Facile Formation of Branched Titanate Nanotubes to Grow a Three

Dec 29, 2009 - Allam , N. K., Shankar , K. and Grimes , C. A. Adv. Mater. 2008 .... Mor , G. K., Varghese , O. K., Paulose , M., Shankar , K. and Grim...
0 downloads 0 Views 3MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Facile Formation of Branched Titanate Nanotubes to Grow a ThreeDimensional Nanotubular Network Directly on a Solid Substrate Haimin Zhang,† Porun Liu,† Hongjuan Wang,†,‡ Hua Yu,† Shanqing Zhang,† Huaiyong Zhu,§ Feng Peng,‡ and Huijun Zhao*,† † Environmental Futures Centre and Griffith School of Environment, Gold Coast Campus, Griffith University, QLD 4222, Australia, ‡School of Chemistry and Chemical Engineering, South China University of Technology, Guangzhou 510640, PR China, and §School of Physical and Chemical Sciences, Queensland University of Technology, Brisbane, QLD 4001, Australia

Received November 3, 2009. Revised Manuscript Received December 20, 2009 The hydrothermal formation of branched titanate nanotubes that grow a 3D nanotubular network directly onto a titanium substrate is reported. The resultant 3D nanotubular network exhibits a unique all-dimensional uniform porous structure. The inner and outer tubular diameters of branched titanate nanotubes were found to be approximately 6 and 12 nm, respectively. For the majority of the nanotubes, the wall is formed from three layers of titanate with an approximate 7.7 A˚ interlayer space. In terms of individual nanotubes, these characteristics are quantitatively similar to those of previously reported nonbranched nanotubes. However, in terms of how nanotubes are arranged in the film, the all-dimensional uniform nanotubular network structure obtained here is distinctively different from those of previously reported structures. The 3D nanotubular network structure was formed by the jointing of branched nanotubes. In contrast, the previously reported nanotubes tend to grow vertically on the substrate, and the resultant tubular films are formed by interwoven nonbranched nanotubes. The branched titanate nanotubes can be readily formed on titanium substrates but not in solution suspension forms. A continuous seed formation-oriented crystal growth mechanism was proposed for the branched titanate nanotubular network formation. Such a network structure could be useful for applications such as photocatalysis, membrane separation, field emission, and photovoltaic devices.

Introduction Since Iijima’s pioneering work in 1991,1 extensive research activities have been conducted to fabricate nanotubes using different materials including TiO2/titanate, CeO2, Al2O3, ZnO, and WO3.2-8 The attraction of such nanotubular materials lies in their unique high-aspect-ratio structural and physiochemical properties.9-12 Among these noncarbon nanotubes, TiO2 nanotubes have occupied an important position because of promising applications in superhydrophilic surfaces, electron field emission, photocatalysis, and solar energy conversion.13-16 Although a *Corresponding author. Fax: 61 7 5552 8067. Tel: 61 7 5552 8261. E-mail: [email protected].

(1) Iijima, S. Nature 1991, 354, 56–58. (2) Zhao, Z.-G.; Miyauchi, M. Angew. Chem., Int. Ed. 2008, 47, 7051–7055. (3) Nah, Y.-C.; Ghicov, A.; Kim, D.; Berger, S.; Schmuki, P. J. Am. Chem. Soc. 2008, 130, 16154–16155. (4) Gonzalez-Rovira, L.; Sanchez-Amaya, J. M.; Lopez-Haro, M.; del Rio, E.; Hungria, A. B.; Midgley, P.; Calvino, J. J.; Bernal, S.; Botana, F. J. Nano Lett. 2009, 9, 1395–1400. (5) Li, Y.; Bando, Y.; Golberg, D. Adv. Mater. 2005, 17, 1401–1405. (6) Allam, N. K.; Shankar, K.; Grimes, C. A. Adv. Mater. 2008, 20, 3942–3946. (7) Armstrong, G.; Armstrong, A. R.; Canales, J.; Bruce, P. G. Chem. Commun. 2005, 2454–2456. (8) Elias, J.; Tena-Zaera, R.; Wang, G.-Y.; Levy-Clement, C. Chem. Mater. 2008, 20, 6633–6637. (9) Ou, H.-H.; Lo, S.-L. Sep. Purif. Technol. 2007, 58, 179–191. (10) Ghicov, A.; Schmuki, P. Chem. Commun. 2009, 2791–2808. (11) Mor, G. K.; Varghese, O. K.; Paulose, M.; Shankar, K.; Grimes, C. A. Sol. Energy Mater. Sol. Cells 2006, 90, 2011–2075. (12) Menzel, R.; Peiro, A. M.; Durrant, J. R.; Shaffer, M. S. P. Chem. Mater. 2006, 18, 6059–6068. (13) Miyauchi, M.; Tokudome, H.; Toda, Y.; Kamiya, T.; Hosono, H. Appl. Phys. Lett. 2006, 89, 043114/1–043114/3. (14) Miyauchi, M.; Tokudome, H. J. Mater. Chem. 2007, 17, 2095–2100. (15) Kim, D.; Ghicov, A.; Albu, S. P.; Schmuki, P. J. Am. Chem. Soc. 2008, 130, 16454–16455. (16) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano Lett. 2005, 5, 191–195.

1574 DOI: 10.1021/la9041869

variety of TiO2 nanotube fabrication methods have been reported, anodization,17 sol-gel template-assisted synthesis,18 and hydrothermal reaction19 methods are the most commonly used methods to obtain a large quantity of nanotubes. Particular attention has been given to low-temperature hydrothermal synthesis because of its simplicity and mass production capacity.7,9,20,21 To date, the vast majority of reported hydrothermal methods produces nanotubes in solution suspension forms20,22,23 whereas only a few directly grows nanotubes on solid substrates.13,14,21,24,25 It is well known that immobilizing nanotubes onto solid substrates is essential for many applications.13-16 In this regard, direct hydrothermal growth of nanotubes onto a solid substrate eliminates the need for postsynthesis immobilization.13,14,21,24,25 More importantly, the ability to create desirable seeding conditions by manipulating the substrate surface enables the direct hydrothermal growth of nanotubes on substrates with various structural alignments that cannot be obtained when nanotubes are produced in solution suspension forms.20,22,23 Tian et al. reported a hydrothermal method to grow a vertically aligned titanate nanotubular (17) Gong, D.; Grimes, C. A.; Varghese, O. K.; Hu, W.; Singh, R. S.; Chen, Z.; Dickey, E. C. J. Mater. Res. 2001, 16, 3331–3334. (18) Lakshmi, B. B.; Patrissi, C. J.; Martin, C. R. Chem. Mater. 1997, 9, 2544– 2550. (19) Kasuga, T.; Hiramatsu, M.; Hoson, A.; Sekino, T.; Niihara, K. Langmuir 1998, 14, 3160–3163. (20) Chen, Q.; Zhou, W.; Du, G.; Peng, L.-M. Adv. Mater. 2002, 14, 1208–1211. (21) Tian, Z. R.; Voigt, J. A.; Liu, J.; McKenzie, B.; Xu, H. J. Am. Chem. Soc. 2003, 125, 12384–12385. (22) Bavykin, D. V.; Parmon, V. N.; Lapkin, A. A.; Walsh, F. C. J. Mater. Chem. 2004, 14, 3370–3377. (23) Kukovecz, A.; Hodos, M.; Horvath, E.; Radnoczi, G.; Konya, Z.; Kiricsi, I. J. Phys. Chem. B 2005, 109, 17781–17783. (24) Peng, X.; Chen, A. Adv. Funct. Mater. 2006, 16, 1355–1362. (25) Yada, M.; Inoue, Y.; Uota, M.; Torikai, T.; Watari, T.; Noda, I.; Hotokebuchi, T. Langmuir 2007, 23, 2815–2823.

Published on Web 12/29/2009

Langmuir 2010, 26(3), 1574–1578

Zhang et al.

film directly on a titanium substrate using Degussa P25 TiO2 power as the starting material.21 Miyauchi et al. later reported a hydrothermal method to grow TiO2 nanotubular arrays directly on titanium substrates (or sapphire substrates) without the use of TiO2 starting material.13,14 The resultant nanotubular array structure exhibits excellent electron field emission and superhydrophilic properties. Yada and co-workers recently utilized titanium metal as the titanium source and the morphological directing material to grow titanate nanotubes directly on plates, wires, meshes, spheres, and tubes.25 Despite these successes, the potential of the direct hydrothermal growth of nanostructures on a solid substrate is far from being fully explored. All reported nanotube formations under hydrothermal conditions are exclusively 1D growth, resulting in nonbranched nanotubes.12,20,21,23 Nanotubular films formed by such nonbranched nanotubes are either in a vertically aligned nanotube form or an interwoven nanotube arrangement.13,14,21,24,25 These nonbranched nanotubes are incapable of forming nanotubular films with 3D network structures, which are highly attractive for membrane separation, field emission, photocatalysis, and solar cell applications. Here we report for the first time the formation of branched titanate nanotubes under hydrothermal conditions to grow a 3D nanotubular network directly that exhibits a unique all-dimensional uniform porous structure. Although a variety of titanate nanotube formation mechanisms under hydrothermal conditions have been proposed, the precise mechanism is still somewhat controversial.23,26-28 The sheet rollup mechanism is a widely accepted formation mechanism, but shortcomings have been identified.27,28 Kukovecz and co-workers recently proposed an oriented crystal growth mechanism from nanoloop seeds.23 They suggested that their new mechanism can explain the experimental findings better than the sheet rollup mechanism. With this oriented crystal growth mechanism, nanotube formation is initiated by the formation of the nanoloop form of titanate seeds resulting from the dissolution of starting material (titanium source) under hydrothermal conditions. A subsequent oriented crystal (basic building blocks resulting from the hydrothermal dissolution of starting material) directly built onto the nanoloop seeds is responsible for nanotube growth. Accordingly, the starting material is an important factor because it strongly affects the formation of building blocks (oriented crystal) and nanoloop seeds and hence nanotube growth. For this reason, we employed a TiO2-porous-film-coated titanium foil as the starting material. To investigate the potentials, the resultant 3D titanate nanotubular network on a titanium substrate as an electrode was evaluated in dye-sensitized solar cells (DSSCs).

Letter

Fabrication of a Three-Dimensional Titanate Nanotubular Network. Titanium foils were degreased prior to experiment

by sonication in acetone, 2-propanol, and methanol, subsequently rinsed with Milli-Q water, and finally dried in a stream of nitrogen. The pretreatment Ti foils were used as substrates for subsequent TiO2 porous films by the sol-gel technique.29 Titanium butoxide (17.02 mL) and diethanolamine (4.8 mL) were dissolved in ethanol (67.28 mL). After the mixture was vigorously stirred for 2 h at room temperature, water (0.9 mL) and ethanol (10.0 mL) were mixed and slowly added to the above solution. The molar ratio of water to alkoxide was controlled at 1:1. After the solution was vigorously stirred for 1 h, 1.0 g of PEG [HOCH2(CH2OCH2)nCH2OH] was added to the above solution. PEG served as a porous creating reagent. The TiO2 coatings were achieved by dip coating onto the pretreated Ti foils. The withdrawal speed was 2 mm/s. The resultant coatings were dried at room temperature for 2 h and then treated at 550 °C in air for 2 h at a heating and cooling rate of 2 °C/min. A three-dimensional titanate nanotubular network film was fabricated by hydrothermal reaction. The above prepared TiO2 porous film on a Ti substrate was placed in an autoclave. The NaOH solution (50 mL, 10 M) was added to the above autoclave. The hydrothermal reaction was kept at 130 °C for 6 h. The resultant 3D titanate nanotubular network film was thoroughly rinsed with deionized water and then treated with 0.1 M HNO3 solution for 30 min. It was again thoroughly rinsed with deionized water and then dried at room temperature for subsequent characterization and measurement. Characterization. The morphologies of the resultant 3D titanate nanotubular network films were investigated using a JEOL JSM-6400F field emission scanning electron microscope (FESEM, Tokyo, Japan). The microstructure and morphology of nanotube samples were also examined using transmission electron microscopy (TEM, FEI Tecnai 20) with an acceleration voltage of 200 kV. It should be noted that to preserve the original morphology of the titanate nanotubular network structure the TEM sample was prepared by directly scratching off samples from the Ti substrate without ultrasonic dispersion. XRD patterns of the TiO2 porous film and titanate nanotube samples were recorded by X-ray diffraction (XRD, LabX-6000, Shimadzu, Japan) employing Cu KR radiation at 40 kV and 40 mA over the 2θ range of 5-75°. Measurements. Photoelectrochemical measurements were performed at 23 °C in a photoelectrochemical cell with a quartz window for illumination.30 It consisted of a TiO2 photoanode, a saturated Ag/AgCl reference electrode, and a platinum mesh auxiliary electrode. A voltammograph (CV-27, BAS) was used for the application of potential bias. Potential and current signals were recorded using a Macintosh (AD Instruments). The illuminated area of the photoanode was 0.785 cm2. A 0.10 M NaNO3 solution was used as the electrolyte. Illumination was carried out using a 150 W xenon arc lamp light source with a focusing lenses (HF-200W-95, Beijing Optical Instruments). To avoid the electrolyte being heated up by the infrared light, a UV band-pass filter (UG 5, Avotronics Pty. Ltd.) was used. The light intensity was regulated and carefully measured at 365 nm. The fabricated 3D titanate nanotube network film was used as photoanode for measurement in dye-sensitized solar cells (DSSCs).31 The nanotube photoanode was first heated to 100 °C for 0.5 h to eliminate adsorbed water and then dipped into the dye solution (5  10-4 M) and kept for 24 h until complete dye absorption was achieved.32 The dye solution was prepared by dissolving N-719 dye (RuL2(NCS)2(TBA)2(H2O)4, L = 2,20 bipyridyl-4,40 -dicarboxylic acid, TBA = tetrabutylammonium, from Dyesol) in butanol and acetonitrile (1:1 v/v). A pyrolytic

(26) Kasuga, T.; Hiramatsu, M.; Hoson, A.; Sekino, T.; Niihara, K. Adv. Mater. 1999, 11, 1307–1311. (27) Yao, B. D.; Chan, Y. F.; Zhang, X. Y.; Zhang, W. F.; Yang, Z. Y.; Wang, N. Appl. Phys. Lett. 2003, 82, 281–283. (28) Wang, Y. Q.; Hu, G. Q.; Duan, X. F.; Sun, H. L.; Xue, Q. K. Chem. Phys. Lett. 2002, 365, 427–431.

(29) Yu, J.; Zhao, X.; Du, J.; Chen, W. J. Sol-Gel Sci. Technol. 2000, 17, 163– 171. (30) Jiang, D.; Zhao, H.; Zhang, S.; John, R. J. Catal. 2004, 223, 212–220. (31) Gratzel, M. Nature 2001, 414, 338–344. (32) Ito, S.; Ha, N.-L. C.; Rothenberger, G.; Liska, P.; Comte, P.; Zakeeruddin, S. M.; Pechy, P.; Nazeeruddin, M. K.; Graetzel, M. Chem. Commun. 2006, 4004– 4006.

Experimental Section Chemicals and Materials. Titanium (Ti) foils (0.25 mm thick, 99.7% purity) were supplied by Aldrich. Titanium butoxide (97%, Aldrich), diethanolamine, ethanol, acetone, 2-propanol, methanol, and sodium hydroxide (analytical grade) were purchased from Aldrich without further treatment prior to use. Highpurity deionized water (Millipore Corp., 18 MΩ cm) was used for solution preparation.

Langmuir 2010, 26(3), 1574–1578

DOI: 10.1021/la9041869

1575

Letter

Zhang et al.

transparent Pt electrode was deposited as described33 and used as the counter electrode for DSSCs. The commercially available EL141 (Dyesol) was used as the electrolyte for DSSC assembly. The DSSCs were illuminated from the Pt counter electrode side with a 500 W Xe lamp (Beijing Changtuo) with an AM 1.5 G filter (Sciencetech Inc. Canada). The active area of DSSCs is 0.15 cm2. The light intensity was measured by a radiant power meter (Newport, 70260) coupled to a broadband probe (Newport, 70268). The measured light intensity is 84 mW/cm2. The photocurrent-voltage curves (J-V) of DSSCs were recorded with a PAR potentiostat (362). For comparison, the TiO2 porous film was also measured in DSSCs under the same experimental conditions. The solar energy-to-electricity conversion efficiency (η) is defined in eq 1,34 η ¼

Jsc Voc FF  100% Pin

ð1Þ

where Jsc is the short-circuit photocurrent density under irradiation, Voc is the open-circuit voltage, Pin is the light power per unit area, and FF is the fill factor calculated in eq 2 FF ¼

Pmax Imax Vmax ¼ Jsc Voc Jsc Voc

ð2Þ

where Imax and Vmax are the current and potential at the maximum power point, respectively. The J-V characterization was conducted under Pin = 84 mW 3 cm-2 for the η calculation.

Results and Discussion Figure 1a shows a typical FESEM image of the starting TiO2 porous film on a titanium substrate. The TiO2 porous film was prepared via a sol-gel method using a precursor solution containing 1.1 wt % PEG.29 A highly porous TiO2 film was obtained by completely decomposing the PEG component in the film during a calcination process at 550 °C.29 The resultant porous coating layer with a thickness of 1.2 μm (inset in Figure 1a) and pore size ranging from 80 to 100 nm were found to be dominated by the anatase phase (98.2%, Figure 2, curve a). The starting TiO2 porous film was hydrothermally treated at 130 °C in 50 mL of 10 M NaOH solution. Figure 1b shows the FESEM image of the sample after 30 min of reaction. It revealed a very rough surface, resulted from the dissolution of the TiO2 porous film. A detailed investigation suggested that for the given conditions the hydrothermal reaction was dominated by the dissolution of the TiO2 porous film at up to 70 min. Nanoloop forms of titanate seeds start to appear on the substrate surface after 70 min of hydrothermal treatment, indicating the beginning of the seeding process. When the reaction proceeded for 90 min, a large number of sproutlike structures appeared on the substrate surface (Figure 1c), signifying the start of network formation. TEM image revealed that the sproutlike structures consisted of infant nanotubes (top inset in Figure 1c). These nanotubes in their infancy exhibit irregular shapes and different layers of wall structures. It is important to note the coexistence of nanoloop seeds on the substrate surface at this stage of the reaction (bottom inset in Figure 1c).23 This observation suggests that the seeding process is continuous. The nanotubular network took form after 180 min of reaction (Figure 1d), resulting from the oriented crystal growth. The nanoloop seeds could still be found at this stage of the reaction but seem to be rather three-dimensional (inset in Figure 1d). Such an oriented crystal growth process (33) Papageorgiou, N.; Maier, W. F.; Gratzel, M. J. Electrochem. Soc. 1997, 144, 876–884. (34) Peng, B.; Jungmann, G.; Jager, C.; Haarer, D.; Schmidt, H.-W.; Thelakkat, M. Coord. Chem. Rev. 2004, 248, 1479–1489.

1576 DOI: 10.1021/la9041869

Figure 1. FESEM and TEM images of the samples with different hydrothermal reaction times: (a) 0, (b) 30, and (c) 90 min, where the insets are of TEM images of infant nanotubes and nanoloop seeds; (d) 180 min, where the inset is a TEM image of 3D nanoloop seeds; (e) 360 min, where the inset is a high-magnification FESEM image; (f) cross-sectional FESEM image of a 360 min sample; and (g) HRTEM image of an individual nanotube for a 360 min sample.

Figure 2. (a) X-ray diffraction patterns of the TiO2 porous film by sol-gel technique annealing at 550 °C for 2 h. (b) Three-dimensional titanate nanotube network film formed by treating the TiO2 porous film with 10 M NaOH at 130 °C for 6 h.

dominated the reaction until the matured 3D nanotubular network was obtained at 360 min (Figure 1e). A cross-sectional FESEM image of the matured 3D nanotubular network is shown in Figure 1f. Interestingly, almost identical surface and cross-sectional morphologies were obtained. A close investigation Langmuir 2010, 26(3), 1574–1578

Zhang et al.

revealed that the entire film (4.5 μm in thickness) was formed by a unique, all-dimensional uniform nanotubular network structure. The porous distribution of the network structure was found to be the same in all directions in the nanotube film. The HRTEM characteristics of individual nanotube were examined (Figure 1g). The inner and outer tubular diameters were found to be approximately 6 and 12 nm, respectively. For a majority of the nanotubes, the wall was formed by three layers of titanate with an approximate 7.7 A˚ interlayer space. The selected-area electron diffraction (SAED) patterns (not shown) display a discontinued ring pattern, suggesting the existence of titanate polycrystallites. XRD patterns of the porous TiO2 film starting material (curve a in Figure 2) confirmed the coexistence of anatase and rutile, signified by the diffraction peaks at 25.3 and 27.3°, respectively. The phase composition was calculated to be 98.2% for anatase and 1.8% for rutile. The XRD patterns of the resultant 3D nanotubular network film revealed typical titanate diffraction peaks (curve b in Figure 2).35,36 The obtained nanotubular structural signature reflection peak at 2θ ≈ 10° corresponding to the interlayer spacing of titanate nanotubes was found to be identical to that previously reported.35,36 In terms of individual nanotubes, these characteristics are quantitatively similar to those that were previously reported.35,36 However, in terms of how nanotubes are arranged in the film, the all-dimensional uniform nanotubular network structure obtained here is distinctively different from those of previously reported structures.13,14,21 Figure 1e,f suggests that the 3D nanotubular network structure may be formed by the jointing of branched nanotubes. In contrast, the previously reported nanotubes tend to grow vertically on the substrate, and resultant tubular films are formed by interwoven nonbranched nanotubes.13,14,21 Further investigations were carried out to confirm the nanotube arrangement within the 3D nanotubular network structure. Figure 3a shows a TEM image of the resultant 3D nanotubular network structure. It should be noted that all TEM samples used in this work were prepared by directly scratching off the materials from the resultant nanotubular film without ultrasonic dispersion to preserve the original network structure. As can be seen in Figure 3a, the 3D nanotubular network was constructed by joining nanotubes together at joint centers (circled area in Figure 3a). Distances between the two directly linked joint centers varied from 80 to 200 nm. For the majority of the nanotubes, their lengths were limited by the distance of the neighboring joint centers. Typical joint-center structure shown in Figure 3a indicates that the joint center is formed by the branched nanotubes. An investigation was therefore carried out to confirm the branched nanotubes. Figure 3b shows an HRTEM image in which branched nanotubes are clearly visible. These observations differed remarkably from those of previously reported nanotube arrangements in nanotubular films.13,14,21 According to existing mechanisms, the formation of nanotubular joint center is unlikely to occur because individual nanotubes tend to grow along the axis of the tube and may be interwoven when they reach a certain length but would not be chemically joined together to form branches.21 The formation of branched nanotubes cannot be explained by the sheet rollup mechanism.28 It may be explained by the oriented crystal growth mechanism if seed formation could continuously occur throughout the growth process.23 Figure 4 illustrates the 3D nanotube network formation mechanism. The anatase TiO2 porous film provides a suitable titanium source. The initial hydrothermal process is dominated by the dissolution of porous TiO2, providing (35) Yuan, Z.-Y.; Su, B.-L. Colloids Surf., A 2004, 241, 173–183. (36) Suzuki, Y.; Yoshikawa, S. J. Mater. Res. 2004, 19, 982–985.

Langmuir 2010, 26(3), 1574–1578

Letter

Figure 3. (a) TEM image of titanate nanotube network structure, with green circles indicating the joint-center structures. (b) HRTEM image of branched titanate nanotubes.

Figure 4. Schematic diagram illustrating the 3D titanate nanotube network film formation mechanism.

the required building block (oriented crystal). Nanoloop seed formation occurs when a dissolution/deposition dynamic equilibrium is reached. The oriented crystal growth and continuous nanoloop seed formation on the tips of the nanotube bundles are responsible for the subsequent branched nanotube growth and joint-center formation that lead to all-dimensional uniform porous network structures. The resultant 3D titanate nanotube network film and the TiO2 porous film on a titanium substrate were used as photoanodes for photocatalytic activity evaluation using water as the probe molecule. Water was selected to be the test compound because of its presence in all aqueous reaction media. All experiments were carried out in an aqueous solution containing a 0.10 M NaNO3 electrolyte under different UV light intensities. Figure S1a (Supporting Information) shows the voltammograms of the 3D titanate nanotube network photoanode with or without UV illumination. For all cases with UV illumination, the photocurrents increased linearly as the applied potential bias increased within the low potential range, attributed to the limitation of free photoelectron transport within the photocatalyst film.37,38 The photocurrents saturated at higher potentials because of either the limitation of photohole generation or capture processes at the catalyst/solution interface.37,38 The results revealed that both the slope and saturation photocurrent (Jsph) values increased as the (37) Zhao, H.; Jiang, D.; Zhang, S.; Catterall, K.; John, R. Anal. Chem. 2004, 76, 155–160. (38) Jiang, D.; Zhao, H.; Zhang, S.; John, R. J. Phys. Chem. B 2003, 107, 12774– 12780.

DOI: 10.1021/la9041869

1577

Letter

UV light intensity increased, suggesting that the photoholegeneration process was the control step of the overall reaction. In contrast, only a negligible current (dark current) was measured when the illumination was switched off, suggesting that the measured currents under UV illumination are pure photocurrents originating from the photoelectrocatalytic oxidation of water. The voltammograms obtained from the TiO2 porous film photoanode (Figure S1b in Supporting Information) were found to be qualitatively similar to those shown in Figure S1a. It was found that for a given light intensity the Jsph value obtained from the TiO2 porous film photoanode was lower than that of the 3D titanate nanotube network photoanode. The magnitude of the saturated photocurrents (Jsph) can be used quantitatively to reflect the photocatalytic activity of a photocatalyst because it represents the maximum rate of oxidation under a given light intensity.30,39 To evaluate the relative photocatalytic activity of the TiO2 porous film and 3D titanate nanotube network photoanodes quantitatively, the Jsph values obtained from the two photoanodes were plotted against the light intensity (Figure S2 in Supporting Information). Linear relationships were obtained for both cases with slope values of 0.0915 (R2 = 0.999) and 0.0651 (R2 = 0.998) mA/mW for the 3D titanate nanotube network and the TiO2 porous film photoanodes, respectively. For a given set of experimental conditions, the slope of the Jsph-j curve quantitatively represents the photoelectrocatalytic activity of the photoanode. The data revealed that the slope value obtained from the 3D titanate nanotube network photoanode was 40% higher than that of the TiO2 porous film photoanode. This demonstrates the significantly improved photocatalytic activity of the 3D titanate nanotube network photoanode. The enhanced photocatalytic activity could be attributed to the uniquely configured 3D titanate nanotube network structure. The resultant 3D titanate nanotube network film on the titanium substrate was also evaluated in dye-sensitized solar cells (DSSCs). Figure 5 compares the photocurrent-voltage (Jph-V) curves of DSSCs using the 3D titanate nanotube network photoanode (curve a) and the TiO2 porous film photoanode (curve b). Under AM 1.5 illumination, the 3D nanotube network photoanode cell exhibits a short circuit current of ∼5.58 mA cm-2, an open circuit voltage of ∼0.74 V, and a fill factor of 0.60. The overall solar energy-to-electricity conversion efficiency (η) was found to be 3.0% whereas for the DSSCs constructed using a TiO2 porous film photoanode the measured η was only 0.33% with a short circuit current of ∼1.01 mA cm-2, an open circuit voltage of ∼0.65 V, and a fill factor of 0.42. These results suggest that in terms of solar energy-to-electricity conversion efficiency the 3D titanate nanotube network photoanode cell is 9 times higher than that of the TiO2 porous film photoanode cell. The enhancement in conversion efficiency may be attributed to the textural and structural properties of 3D nanotubular network architecture. This unique nanotube network structure possesses a large surface area, increasing the number of chemisorbed dye (39) Zhang, S.; Jiang, D.; Zhao, H. Environ. Sci. Technol. 2006, 40, 2363–2368.

1578 DOI: 10.1021/la9041869

Zhang et al.

Figure 5. (a) Photocurrent-voltage curves of the 3D titanate nanotube network photoanode DSSC. (b) TiO2 porous film photoanode DSSC. (c) Photocurrent-voltage curve of the 3D titanate nanotube network-based DSSCs in the dark state.

molecules. The porous tubular texture of the network is favorable for the mass transport of redox electrolyte (I2/I3-) into and out of the photocatalyst layer. The enhancement in conversion efficiency may also be attributed to the improved electron transport properties within the catalyst layer as a result of the favorable vectorial electron transport along the axis of the nanotubes.40-42 The lower light-to-electricity conversion efficiency of the 3D titanate nanotube network cell in comparison to the reported result43 was mainly due to the thickness of the titanate nanotube network film, 4.5 μm in our case. In theory, higher efficiencies could be achieved by growing thicker nanotube network films. However, growing thicker nanotube network films that possess strong adhesion to the substrate remains a challenge under investigation in our experiments. In summary, we have demonstrated a hydrothermal method to grow a 3D nanotube network structure directly on a titanium substrate. Most noticeably, the resultant structure is formed by branched nanotubes and possesses unique all-dimensional uniform porous structural characteristics. Such a 3D nanotube network structure on a solid substrate could be useful for a wide range applications in photocatalysis, field emission, membrane separation, and photovoltaic devices. Acknowledgment. This work was financially supported by the Australian Research Council. Supporting Information Available: Voltammograms of the 3D titanate nanotube network photoanode and the TiO2 porous film photoanode obtained from an aqueous solution containing 0.10 M NaNO3 under different light intensities. Relationships between the saturation photocurrent and light intensity. This material is available free of charge via the Internet at http://pubs.acs.org. (40) Zhang, H.; Zhao, H.; Zhang, S.; Quan, X. ChemPhysChem 2008, 9, 117– 123. (41) Zhu, K.; Neale, N. R.; Miedaner, A.; Frank, A. J. Nano Lett. 2007, 7, 69–74. (42) Paulose, M.; Shankar, K.; Varghese, O. K.; Mor, G. K.; Grimes, C. A. J. Phys. D 2006, 39, 2498–2503. (43) Gratzel, M. J. Photochem. Photobiol., A 2004, 164, 3–14.

Langmuir 2010, 26(3), 1574–1578