapping the photo-generated electrons in TiO₂, inhibiting the charge recombination process, and extending TiO₂ absorption to the visible light range. The improvement mainly benefits from the collective oscillation of free electron in plasmonic materials, namely surface plasmon resonance (SPR), in phase with the varying electric field of the incident light. The SPR creates an intensive local electromagnetic (EM) field which favors the photocatalytic reaction in several ways, including powering the excitation of electron–hole pairs and heating up the surrounding environment to increase the redox reaction and the mass transfer.⁶ Conventional plasmonic materials using the three coinage metals (Au, Ag, and Cu) which are conceived for practical applications, however, are in challenge in long-time photocatalysis experiments due to chemical and thermal unstableness.¹⁵

Indium tin oxide (ITO) is a possible alternative plasmonic material with tunable SPR frequencies in the near-infrared region.^16,17 It is supposed to widely broaden absorption band of TiO₂ to the infrared. Moreover, one of the distinct advantages of ITO is its optical transparency and chemical stableness.¹⁸ Although ITO have potentials and advantages in high-performance photocatalysis, few efforts have been made to explore its feasibility and the extent of its effects. In this paper, we describe a convenient approach to fabricate ITO capped TiO₂ nanolace arrays as cost-efficient plasmonic photocatalysts. It is notable that the lace-like nanostructures provide a very large lateral aspect ratio due to its two-layered morphology.¹⁹ The topography can further be tuned to optimize the plasmonic photocatalytic activities by adjusting anodic voltages or the time of ITO magnetron sputtering.

2 Experimental

2.1 Fabrication of TiO₂ nanolaces

Schematic of TiO₂ fabrication process is shown in Fig. 1. High purity titanium (Ti) foils (99.99%, 10 mm × 10 mm × 1 mm in size) were degreased by acetone, followed by chemical polishing for 5 min in a mixture of hydrofluoric acid (40 wt%), nitric acid (65 wt%), and water with volume ratio of 1 : 4 : 5, to further remove surface impurities (Fig. 1a). After rinsing in distilled water and drying, the Ti foils were anodized separately in a fresh electrolyte under a constant DC voltage of 40 V (50 V, 60 V, 70 V and 80 V) at 25 °C, respectively. The electrolyte consists of a mixture of ethylene glycol and deionized water with volume ratio of 100 : 2, containing 0.3 wt% NH₄F. In order to obtain regular TiO₂ nanolace arrays, a two-step anodizing process was adopted. The Ti foils were first anodized for 3 h (Fig. 1b), followed by polarizing with a cathodic pulse in 1 M sulfuric acid at −3 V to remove the TiO₂ nanotubes layer (Fig. 1c). Subsequently, the Ti foil were immediately anodized at the half voltage of the first step for 1 min, then at the same voltage as the first step for 1 h and finally, TiO₂ nanolaces were obtained (Fig. 1d).

Fig. 1 Schematics of the fabrication method of ITO capped TiO₂ nanolace arrays: (a) pretreated Ti foils; (b) TiO₂ nanotube arrays after first-step anodization; (c) patterned Ti foils after the removal of TiO₂ layer; (d) TiO₂ nanolace arrays after the second-step anodization; (e) ITO capped TiO₂ nanolace arrays; (f) the top view of TiO₂ nanolace arrays.

2.2 Deposition of ITO films

ITO (10% Sn-doped) films were prepared on the TiO₂ lace-like templates by magnetron sputtering (MSP-300CT) in an oxygen/argon plasma with a mixing ratio at 1 : 32 at room temperature (Fig. 1e). The power of the magnetron was 60 W. During the deposition process, the pressure in the magnetron chamber was 1 Pa and the distance between the target and substrate was 10 cm. Based on the given parameters, the deposition rate was calculated to be 0.125 nm s⁻¹. In order to acquire TiO₂ in anatase phase and improve the crystallization of ITO films, all samples were annealed at 400 °C for 0.5 h in vacuum.

2.3 Characterization

Scanning electron microscopy (SEM, FEI Inspect F50) was used to investigate the morphologies of the photocatalysts. X-ray diffraction (XRD, Smartlab, Cu K_α1) was performed to investigate the crystal structure. Commercial finite-difference time domain (FDTD) software (CST MWS 2010) was employed to calculate the EM field properties of the loaded ITO film. The Drude model was employed to describe the properties of ITO film and the parameters of nanostructures in the FDTD simulations were set the same as experimental measurements.

2.4 Evaluation of photocatalytic activities

The photocatalytic activities of the ITO capped TiO₂ nanolace arrays were evaluated via the photo-decomposition of methyl orange (MO) under irradiation of a 300 W Xe lamp at room temperature. In a typical experiment, the sample was settled in 10 mL MO aqueous solution with a concentration of 1 × 10⁻⁵ M in a 10 mL beaker. The active area of the photocatalysts used for each experiment was about 100 mm² (10 mm × 10 mm). Prior to irradiation, the reaction system was kept in dark for 20 min to establish adsorption/desorption equilibrium between the dye and the surface of the photocatalysts under room air equilibrated conditions. At given irradiation intervals, the solution was collected for analysis. The residual MO concentration was monitored using a UV-Vis spectrophotometer (Hitachi U-3900).

3 Results and discussion

XRD was performed to investigate the crystalline phase and crystal structure of the as-prepared and annealed TiO₂ nanolaces. The as-prepared samples are amorphous due to no diffraction peaks associated with crystalline ITO or TiO₂ shown in Fig. 2. For annealed samples, all of the intensive diffraction peaks could be fully assigned to well-crystallized ITO and anatase phase TiO₂ except peaks corresponded to underlying Ti substrate. The sharp peaks at 2θ = 25° and 48° are indexed to TiO₂ anatase phase (101) and (200). Anatase is generally recognized to be the most active in photocatalysis among the common crystal phases of TiO₂.^20–22

Fig. 2 XRD spectra of as-prepared and annealed ITO capped TiO₂ nanolaces.

The top view and cross-sectional view morphologies of the self-organized TiO₂ nanolaces before and after ITO deposition were examined using SEM. Fig. 3a presents a top view of a 60 V as-synthesized TiO₂ nanolace array, showing a regularly arranged multilayer structures with a flat hexagonal TiO₂ crystallographic plane. The cell size of upper lace layer is about 180 nm and of the lower nanotubes is 120 nm. The two-layer structure is also clearly observed from the cross-sectional view in Fig. 3c. The nanolace structure extending over the entire sample surface provides a very large lateral aspect ratio benefitting molecule adsorption in photocatalysis.¹⁹ Fig. 3b and d shows SEM images of TiO₂ nanolaces after ITO deposition, indicating that well-ordered structure still exists after ITO sputtering and annealing process.

Fig. 3 Top view SEM images of (a) a 60 V TiO₂ nanolace array; (b) a 60 V ITO capped TiO₂ nanolace array; and cross-sectional view SEM images of (c) a 60 V TiO₂ nanolace array; (d) a 60 V ITO capped TiO₂ nanolace array.

A series of ITO capped TiO₂ nanolace arrays with different cell sizes were obtained by adjusting the voltage of anodization and subsequently the deposition processing (ITO layer of 60 nm thick). Fig. 4a depicts typical SEM images and the functional relationship between the values of the cell size D and first step anodic voltage V_1st. In the SEM results, each hexagonal cell is surrounded by six equivalent adjacent ones forming the nanostructured unit and the average value of D is a function of V_1st with a slope of 2 nm V⁻¹ in the voltage region of 40–80 V.

Fig. 4 (a) Average D size as a function of V_1st and corresponding SEM images. The scale bar is 100 nm. (b) Representative UV-Vis spectra changes during MO degradation by ITO capped TiO₂ nanolace arrays. (c) The functional relationships between ln(c₀/c) and reaction time t for different V_1st (40 V, 50 V, 60 V, 70 V and 80 V, respectively). (d) The relationships of reaction rate k and V_1st of ITO capped TiO₂ nanolace arrays.

Fig. 4b shows the typical absorption spectra of the MO solutions (initial concentration of 10⁻⁵ M) with different irradiation time in the presence of as-prepared photocatalysts. Here the concentration and absorbency of solution have good linear relationship, indicating the concentration of MO decreases linearly with increase of irradiation time. The photocatalytic decomposition of MO is a pseudo-first-order reaction and its kinetics might be expressed as follows:

where k is the apparent reaction rate constant; c₀ and c are the initial concentration and the reaction concentration of MO, respectively. In our experiments, the change in intensity of MO absorption peak at λ = 463.5 nm is used to determine the reaction rate. Fig. 4c shows the plots of ln(c₀/c) versus reaction time t of ITO capped TiO₂ nanolace arrays at different V_1st = 40, 50, 60, 70, 80 V, and the solid lines are the fitting results. Values of k were obtained by calculating the slope of the fitting lines and summarized in Table 1. The reaction rates of samples are given in Fig. 4d, indicating that k increases and reaches the maximum at 70 V with the increase of V_1st, and then decreases.

Table 1 Calculated data of k under different V_1st

V1st/V	40	50	60	70	80
k/10^−2 min^−1	0.286	0.331	1.180	1.210	0.791

---

The photocatalytic activities are surely enhanced after ITO film loading on the surface of TiO₂ nanolaces, and the underlying mechanisms can be explained mainly by plasmonic effect of ITO nanolace arrays. Plasmonic photocatalysis makes use of plasmonic materials to strongly absorb light and then transfer the active charge carriers to semiconductor photocatalysts, leading to the improvement of photocatalytic activity.⁶ ITO has been proposed to show plasmonic effect with tunable SPR frequencies in the near-infrared region.¹⁶ The plasmonic effects couple with TiO₂ can extend the spectral response of TiO₂ and inhibit the charge recombination process.^1,2

One mechanism responsible for the enhanced photocatalytic activity of ITO–TiO₂ nanolace composite is photo-generated exciton separation in ITO–TiO₂ heterojunctions. It is well known that a vectorial transfer of electrons and holes from one semiconductor to another can result in the improvement of photocatalytic efficiency in a coupled semiconductor system with favorable energetics.²³ ITO and TiO₂ are both n-type semiconductors, so they form n–n type heterojunction owing to the direct contact.²⁴ The electrons excited to conduction band from valence band of TiO₂ by near-UV light can flow along the band bending to the conduction band of ITO, thus facilitating the charge excitons separation.²⁵ Therefore, there are more holes on TiO₂ surface which can oxidize the adsorbents and decompose MO molecules subsequently.

Another possible mechanism responsible for the enhanced photocatalytic activity is the plasmonic effect of coated ITO films. The SPR modes of ITO depend on the coupling effect among adjacent ITO nanoparticles¹⁶ and can be tuned by optimizing the nanostructures. In our experiments, ITO films composed of ITO nanoparticles were prepared with different cell sizes, thus the SPR modes can be altered. FDTD method was employed to calculate the local EM-field distribution of the ITO capped nanolaces. In the simulation process, the ITO capped nanolace arrays are approximated by the fundamental nanostructure units composed of seven hexagonally arranged nanoholes and the dimensional parameters of simulated ITO capped nanolace arrays were equal to the average values of the samples fabricated experimentally. Additionally, the incident light was assumed to be normal to the sample surface and the wavelength range was set to vary from 400 to 2000 nm. The relationship of calculated maximal local electric intensity at different wavelength is given in Fig. 5a. Obviously, the incident radiation excites SPR in near-infrared region which creates an enhanced EM field. A typical planar view of the calculated local electric field components at the ITO–TiO₂ interface when the wavelength of incident light was set to be 2000 nm is displayed in the inset of Fig. 5a. The enhancement factor (EF) is simplified as the fourth power of the local electric field enhancement: EF = (E_loc/E₀)⁴. In this case, the calculated data of the EF are summarized in Table 2. Fig. 5b shows the reaction rates and the EF as a function of V_1st, demonstrating good correlation between experiments and theory. Our simulation results indicate that ITO films surely have response in near-infrared region and thus SPR of ITO in near-infrared region can generate “hot” electrons, facilitating the reduction reaction process on the surface of ITO, then boosting the photocatalytic activity of ITO capped TiO₂ nanolace arrays as well.

Fig. 5 (a) The relationship of calculated maximal local electric intensity at different wavelength. The inset in (a) shows a typical planar view of the calculated local electric field components at the ITO–TiO₂ interface. The scale bar is 100 nm. (b) The reaction rates and the EFs as a function of V_1st.

Table 2 Calculated data of the maximum local electric field under different V_1st

V1st/V	40	50	60	70	80
EF	1.937	1.991	2.405	2.465	2.263

---

Apart from the observation that the ITO capped TiO₂ nanolace arrays can be tuned by varying anodic voltage to optimize the plasmonic photocatalysis, we also increased the deposition thickness of ITO from 5 nm to 120 nm to tailor the geometries of ITO films (V_1st = 60 V). The photocatalytic ability to decompose MO molecules of these samples was experimentally evaluated (as shown in Fig. 6). The tests of each sample were repeated three times to avoid random errors. When the concentration of ITO is low, the increase of ITO loading results in larger interface area between TiO₂ and ITO as well as more “hot electrons”, which facilitate the electron transfer and enhance the photocatalytic activity. However, as more ITO is loaded on TiO₂ nanolace arrays, less TiO₂ is exposed to MO molecules, which results in less electron transferred from TiO₂ to MO and thus reduces the photocatalytic activity.

Fig. 6 Comparison of photocatalytic degradation efficiency of MO by ITO capped TiO₂ nanolace arrays with different ITO thickness (5 nm, 15 nm, 30 nm, 60 nm, 90 nm and 120 nm, respectively).

4 Conclusion

ITO capped TiO₂ nanolace arrays have been fabricated by a voltage switched two-step anodic oxidation process to form TiO₂ nanolaces and subsequent ITO magnetron sputtering deposition. By adjusting the anodic voltage and the ITO deposition time, the morphologies of TiO₂ nanolaces and ITO film thickness have been tuned, respectively, to optimize the photocatalytic activity of the photocatalysts. The underlying mechanisms of enhanced photocatalytic activity can be explained by the plasmonic properties of ITO films and the photo-generated exciton separation in ITO–TiO₂ heterojunction. SPR of ITO in near-infrared region generates “hot” electrons, which transfer from TiO₂ to ITO, thus boosting the photocatalytic activity of TiO₂ nanolaces. Additionally, the direct contact of ITO and TiO₂ forms n–n type semiconductor heterojunctions, which facilitates exciton separation and benefit photocatalytic activity. Our results can provide insight of plasmonic photocatalysis of ITO–TiO₂ composite system, and may motivate the invention of plasmonic photocatalysts in near-infrared region with much higher photocatalytic activities.

Acknowledgements

This work was jointly supported by the National Natural Science Foundation of China under Grant No. 51271057 and Jiangsu Key Laboratory of Advanced Metallic Materials (BM2007204).

References

1. Y. Z. He, P. Basnet, S. E. H. Murph and Y. P. Zhao, ACS Appl. Mater. Interfaces, 2013, 5, 11818–11827.
2. Y. Wen, B. T. Liu, W. Zeng and Y. H. Wang, Nanoscale, 2013, 5, 9739–9746.
3. A. Yamakata, T. Ishibashi and H. Onishi, J. Phys. Chem. B, 2002, 106, 9122–9125.
4. Q. Hao, C. X. Wang, H. Huang, W. Li, D. Y. Du, D. Han, T. Qiu and P. K. Chu, Sci. Rep., 2015, 5, 15288.
5. L. J. Bu, W. J. Yang and H. Ming, RSC Adv., 2015, 5, 45122–45128.
6. X. M. Zhang, Y. L. Chen, R. S. Liu and D. P. Tsai, Rep. Prog. Phys., 2013, 76, 046401.
7. E. H. Wang, S. W. Liu, Q. F. Lu, Z. L. Xiu, T. G. Li and L. J. Song, J. Sol-Gel Sci. Technol., 2011, 58, 705–710.
8. D. W. Wang, Y. Li, G. L. Puma, C. Wang, P. F. Wang, W. L. Zhang and Q. Wang, Appl. Catal., B, 2015, 168, 25–32.
9. W. H. Hung, T. M. Chien, A. Y. Lo, C. M. Tseng and D. D. Li, RSC Adv., 2014, 4, 45710–45714.
10. Z. W. Liu, W. B. Hou, P. Pavaskar, M. Aykol and S. B. Cronin, Nano Lett., 2011, 11, 1111–1116.
11. K. Koci, L. Obalova and Z. Lacny, Chem. Pap., 2008, 62, 1–9.
12. J. Z. Y. Tan, Y. Fernandez, D. Liu, M. Maroto-Valer, J. C. Bian and X. W. Zhang, Chem. Phys. Lett., 2012, 531, 149–154.
13. M. L. de Souza, D. C. Tristao and P. Corio, RSC Adv., 2014, 4, 23351–23358.
14. W. B. Hou, Z. W. Liu, P. Pavaskar, W. H. Hung and S. B. Cronin, J. Catal., 2011, 277, 149–153.
15. P. R. West, S. Ishii, G. V. Naik, N. K. Emani, V. M. Shalaev and A. Boltasseva, Laser Photonics Rev., 2010, 4, 795–808.
16. M. Kanehara, H. Koike, T. Yoshinaga and T. Teranishi, J. Am. Chem. Soc., 2009, 131, 17736–17737.
17. G. V. Naik, J. Kim and A. Boltasseva, Opt. Mater. Express, 2011, 1, 1090–1099.
18. C. G. Granqvist and A. Hultaker, Thin Solid Films, 2002, 411, 1–5.
19. S. R. Albu, D. Kim and P. Schmuki, Angew. Chem., Int. Ed., 2008, 47, 1916–1919.
20. K. Yanagisawa and J. Ovenstone, J. Phys. Chem. B, 1999, 103, 7781–7787.
21. K. Lee, D. Kim, P. Roy, I. Paramasivam, B. I. Birajdar, E. Spiecker and P. Schmuki, J. Am. Chem. Soc., 2010, 132, 1478–1479.
22. G. Liu, J. C. Yu, G. Q. Lu and H. M. Cheng, Chem. Commun., 2011, 47, 6763–6783.
23. J. Zhang, L. Z. Liu, L. Yang, Z. X. Gan, X. L. Wu and P. K. Chu, Appl. Phys. Lett., 2014, 104, 231902.
24. H. Chorfi, G. Zayani, M. Saadoun, L. Bousselmi and B. Bessais, Appl. Surf. Sci., 2010, 256, 2170–2175.
25. Y. Ma, J. B. Qiu, Y. A. Cao, Z. S. Guan and J. N. Yao, Chemosphere, 2001, 44, 1087–1092.

---