4316
Langmuir 2008, 24, 4316-4319
Surfactant-Assisted Shape Evolution of Thermally Synthesized TiO2 Nanocrystals and Their Applications to Efficient Photoelectrodes Jin Young Kim,*,† Sung Bum Choi,‡ Dong Wook Kim,‡ Sangwook Lee,‡ Hyun Suk Jung,§ Jung-Kun Lee,| and Kug Sun Hong*,‡ Chemical and Biosciences Center, National Renewable Energy Laboratory, Golden, Colorado 80401, Department of Materials Science and Engineering, Seoul National UniVersity, San 56-1 Sillim-dong, Gwanak-gu, Seoul 151-744, Korea, School of AdVanced Materials Engineering, Kookmin UniVersity, Jeongneung-dong Seongbuk-gu, Seoul 136-702, Korea, and Department of Mechanical Engineering and Materials Science, UniVersity of Pittsburgh, Pittsburgh, PennsylVania 15261 ReceiVed NoVember 8, 2007. In Final Form: December 31, 2007 TiO2 nanocrystals were synthesized via a two-phase thermal process, and the shape of the nanocrystals was controlled from nanospheres to nanorods by the ratio of two surfactants. The shape control of nanocrystals was ascribed to the selective adsorption of the two surfactants. The shape of TiO2 nanocrystals influenced the photocatalytic performances of the photoelectrodes through two compromising factors: the relative surface area and the electron transport. The photoelectrode composed of nanorods showed a slower charge recombination rate, while it showed a smaller specific surface area, compared to nanospheres. As a result, the photoelectrodes showed the optimal photocatalytic performance when the nanospheres and the nanorods were mixed.
1. Introduction TiO2 nanocrystals have been intensively investigated as wide band gap semiconductors due to their potential applications, including solar energy conversion,1 photocatalysis,2 and gas sensors.3 Recently, TiO2 nanocrystals with 1-D structures, such as nanorods, nanowires, and nanotubes, have attracted special interest, since they demonstrate superior properties in comparison with normal nanocrystals.4 Various methods have been developed for the preparation of TiO2 nanocrystals, such as the sol-gel process,5 the hydrothermal process,6 the solvothermal process,7 and the thermal decomposition process.8 Although there have been numerous reports on the synthesis of nanocrystals,5-8 most of these reports have focused only on the controlled synthesis * Corresponding author. Tel: +82-2-880-8316. Fax: +82-2-886-4156. E-mail:
[email protected] (K.S.H.),
[email protected] (J.Y.K.). † National Renewable Energy Laboratory. ‡ Seoul National University. § Kookmin University. | University of Pittsburgh. (1) (a) O’Regan, B.; Gratzel, M. Nature 1991, 353, 737. (b) Jung, H. S.; Lee, J. K.; Nastasi, M.; Lee, S. W.; Kim, J. Y.; Park, J. S.; Hong, K. S.; Shin, H. Langmuir 2005, 21, 10332. (c) Nakade, S.; Matsuda, M.; Kambe, S.; Saito, Y.; Kitamura, T.; Sakata, T.; Wada, Y.; Mori, H.; Yanagida, S. J. Phys. Chem. B 2002, 106, 10004. (2) (a) Hoffmann, M. R.; Martin, S. T.; Choi, W.; Bahnemann, D. W. Chem. ReV. 1995, 95, 69. (b) Gao, L.; Zhang, Q. H. Scr. Mater. 2001, 44, 1195. (c) Chae, S. Y.; Park, M. K.; Lee, S. K.; Kim, T. Y.; Kim, S. K.; Lee, W. I. Chem. Mater. 2003, 15, 3326. (3) (a) Zhu, Y.; Shi, J.; Zhang, Z.; Zhang, C.; Zhang, X. Anal. Chem. 2002, 74, 120. (b) Wu, N.; Wang, S.; Rusakova, I. A. Science 1999, 285, 1375. (4) (a) Mor, G. K.; Shankar, K.; Paulose, M.; Varghese, O. K.; Grimes, C. A. Nano Lett. 2005, 6, 215. (b) Jiu, J.; Isoda, S.; Wang, F.; Adachi, M. J. Phys. Chem. B 2006, 110, 2087. (5) (a) Bischoff, B. L.; Anderson, M. A. Chem. Mater. 1995, 7, 2. (b) Jung, H. S.; Shin, H.; Kim, J. R.; Kim, J. Y.; Hong, K. S.; Lee, J. K. Langmuir 2004, 20, 11732. (6) (a) Yuan, Z.-Y.; Su, B.-L. Colloids Surf., A 2004, 241, 173. (b) Nian, J.-N.; Teng, H. J. Phys. Chem. B 2006, 110, 4193. (7) Kim, C. S.; Moon, B. K.; Park, J. H.; Choi, B. C.; Seo, H. J. J. Cryst. Growth 2003, 257, 309. (8) (a) Joo, J.; Kwon, S. G.; Yu, T.; Cho, M.; Lee, J.; Yoon, J.; Hyeon, T. J. Phys. Chem. B 2005, 109, 15297. (b) Trentler, T. J.; Denler, T. E.; Bertone, J. F.; Agrawal, A.; Colvin, V. L. J. Am. Chem. Soc. 1999, 121, 1613. (c) Tang, J.; Redl, F.; Zhu, Y.; Siegrist, T.; Brus, L. E.; Steigerwald, M. L. Nano Lett. 2005, 5, 543.
of monodispersed nanocrystals and have paid little attention to their applications. In many synthetic procedures of nanocrystals, various surfactants, such as carboxylic acids, alkylamines, and alkylphosphonic acids, have been adopted to control the shape, particularly the aspect ratio, of nanocrystals.9 This method is usually based on the fact that the crystal growth rates are inversely proportion to the surface energies of the crystallographic planes, and therefore, the anisotropic crystal growth can be enhanced by controlling the surface energy via the surfactant adhesion. The anisotropic growth of nanocrystals can be more precisely regulated by adopting different surfactants, which collectively show the selective adhesion to the specific crystal surfaces.9a In this paper, anatase-phase TiO2 nanocrystals with tunable aspect ratios were synthesized thermally in two-phase reaction media (water and toluene) using an autoclave. In order to control the aspect ratio of nanocrystals, we adopted two kinds of surfactants: water-soluble surfactant (ethanolamine) and waterinsoluble surfactant (oleic acid). In particular, we focused on the effects of the relative ratio of the two surfactants on the shape evolution of the TiO2 nanocrystals. Two-phase thermal synthesis has advantages over single-phase methods such as hydrothermal and solvothermal synthesis, since it can effectively separate the surfactants with different functionalities. The effects of the reaction variables can be more evident through this separation. In addition to the synthesis, the photocatalytic performance of TiO2 photoelectrodes, composed of the nanocrystals with controlled aspect ratios, was evaluated. Although there are some reports concerning the photocatalytic properties of TiO2 nanorods,9d few of them have focused on the origin of the enhanced properties. However, in this paper, the photocatalytic properties were analyzed from the viewpoints of electron transport and the relative surface area. (9) (a) Jun, Y.-W.; Casula, M. F.; Sim, J.-H.; Kim, S. Y.; Cheon, J.; Alivisatos, A. P. J. Am. Chem. Soc. 2003, 125, 15981. (b) Zhang, Z.; Zhang, X.; Liu, S.; Li, D.; Han, M. Angew. Chem. Int. Ed. 2005, 44, 3466. (c) Peng, X.; Manna, L.; Yang, W.; Wickham, J.; Scher, E.; Kadavanich, A.; Alivisatos, A. P. Nature 2000, 404, 59. (d) Liao, D. L.; Liao, B. Q. J. Photochem. Photobiol. A: Chem. 2007, 187, 363.
10.1021/la703497e CCC: $40.75 © 2008 American Chemical Society Published on Web 03/07/2008
Photoelectrodes Composed of TiO2 Nanocrystals
Langmuir, Vol. 24, No. 8, 2008 4317
2. Experimental Section Titanium isopropoxide (TIP, 97%, Aldrich), toluene (99.9%, Fisher), oleic acid (OLA, 90%, Aldrich), ethanolamine (EA, 99%, Aldrich), absolute ethanol (99.9%, Hayman), and hexane (95%, Aldrich) were used as received. In a typical preparation of TiO2 nanocrystals with tunable aspect ratios, OLA (10 mL, 30 mmol) and TIP (1.5 mL, 5 mmol) were dissolved in toluene (100 mL) in a Teflon vessel. Then, the TIP in toluene was hydrolyzed with distilled water (100 mL) containing predetermined amounts of EA ranging from 5 to 100 mmol (0.3-6 mL). After hydrolysis, the Teflon vessel was sealed and the solution was reacted at 170 °C for 12 h (heating rate ∼ 1.5 °C/min) with vigorous stirring (450 rpm) before being cooled to room temperature. The resulting solution of TiO2 nanocrystals was precipitated with dry ethanol (150 mL) followed by washing six times with dry ethanol to remove the remaining organics and water. The product of the centrifugation was redispersed in the hexane for the TEM analysis, and the TiO2 powders for the XRD analysis and the fabrication of photoelectrodes were obtained through additional water-washing and freeze-drying. TiO2 photoelectrodes were fabricated by sedimentation. Some pieces of slide glass were bonded on a transparent conducting glass substrate (indium tin oxide, Samsung SDI) in order to make a rectangular guide (3 × 4 cm2). TiO2 nanocrystals were dispersed in ethanol at fixed solid contents (3 mg/mL), and 1 mL of the suspension was put inside the guide followed by drying in an ambient atmosphere while the level was carefully maintained. The crystal structure of the nanocrystals was identified using XRD with Cu KR radiation (M18XHF-SRA, MAC Science) and the morphology of the nanocrystals was investigated using HRTEM (JEM-3000F, JEOL) operating at an accelerating voltage of 300 kV. TEM samples were prepared by placing a drop of the nanocrystal solution in hexane onto carbon-coated copper grids before drying in an ambient atmosphere. The relative surface area of the photoelectrodes was performed through comparison of the amounts of dye-adsorption. The electrodes were immersed in a solution of ruthenium dye [ruthenium (2,2′-bipyridyl-4,4′-dicarboxylate)2(NCS)2 (SOLARONIX), dissolved in ethanol] for 2 h at 50 °C. The amounts of adsorbed dye molecules were measured using a UV-vis spectroscope (Model LAMBDA 650, Perkin-Elmer).1b The open circuit voltage transient profiles and impedance spectra were measured using an electrochemical analyzer (Model CHI 608C, CH Instruments). Photocatalytic decomposition of the aqueous phenol solution (10 ppm) was evaluated using a photoelectrocatalytic system with a Pt counter electrode (0.5 × 4 cm2) and 0.1 M NaCl aqueous electrolyte. The reaction was conducted for 2 h under UV irradiation from a UV lamp (10 W, 253.7 nm, Sankyo). The amounts of decomposed phenol were measured using a UV-vis spectroscope (Model LAMBDA 650, Perkin-Elmer).
3. Results and Discussion The shape of TiO2 nanocrystals, particularly the aspect ratio, was controlled through varying the relative ratio of two surfactants (EA and OLA). The amounts of EA added were varied from 5 mmol (N1 condition) to 100 mmol (N5 condition) with the amounts of TIP (titanium isopropoxide) and OLA fixed at 5 and 30 mmol, respectively. Figure 1 shows the shape evolution of TiO2 nanocrystals with various ratios of EA to OLA. As seen in Figure 1a-c, the aspect ratio of TiO2 nanocrystals increases as the ratio of EA to OLA increases. In the case of the N1 condition, TiO2 nanocrystals with an aspect ratio of approximately 1 are synthesized. Although the aspect ratio of TiO2 nanocrystals increases with the EA concentrations, small amounts of spherical nanocrystals still remain until the N3 condition (Supporting Information). Pure nanorods with an aspect ratio of approximately 10 without any nanospheres can be synthesized only at the N4 condition. In the case of more EA addition, the shape of nanocrystals becomes irregular, as can be seen in Figure 1c (N5
Figure 1. TEM and HRTEM images of TiO2 nanocrystals with ratios of EA/OLA of (a) 1 (N1), (b) 10 (N4), (c) 20 (N5), and (d) 10 (N4, HRTEM image). The growth directions of the nanorods are parallel to the c-axis of the anatase structure. Table 1. Synthetic Conditions and Resulting Shapes of TiO2 Nanocrystals EA TIP toluene OLA H2O sample (mmol) (mL) (mmol) (mL) (mmol) N1 N2 N3 N4 N5
5 5 5 5 5
100 100 100 100 100
30 30 30 30 30
100 100 100 100 100
5 15 25 50 100
shape spheres short rods short and long rods long rods irregular rods
condition). Shape evolution of nanocrystals with various conditions is summarized in Table 1. The formation of nanorods can be ascribed to the selective stabilization of the crystallographic planes of TiO2 crystals by EA and OLA. Since the amine groups are known to form chemical bonds to the {101} planes of TiO2 strongly,8a more {101} planes of TiO2 nanocrystals will be stabilized with increasing amounts of the EA. This selective stabilization of the {101} planes will result in selective growth on the {001} planes, i.e., along the [001] direction. The highresolution image of nanorods in Figure 1d also confirms the growth along the [001] direction. When excessive EA is introduced, the area of {001} planes will overly decrease before the nanorods successfully form. For this reason, the {101} facets are extremely developed in the N5 nanorods, as seen in Figure 1c. Among the TiO2 nanocrystals synthesized under various conditions, we selected N1 and N4 nanocrystals as nanospheres and nanorods, respectively. Figure 2 shows the powder X-ray diffraction (XRD) patterns of TiO2 nanospheres and nanorods, respectively. The XRD patterns show a pure anatase phase (JCPDS #21-1272) without any polymorphs, regardless of the shapes. The anisotropic growth of TiO2 nanocrystals along the [001] direction can be also confirmed in the XRD patterns. As can be seen in Figure 2a, the (004) diffraction peak is strengthened and sharpened for the nanorods. The intensity ratio of the (004) peak relative to the (200) peak represents a meaningful parameter for the anisotropic growth along the [001] direction, since the intensities of two peaks are always independent. The relative intensity ratios of the (004) peaks are 1.25 and 2.24 for
4318 Langmuir, Vol. 24, No. 8, 2008
Figure 2. XRD patterns of TiO2 nanocrystals composed of (a) spheres and (b) rods, prepared at N1 and N4 conditions, respectively.
Figure 3. Decomposition of phenol and the relative surface area of the TiO2 photoelectrodes composed of nanocrystals with various shapes. The relative surface area can be assumed to be proportional to the amounts of dye adsorbed (number of the adsorbed dye molecules per 1 g of TiO2). Samples denoted as “sphere”, “rod”, and “mix” are made of N1, N4, and a 1:1 mixture of N1 and N4 nanocrystals, respectively.
nanospheres and nanorods, respectively. This result can be also ascribed to the preferred anisotropic growth along c-axis of the anatase nanocrystals. In order to compare the photocatalytic performance of the nanocrystals with various aspect ratios, three kinds of TiO2 photoelectrodes composed of N1, N4, and a combined 1:1 mixture of Ni and N4 were prepared. The photoelectrodes were prepared by sedimentation from the ethanol-based suspensions containing nanocrystals so as to prevent the separation of nanospheres and nanorods in the mixture photoelectrodes during coating processes such as dipping and spin-coating. As-deposited photoelectrodes were dried in an ambient atmosphere and their photocatalytic performances were evaluated without further heat treatments. Figure 3 shows the effects of the particle shape on the photocatalytic performances of TiO2 photoelectrodes; 35%, 64%, and 48% of phenols are decomposed after 2-h reactions for nanospheres, mixtures, and nanorods, respectively (indicated with hollow circles in Figure 3). This result can be understood as a compromise between two factors: the relative surface area and the charge recombination. The photocatalytic performance will be enhanced if the relative surface area of the photoelectrodes is increased, since the number of sites for the photocatalytic reaction increases. The relative surface area of the photoelectrodes was compared through the amounts of dye-adsorption, as described in the Experimental Section. As can be seen in Figure 3, the photoelectrode composed
Kim et al.
Figure 4. Open circuit voltage (Voc) transient profiles and ColeCole plots of the TiO2 photoelectrodes composed of nanospheres (black circles) and nanorods (hollow circles).
of the nanorods shows a comparatively lower relative surface area. This result can be ascribed to the packing density of the photoelectrodes, since close-packing is more difficult for nanorods than for nanospheres. For the mixture photoelectrode, the relative surface area is similar to that of the nanospheres, since the nanospheres can fill the open spaces between the nanorods. At this point, it is noticeable that the photocatalytic performances show different behavior to the relative surface area. The photoelectrode composed of nanospheres shows the lowest photocatalytic performance, although it shows the highest relative surface area. On the other hand, the mixture photoelectrode whose relative surface area is similar to the nanospheres shows the best photocatalytic performance. The photocatalytic performance will be also enhanced if the charge recombination is reduced, since the surviving holes can be utilized in the oxidation of the organic compounds. In photoelectrocatalytic systems, photogenerated electrons in the TiO2 layer will move to the conducting layer through diffusion. Assuming that the electron transport inside the particle is faster than the interparticle transport, electrons in the photoelectrodes containing more nanorods will move to the conducting layer (i.e., the counter electrode) more quickly. Therefore, the relatively lower population of electrons compared to the holes in the photoelectrodes will reduce the total amounts of the charge recombination, which results in enhanced photocatalytic performance. This hypothesis can be supported experimentally through the impedance analysis and the open circuit voltage (Voc) transient profile, as seen in Figure 4. As can be seen in the Cole-Cole plots (inset of Figure 4), the photoelectrode composed of nanorods shows lower resistance, i.e., faster electron transport compared to nanospheres.10 The reduced charge recombination can also be confirmed by the Voc transient profiles, where the Voc of the photoelectrode composed of nanorods decays slower than nanospheres.11 As a result, the photoelectrode composed of the mixtures shows the best photocatalytic performance due to the reduced charge recombination and the high relative surface area. The photoelectrode composed of nanospheres shows the lowest photocatalytic performance due to the poor electron transport, in spite of the highest relative surface area. On the other hand, the photoelectrode composed of nanorods shows better photocatalytic performance than nanospheres due to the reduced charge (10) Hoshikawa, T.; Yamada, M.; Kikuchi, R.; Eguchi, K. J. Electrochem. Soc. 2005, 152, E62. (11) (a) Zaban, A.; Greenshtein, M.; Bisquert, J. Chem. Phys. Chem. 2003, 4, 859. (b) Bisquert, J.; Zaban, A.; Greenshtein, M.; Mora-Sero, I. J. Am. Chem. Soc. 2004, 126, 13550.
Photoelectrodes Composed of TiO2 Nanocrystals
Figure 5. Schematic illustrations of the photocatalytic mechanism for the photoelectrodes composed of (a) spheres, (b) mixture, and (c) rods.
recombination, in spite of the lowest relative surface area. Schematic illustrations of the photocatalytic mechanism for each case are presented in Figure 5.
4. Conclusions In conclusion, TiO2 nanocrystals were synthesized via a twophase thermal process, and the aspect ratio of nanocrystals was controlled with the relative ratio of EA to OLA. The anisotropic crystal growth was enhanced with increasing amounts of EA, and pure nanospheres and nanorods were synthesized when 5 mmol (N1 condition) and 50 mmol (N4 condition) of EA were added, respectively. This shape evolution was ascribed to the selective adsorption of surfactants and the resulting selective stabilization of the crystallographic planes. The shape of
Langmuir, Vol. 24, No. 8, 2008 4319
nanocrystals influenced the photocatalytic performances of the TiO2 photoelectrodes in two ways: the relative surface area and the electron transport. Nanospheres displayed higher relative surface area compared to nanorods while showing poorer electron transport. These two compromising effects resulted in optimal photocatalytic performance for the photoelectrode composed of a 1:1 mixture of the nanospheres and the nanorods. From this result, actual applications of nanocrystals in the photocatalysts were demonstrated and their potential applications as photoelectrodes in dye-sensitized solar cells were also confirmed. Acknowledgment. The works done in SNU was partially supported by the Korea Science and Engineering Foundation (KOSEF) grant funded by the Korea government (MOST) (R012007-000-11075-0) and partially supported by the Core Environmental Technology Development Project for Next Generation (Eco-Technopia-21) funded by the Korea Institute of Environmental Science and Technology under the Ministry of Environment, Republic of Korea. The works done in KMU was supported by ERC Program (CMPS, Center for Materials and Processes of Self-Assembly) program of MOST/KOSEF (R11-2005-04800000-0). Supporting Information Available: Additional TEM images. This material is available free of charge via the Internet at http://pubs.acs.org. LA703497E
