Shape and Size Controlled Synthesis of Anatase Nanocrystals with

Dec 23, 2009 - Study on the Anatase to Rutile Phase Transformation and Controlled Synthesis of Rutile Nanocrystals with the Assistance of Ionic Liquid...
0 downloads 22 Views 4MB Size
Download PDF
pubs.acs.org/Langmuir © 2009 American Chemical Society

Shape and Size Controlled Synthesis of Anatase Nanocrystals with the Assistance of Ionic Liquid Kunlun Ding, Zhenjiang Miao, Baoji Hu, Guimin An, Zhenyu Sun, Buxing Han, and Zhimin Liu* Beijing National Laboratory for Molecular Science (BNLMS), Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China Received September 23, 2009. Revised Manuscript Received November 17, 2009 We report an ionic liquid (IL) assisted hydrothermal method to synthesize anatase TiO2 nanocrystals (NCs), in which TiCl4 was used as precursor, 1-butyl-3-methylimidazolium chloride (bmimþCl-) as IL, and F- or SO42- ions as phase transformation inhibitor. The surfactant-like nature of IL was found to play a key role in controlling the crystallization process via controlling the aggregation manner of the NCs. The fine-tuning abilities of the operating parameters of the bmimþCl-/TiCl4/H2O system facilitated the controlling over the shape and size of TiO2 NCs. Phase-pure anatase monodisperse NCs with various shape and size were controllably obtained. Moreover, the aggregation manners of anatase NCs were also studied, and it was demonstrated that the high concentration of HF or H2SO4 could result in aggregation of anatase NCs to form pseudo single crystals.

1. Introduction Titanium dioxide (TiO2) is one of the most studied transition metal oxides and has attracted great attention due to its chemical stability, nontoxicity, low cost, and many other advantageous properties. TiO2 has been widely used as pigment, UV absorbant in many cosmetics, photovoltaics, photocatalysts, catalyst supports, gas sensors, Li batteries, and so on.1 TiO2 has four major polymorphs: anatase (tetragonal), rutile (tetragonal), brookite (orthorhombic), and TiO2 (B) (monoclinic).1a,1b Anatase and rutile are the most important crystal polymorphs of TiO2 owing to their relatively higher stability and easier synthesis. The lattice structure differences among these phases lead to many different physical and physical chemical properties, thus leading to different performances in applications. In addition to the phase effects, the shape and size of TiO2 nanoparticles also influence the properties and performances of the TiO2-based materials. The shape indicates the percentages of different facets. As a result, different shaped nanoparticles exhibit different surface properties, finally leading to different performances.2 For example, most anatase crystals (more than 94%, according to the Wulff construction) are dominated by the thermodynamically stable {101} facets, rather than the much more reactive {001} facets.3,4 Only very recently, some methods *Corresponding author: Fax 8610-62562821; Tel 8610-62562821; e-mail [email protected]. (1) (a) Diebold, U. Surf. Sci. Rep. 2003, 48, 53. (b) Carp, O.; Huisman, C.; Reller, A. Prog. Solid State Chem. 2004, 32, 33. (c) Chen, X. B.; Mao, S. S. Chem. Rev. 2007, 107, 2891. (d) Hagfeldt, A.; Gr€atzel, M. Chem. Rev. 1995, 95, 49. (e) Fujishima, A.; Honda, K. Nature 1972, 238, 37. (2) (a) Somorjai, G. A.; Rioux, R. M. Catal. Today 2005, 100, 201. (b) Yin, Y. D.; Alivisatos, A. P. Nature 2005, 437, 664. (c) Jun, Y.; Choi, J.; Cheon, J. Angew. Chem., Int. Ed. 2006, 45, 3414. (d) Tao, A. R.; Habas, S.; Yang, P. D. Small 2008, 4, 310. (e) Xia, Y. N.; Xiong, Y. J.; Lim, B.; Skrabalak, S. E. Angew. Chem., Int. Ed. 2009, 48, 60. (3) (a) Barnard, A. S.; Curtiss, L. A. Nano Lett. 2005, 5, 1261. (b) Barnard, A. S.; Zapol, P.; Curtiss, L. A. Surf. Sci. 2005, 582, 2173. (4) (a) Donnay, J. D.; Harker, D. Am. Mineral. 1937, 22, 446. (b) Lazzeri, M.; Vittadini, A.; Selloni, A. Phys. Rev. B 2001, 63, 155409. (5) Ding, K. L.; Miao, Z. J.; Liu, Z. M.; Zhang, Z. F.; Han, B. X.; An, G. M.; Miao, S. D.; Xie, Y. J. Am. Chem. Soc. 2007, 129, 6362. (6) (a) Yang, H. G.; Sun, C. H.; Qiao, S. G.; Zou, J.; Liu, G.; Smith, S. C.; Cheng, H. M.; Lu, G. Q. Nature 2008, 453, 638. (b) Yang, H. G.; Liu, G.; Qiao, S. G.; Sun, C. H.; Jin, Y. G.; Smith, S. C.; Zou, J.; Cheng, H. M.; Lu, G. Q. J. Am. Chem. Soc. 2009, 131, 4078.

Langmuir 2010, 26(7), 5129–5134

have been developed to synthesize {001} facets dominated anatase micro/nanocrystals by several groups (including our group).5-7 Unlike the II-VI semicondutors, TiO2 has a relatively small exciton radius (∼1 nm); thus, its quantum confinement effect has been seldom reported.8 However, crystal size is still a key factor determining the application of the TiO2 nanomaterials. For the photoinduced processes including photovoltaics and photocatalysis, the decrease in crystal size causes the increase of e-/hþ recombination probability. Therefore, proper size is needed to balance the specific surface area and e-/hþ recombination probability, so as to obtain the optimized performances.9,10 Massive work has been done on the phase, shape, and size effects on the applications of TiO2. But unfortunately, a survey of the literature reveals widespread disagreement.1b,1c,11 Many contrasting results have been reported. One of the main obstacles is the phase-, shape-, and size-controlled synthesis of TiO2 nanocrystals (NCs). According to the theoretical calculation, anatase possesses lower surface Gibbs free energy and higher bulk one than rutile.3,12 Therefore, the thermodynamic stability of TiO2 NCs is size dependent due to the balance of bulk and surface contribution. Anatase nanoparticles were reported to be thermodynamically more stable than equal-sized rutile in the size range smaller than a critical size, which is highly dependent on the (7) Han, X. G.; Kuang, Q.; Jin, M. S.; Xie, Z. X.; Zheng, L. S. J. Am. Chem. Soc. 2009, 131, 3152. (8) (a) Zhai, H. J.; Wang, L. S. J. Am. Chem. Soc. 2007, 129, 3022. (b) Satoh, N.; Nakashima, T.; Kamikura, K.; Yamamoto, K. Nat. Nanotechnol. 2008, 3, 106. (9) Cozzoli, P. D.; Kornowski, A.; Weller, H. J. Am. Chem. Soc. 2003, 125, 14539. (10) (a) Grela, M.; Colussi, A. J. J. Phys. Chem. 1996, 100, 18214. (b) Zhang, Z.; Wang, C. C.; Zakaria, R.; Ying, J. Y. J. Phys. Chem. B 1998, 102, 10871. (c) Kominami, H.; Muratami, S.; Kato, J.; Kera, Y.; Ohtani, B. J. Phys. Chem. B 2002, 106, 10501. (d) Almquist, C. B.; Biswas, P. J. Catal. 2002, 212, 145. (11) Testino, A.; Bellobono, I. R.; Buscaglia, V.; Canevali, C.; D’Arienzo, M.; Polizzi, S.; Scotti, R.; Morazzoni, F. J. Am. Chem. Soc. 2007, 129, 3564. (12) (a) Gribb, A. A.; Banfield, J. F. Am. Mineral. 1997, 82, 717. (b) Zhang, H. Z.; Banfield, J. F. J. Mater. Chem. 1998, 8, 2073. (c) Zhang, H. Z.; Banfield, J. F. Am. Mineral. 1999, 84, 528. (d) Zhang, H. Z.; Banfield, J. F. J. Mater. Res. 2000, 15, 437. (e) Zhang, H. Z.; Banfield, J. F. J. Phys. Chem. B 2000, 104, 3481. (f ) Gilbert, B.; Zhang, H. Z.; Huang, F.; Finnegan, M. P.; Waychunas, G. A.; Banfield, J. F. Geochem. Trans. 2003, 4, 20. (g) Finnegan, M. P.; Zhang, H. Z.; Banfield, J. F. J. Phys. Chem. B 2007, 111, 1962. (h) Finnegan, M. P.; Zhang, H. Z.; Banfield, J. F. Chem. Mater. 2008, 20, 3443.

Published on Web 12/23/2009

DOI: 10.1021/la903600q

5129

Article

surface complexation conditions.3,12 From this point of view, it is reasonable that TiO2 prefers to homogeneously nucleate into anatase phase, and rutile phase can only heterogeneously nucleate via the aggregation of anatase NCs.13 Various methods have been developed to synthesize TiO2, mainly including sol-gel methods, direct oxidation methods, and vapor phase methods (CVD PVD).1b,1c Among these methods, direct oxidation and vapor phase deposition methods are suitable for the synthesis of thin films but cannot be adopted to synthesize nonagglomerate NCs. Flame-based gas-phase synthesis is probably the most common industrial route for TiO2 synthesis, which usually yields phase-mixed products with lack control over shape and size. Sol-gel methods (including hydrolytic and nonhydrolytic routes) are the most successful synthetic methods for inorganic nanocrystals, including TiO2.2c,14 Phase-controlled synthesis of anatase NCs has been achieved via several routes,1b,1c,9,15 but the shape and size controlling over the anatse NCs is still a great challenge. Ionic liquids (ILs), which can be also called room temperature molten salts, typically composed of organic cations and large anions, have attracted great attention due to their many unique properties such as low volatility, wide liquidus, good thermal stability, good dissolving ability, strongly designable structures, high ionic conductivity, and a large electrochemical window.16 These excellent properties have stimulated numerous studies on the application of ILs. In recent years, many inorganic nanostructures have been fabricated via various IL-involved processes, including electrodeposition, chemoreduction, sol-gel, and solvothermal routes.17 The most attractive properties of ILs for the synthesis of TiO2 NCs are the limited ligand protection and the facility on tuning the composition of the reactants, thus controlling over the aggregation manner of anatase and rutile NCs. In Zhou’s work17d reported in 2003, mesoporous spherical aggregates composed of unltrasmall anatase NCs were obtained from TiCl4/water/[bmim]þ[BF4]- (volume ratio, 1/2/10) system after heating 12 h at 80 °C. In our recent work,5 anatase NCs with uniform size and shape were also synthesized via a microwaveassisted route in [bmim]þ[BF4]-. We are rather wondering why these anatase NCs did not transformed into rutile phase. In our subsequent study, we discovered that a lot of F- ions existed in the anatase NCs confirmed by XPS analysis. It has been reported18 (13) (a) Penn, R. L.; Banfield, J. F. Am. Mineral. 1998, 83, 1077. (b) Penn, R. L.; Banfield, J. F. Am. Mineral. 1999, 84, 871. (c) Penn, R. L.; Banfield, J. F. Science 1998, 281, 969. (d) Penn, R. L.; Banfield, J. F. Geochim. Cosmochim. Acta 1999, 63, 1549. (14) (a) Livage, J.; Henry, C.; Sanchez, C. Prog. Solid State Chem. 1988, 18, 259. (b) Sugimoto, T. Monodispersed Particles; Elsevier: Amsterdam, 1988. (c) Park, J.; Joo, J.; Kwon, S. G.; Jang, Y.; Hyeon, T. Angew. Chem., Int. Ed. 2007, 46, 4630. (d) Niederberger, M.; Garnweitner, G. Chem.;Eur. J. 2006, 12, 7282. (15) (a) Chemseddine, A.; Moritz, T. Eur. J. Inorg. Chem. 1999, 1999, 235. (b) Trentler, T. J.; Denler, T. E.; Bertone, J. F.; Agrawal, A.; Colvin, V. L. J. Am. Chem. Soc. 1999, 121, 1613. (c) Jun, Y.; Casula, M. F.; Sim, J.; Kim, S. Y.; Cheon, J.; Alivisatos, A. P. J. Am. Chem. Soc. 2003, 125, 15981. (d) Zhang, Z. H.; Zhong, X. H.; Liu, S. H.; Li, D. F.; Han, M. Y. Angew. Chem., Int. Ed. 2005, 44, 3466. (e) Joo, J.; Kwon, S. G.; Yu, T.; Cho, M.; Lee, J.; Yoon, J.; Hyeon, T. J. Phys. Chem. B 2005, 109, 15297. (f ) Tang, J.; Redl, F.; Zhu, Y. M.; Siegrist, T.; Brus, L. E.; Steigerwald, M. L. Nano Lett. 2005, 5, 543. (g) Pan, D. C.; Zhao, N. N.; Wang, Q.; Jiang, S. C.; Ji, X. L.; An, L. J. Adv. Mater. 2005, 17, 1991. (h) Li, X. L.; Peng, Q.; Yi, J. X.; Wang, X.; Li, Y. D. Chem.;Eur. J. 2006, 12, 2383. (i) Wu, B. H.; Guo, C. Y.; Zheng, N. F.; Xie, Z. X.; Stucky, G. D. J. Am. Chem. Soc. 2008, 130, 17563. (16) (a) Welton, T. Chem. Rev. 1999, 99, 2071. (b) Wasserscheid, P.; Welton, T. Inonic Liquids in Synthesis; Wiley-VCH: Weinheim, 2002.(c) Dupont, J.; Suoza, R. F.; Suarez, P. A. Z. Chem. Rev. 2002, 102, 2667. (d) P^arvulescu, V. I.; Hardacre, C. Chem. Rev. 2007, 107, 2615. (17) (a) Antonietti, M.; Kuang, D. B.; Smarsly, B.; Zhou, Y. Angew. Chem., Int. Ed. 2004, 43, 4988. (b) Taubert, A.; Li, Z. H. Dalton Trans. 2007, 723. (c) Endres, F.; Abedin, S. Z. E. Phys. Chem. Chem. Phys. 2006, 8, 2101. (d) Zhou, Y.; Antonietti, M. J. Am. Chem. Soc. 2003, 125, 14960. (e) Kaper, H.; Endres, S.; Djerdj, I.; Antonietti, M.; Smarsly, B. M.; Maier, J.; Hu, Y. S. Small 2007, 3, 1753. (f) Yu, N. Y.; Gong, L. M.; Song, H. J.; Liu, Y.; Yin, D. H. J. Solid State Chem. 2007, 180, 799. (g) Zheng, W. J.; Liu, X. D.; Yan, Z. Y.; Zhu, L. J. ACS Nano 2009, 3, 115. (18) Miao, Z. J.; Liu, Z. M.; Ding, K. L.; Han, B. X.; Miao, S. D.; An, G. M. Nanotechnology 2007, 18, 125605.

5130 DOI: 10.1021/la903600q

Ding et al.

that BF4- could slowly decompose into F- and BO33-. Since Fion was reported19 to be an inhibitor for phase transformation from anatase to rutile, it is quite reasonable to suppose that Ftook charge of the inhibition of the rutile phase formation in the [bmim]þ[BF4]- reaction systems. Herein we use the ionic liquid 1-butyl-3-methylimidazolium chloride (bmimþCl-) as the solvent and TiCl4 as the precursor to synthesize TiO2 NCs via a hydrothermal route. HF and H2SO4 were used as additives to inhibit the phase transformation from anatase to rutile, and their effects on the shape and size of phasepure anatase NCs were studied via tuning the reactant composition. In addition, the aggregation of the NCs was also investigated.

2. Experimental Section Reagents. 1-Methylimidazole was provided by Kai Le Chemical Plant; all the other chemicals were purchased from Sinopharm Chemical Reagent Beijing Co., Ltd. 1-Methylimidazole and 1-chlorobutane were distilled before use. Synthesis of bmimþCl- and [bmimþ]2[TiCl6-x(OH)x]2-.

The procedures to synthesize bmimþCl- and [bmimþ]2[TiCl6-x(OH)x]2- were the same as those described in a separate paper which is on the synthesis of rutile NCs.20 Synthesis of TiO2 NCs. The proper amount of [bmimþ]2[TiCl6-x(OH)x]2- was mixed with water and HF solution (or H2SO4) to form a transparent yellowish solution, and bmimþClwas then added, followed by sonication. The resulting transparent solution was transferred into a Teflon-lined stainless autoclave and heated. (Detailed experimental parameters are given in Table S1 in the Supporting Information, where [bmimþ]2[TiCl6-x(OH)x]2- was divided into bmimþCl- and TiCl4-x(OH)x for the convenience of comparison.) After cooling to room temperature, the white solid product was washed at least five times by centrifugation and redispersion in ethanol. Characterization. The SEM observation was conducted on a Hitachi-s4300 electron microscope operated at 15.0 kV. The TEM observation was performed on a transmission electron microscope (JEM 1011) at an accelerating voltage of 100 kV. Highresolution TEM (HRTEM) observation was carried out on JEM 2011 at an accelerating voltage of 200 kV. X-ray photoelectron spectroscopy (XPS) measurement was performed on an ESCALab220i-XL spectrometer using Al KR as the exciting source (hν = 1486.6 eV). The X-ray diffraction (XRD) patterns were collected on an X-ray diffracometer (Rigaku D/max2500) with Cu KR1 radiation (λ = 1.540 60 A˚). The phase composition of the products was estimated from the following equation: R% = (1 þ 0.8IA/IR)-1, where R% is the weight fraction of rutile in the powder and IA and IR are the peak intensities of the anatase (101) and rutile (110) diffractions, respectively.21

3. Results and Discussion 3.1. Phase Control. In a separate paper,20 it was demonstrated that anatase NCs were formed at early stage in the TiCl4/ H2O/bmimþCl- system and would gradually phase transform into rutile phase. In order to obtain phase-pure anatase NCs in hydrothermal synthesis, phase transformation inhibitor is always needed. F- and SO42- ions are two most commonly used inhibitors for anatase to rutile phase transformation.19b,22-24 In (19) (a) Yin, H. B.; Wada, Y.; Kitamura, T.; Kambe, S.; Murasawa, S.; Mori, H.; Sakata, T.; Yanagida, S. J. Mater. Chem. 2001, 11, 1694. (b) Wu, M. M.; Lin, G.; Chen, D. H.; Wang, G. G.; He, D.; Feng, S. H.; Xu, R. R. Chem. Mater. 2002, 14, 1974. (20) Ding, K. L.; Miao, Z. J.; Hu, B. J.; An, G. M.; Sun, Z. Y.; Han, B. X.; Liu, Z. M., submitted. (21) Spurr, R. A.; Myers, H. Anal. Chem. 1957, 29, 760. (22) Yan, M.; Chen, F.; Zhang, J.; Anpo, M. J. Phys. Chem. B 2005, 109, 8673. (23) Li, J.; Zeng, H. C. J. Am. Chem. Soc. 2007, 129, 15839. (24) (a) Pan, J. H.; Zhang, X. W.; Du, J. H.; Sun, D.; Leckie, J. O. J. Am. Chem. Soc. 2008, 130, 11256. (b) Zhou, L.; Smyth-Boyle, D.; O'Brien, P. J. Am. Chem. Soc. 2008, 130, 1309.

Langmuir 2010, 26(7), 5129–5134

Ding et al.

Figure 1. XRD patterns of samples listed in Table S1.

this work, HF and H2SO4 were used as additives in the TiCl4/ H2O/bmimþCl- system, and the phase transformation was totally inhibited, leading to phase-pure anatase NCs. Experimental details and phase compositions of the samples are shown in Table S1 and Figure 1. While phase-pure anatase NCs can be obtained in the presence of F- and SO42- ions, to the best of our knowledge, there is no convincing mechanism for the formation of phase-pure anatase NCs induced by F- and SO42- ions. Researchers used to attribute this to the inhibition of the homogeneous rutile nucleation. In our knowledge, the most acceptable mechanism of anatase to rutile phase transformation is based on the heterogeneous rutile nucleation. Thus, the theory of inhibiting the phase transformation via the inhibition of the homogeneous rutile nucleation seems invalid. According to the “soft hard acid base” principle, Ti4þ should be classified as a hard acid, while F- and SO42- as hard bases. Thus, the F- and SO42- ions can be strongly adsorbed on the surfaces of anatase NCs and alter their surface structures. Since the formation of rutile-structured interfacial nuclei was highly dependent on the anatase surface structure, it is not surprising that the adsorption of F- and SO42- ions would greatly hinder the phase transformation process. A considerable amount of F and S elements was found in samples F2 and S2 by the XPS analysis. As shown in Figure S1, the XPS peak at 684 eV indicates that the F element existed in the form of F- ions, which may be adsorbed on the surfaces but not substituting the O2- ions in crystal lattice.23,25 For sample S2, the peak at 168 eV shows that S element was in the form of SO42ions.26 The XPS spectra of both samples clearly show the presence of N element, which should originate from bmimþ, suggesting that bmimþ cations were adsorbed on the surface of TiO2 NCs. The bmimþ cation adsorption layer formed on the particle surface could be regarded as a steric protection shell, which reduced the aggregation of anatase NCs. 3.2. Shape Control. In this work, HF and H2SO4 not only acted as phase transfomation inhibitor but also played important roles in controlling over the shape of anatase NCs. For the HFmediated synthesis, most anatase NCs displayed square-plate shape under SEM and low-magnification TEM observations, as shown in Figure 2. HRTEM analysis were carried out to determine the shape of the anatase NCs. Figure 2f shows a HRTEM image of the top view of a platelike NC. The lattice fringes could be indexed into (200) and (020) planes via measuring the lattice spacing and interfacial angle. The side view HRTEM (25) Yu, J. C.; Yu, J. G.; Ho, W. K.; Jiang, Z. T.; Zhang, L. Z. Chem. Mater. 2002, 14, 3808. (26) Gichuhi, A.; Shannon, C.; Perry, S. S. Langmuir 1999, 15, 5654.

Langmuir 2010, 26(7), 5129–5134

Article

image showed (101) and (101) lattice fringes, as shown in Figure 2g. These analyses clearly indicate that the square-platelike NCs were actually in highly truncated bipyramid shape. A simulated shape is given in Figure 2i. The truncated bipyramid possesses two large {001} facets and eight small {101} facets. The average sizes along [001] and [100] directions were 20.1 and 23.4 nm, respectively, calculated from the XRD patterns using the Scherrer equation. Most of these NCs preferred to sitting on their larger facets and thus exhibited square-plate shape under electron microscopy observations. These results are quite similar to that obtained in our previous work.5 For the H2SO4-mediated synthesis, elongated NCs were obtained, as shown in Figure 2j. Using the Scherrer equation, the average sizes along [001] and [100] directions were calculated to be 23.2 and 15.3 nm, respectively. The HRTEM image shown in Figure 2k displayed (101) and (101) lattice fringes. These NCs were elongated along the [001] direction, and the zigzag side surfaces were mainly {101} facets. A simulated shape of the NCs is shown in Figure 2l. For convenience to understand, the structure could be regarded as composing of two kinds of building blocks: truncated pyramid and parallelepiped. These building blocks were connected by sharing the {001} facets. Many holes and cracks (marked by the arrows in Figure 2b,j) were observed via careful view, revealing that these NCs may be formed by the aggregation of small NCs and the subsequent ripening process. To explore the formation mechanism of these anatase NCs, samples F1 and S1 obtained at short reaction time of 3 h were examined. As shown in Figure 3, NCs with size about 10 nm were observed, along with the oriented aggregated larger NCs. Since the anatase yields almost remained unvaried (∼70%) from 3 to 48 h for both HF- and H2SO4-mediated synthesis, there is no doubt that the aggregation and ripening process did exist and determine the final shape and size of the products. Here we denote the small-sized nonaggregated NCs as primary NCs and the aggregated ones as secondary NCs. The HRTEM images (Figure 3b,e) show that the shapes of the primary NCs obtained from the HF- and H2SO4-mediated processes were quite different. This could be owed to the surface complexation of Fand SO42- ions, which altered the surface free energy comparisons. The F- ions were supposed to be preferentially adsorbed on the anatase {001} facet, leading to the highly truncated bipyramidshaped primary NCs. Independent of our findings, several excellent works6,7 have also reported the F- effect on the shape controlling over anatase NCs, in all of which F- ions were discovered to be adsorbed strongly on {001} facet of anatase NCs, inhibiting the growth of NCs along [001] direction. Compared to the F- ions, SO42- ions were supposed to be less selectively adsorbed, thus leading to slightly truncated bipyramid-shaped primary NCs. The formation model of the anatase NCs is described in Scheme 1, in which the whole process is divided into four stages. The first stage was the hydrolysis of the Ti precursor, during which the ligands around the Ti atoms were substituted by hydroxyl groups via the attack by water molecules; thus, the Ti-oxo species were formed and could serve as the monomers for the nucleation and growth of TiO2 NCs. Stage 2 was the nucleation of anatase NCs. The classical LaMer model can be introduced here to deal with this stage.14b,27 The concentration of the Ti-oxo species increased rapidly to an extremely high level due to the fast hydrolysis reaction. Once the degree of supersaturation was high enough to overcome the energy barrier for nucleation, nucleation burst occurred. Thus, numerous anatase primary NCs were formed, and the concentration of the Ti-oxo (27) LaMer, V. K.; Dinegar, R. H. J. Am. Chem. Soc. 1950, 72, 4847.

DOI: 10.1021/la903600q

5131

Article

Ding et al.

Figure 2. SEM (a), TEM (b, c), and HRTEM (d, f, g) images and shape simulation (i) of sample F2 (HF, 150 °C, 2 days); (f) magnified TEM image of (d); (e, h) FFT images corresponding to (f, g), respectively; TEM (j) and HRTEM (k) images and shape simulation (l) of sample S2 (H2SO4, 150 °C, 2 days).

species rapidly decreased into a lower level. The third and last stage were the aggregation and Ostwald ripening process, respectively, resulting in the formation of anatase secondary NCs, i.e., the final product. Different aggregation manners occurred in HFand H2SO4-mediated processes due to the natures of primary NCs. Anatase NCs tend to nucleate in truncated bipyramid shape with eight {101} facets and two {001} facets exposed. For Fcomplexed anatase primary NCs, {101} facets possessed higher surface energy, and they tended to aggregate toward {101} facets 5132 DOI: 10.1021/la903600q

to eliminate high-energy facets. While for SO42- complexed anatase primary NCs {001} facets possessed higher surface energy, they tended to aggregate along [001] directions. Thus, platelike and elongated secondary NCs were formed via these two different routes, respectively. After the ripening process, most of the defects on NCs surfaces were eliminated. However, there still remained some holes and cracks inside the NCs. Furthermore, some primary NCs could grow up by Ostwald ripening without aggregation, yielding defect-free secondary NCs. Langmuir 2010, 26(7), 5129–5134

Ding et al.

Article

Figure 3. TEM (a) and HRTEM (b) images and ED pattern of sample F1 (HF, 150 °C, 3 h); (c) was taken from the whole partile in (b); TEM (d) and HRTEM (e) images of sample S1 (H2SO4, 150 °C, 3 h).

Scheme 1. Brief Model for the Anatase NCs Formation and Shape Evolution

3.3. Size Control. As indicated in the formation mechanism of the anatase NCs, the aggregation and Ostwald ripening play the most important roles in controlling the size of anatase NCs. Several experiments have been carried out to study the influences of the hydrothermal conditions on the NC sizes. SEM and TEM images of these samples are given in Figure 4. It is clearly shown that the NC sizes could be tuned in the range of 10-60 nm with low polydispersity. Samples F2 and S2 were obtained at 150 °C, while F3 and S3 were prepared at 200 °C under the conditions as listed in Table S1. From the images shown in Figure 4a,b,e,f and Figure 4i,j,m,n, it can be observed that higher temperature led to larger NCs. Not only the degree of aggregation of the NCs was increased, but also the Ostwald ripening process was accelerated, both of which decreased the particle number of the secondary NCs, thus increasing the size of the NCs. Langmuir 2010, 26(7), 5129–5134

Comparing the images of samples F2, F4, F5 and samples S2, S4, S5 displayed in Figure 4, it can be concluded that higher Ti precursor concentration resulted in larger NCs. On one hand, increasing the Ti/ILs ratio lowered the protecting ability of the IL adsorption layer to the primary NCs, facilitating the aggregation of NCs. On the other hand, the increasing of Ti precursor concentration enhanced the solution acidity, which increased the Ti-O bond activity and Ti-oxo concentration, thus accelerating the Ostwald ripening process. For the H2SO4mediated syntheses, higher Ti precursor concentration increased the anisotropy of the elongated NCs due to the enhanced aggregation. Interesting phenomena were observed as studying the influences of the HF and H2SO4 concentration on the anatase NCs morphology. It was found that higher additive (F- and SO42ions) concentration led to higher degree of aggregation and larger sized NCs. If the concentration was kept increased, it would result DOI: 10.1021/la903600q

5133

Article

Ding et al.

Figure 4. SEM (a-d, i-l) and TEM (e-h, m-p) images of samples F2-F5 and S2-S5: (a, e) sample F3; (b, f) sample F2; (c, g) sample F4; (d, h) sample F5; (i, m) sample S3; ( j, n) sample S2; (k, o) sample S4; (l, p) sample S5. Samples F3 and S3 were obtained at 200 °C, while all the others were obtained at 150 °C with different Ti precursor concentration (see the details in Table S1).

in extensive aggregation of the NCs. Figure S2 shows the TEM images and SAED patterns of sample F6 along different directions, indicating the pseudo single crystal nature of the aggregates. These aggregates were formed by {101} facets oriented attachment of highly truncated anatase NCs with small size. As shown in Figure S2i, hundreds-nanometer-sized aggregates were formed at high H2SO4 concentration. The HRTEM image (Figure S2j) and SAED pattern (Figure S2l) of the small aggregates clearly show the oriented attachment of small anatase NCs along the [001] direction. The SAED pattern of the larger sized aggregate (Figure S2n) shows the characteristic of imperfect oriented attachment. As indicated in Scheme 1, the different aggregation manner may originate from the surface complexation by different ligands, e.g., F- and SO42- ions. Although there have been many reports about the oriented attachment of anatase NCs,13,15c,28,29 this kind of the direction modulation of the anatase NCs oriented attachment has never been reported before. (28) (a) Lee, E. J. H.; Ribeiro, C.; Longo, E.; Leite, E. R. J. Phys. Chem. B 2005, 109, 20842. (b) Ribeiro, C.; Vila, C.; Matos, M. E.; Bettini, J.; Longo, E.; Leite, E. R. Chem.;Eur. J. 2007, 13, 5798. (29) (a) Polleux, J.; Pinna, N.; Antonietti, M.; Niederberger, M. Adv. Mater. 2004, 16, 436. (b) Polleux, J.; Pinna, N.; Antonietti, M.; Hess, C.; Wild, U.; Schl€ogl, R.; Niederberger, M. Chem.;Eur. J. 2005, 11, 3541.

5134 DOI: 10.1021/la903600q

4. Conclusions In conclusion, we have demonstrated an IL-assisted hydrothermal method to synthesize anatase TiO2 NCs. The surfactantlike nature of IL played key roles in controlling the crystallization process via controlling the aggregation manner of the NCs. Shape- and size-controlled anatase NCs were obtained in the presence of HF or H2SO4. F- and SO42- ions were supposed to be selectively adsorbed on different facets of anatase NCs, which greatly influenced the shape and aggregation manners of primary anatase NCs, thus determining the final shape of the NCs. We believe that this study will open a new way for controlled synthesis of TiO2 NCs and will facilitate the studies on the phase, shape, and size effect of TiO2 in various applications. Acknowledgment. We thank Dr. Hongbo Li for help in XRD analysis. This work is financially supported by the Ministry of Science and Technology of China (973 project, 2009CB930802) and the Chinese Academy of Sciences (KJCX2.YW.H16). Supporting Information Available: Synthetic conditions and the resultant anatase samples (Table S1); XPS spectra (Figure S1) and SEM, HRTEM, and ED patterns (Figure S2). This material is available free of charge via the Internet at http://pubs.acs.org.

Langmuir 2010, 26(7), 5129–5134