A “teardown” method to create large mesotunnels on the pore walls of ordered mesoporous silica

A “teardown” method to create large mesotunnels on the pore walls of ordered mesoporous silica

Journal of Colloid and Interface Science 328 (2008) 338–343 Contents lists available at ScienceDirect Journal of Colloid and Interface Science www.e...

891KB Sizes 0 Downloads 8 Views

Journal of Colloid and Interface Science 328 (2008) 338–343

Contents lists available at ScienceDirect

Journal of Colloid and Interface Science www.elsevier.com/locate/jcis

A “teardown” method to create large mesotunnels on the pore walls of ordered mesoporous silica Dong Gu, Fuqiang Zhang, Yifeng Shi, Fan Zhang, Zhangxiong Wu, Yonghui Deng, Lijuan Zhang, Bo Tu, Dongyuan Zhao ∗ Department of Chemistry, Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, and Laboratory of Advanced Materials, Fudan University, Shanghai 200433, PR China

a r t i c l e

i n f o

a b s t r a c t A “teardown” method to create large mesotunnels (∼9 nm) on the pore walls of ordered mesoporous silicas is demonstrated by digesting the organic constituents from polymer–silicate nanocomposites. The ordered mesostructured polymer–silicate composites were first obtained via the evaporation-induced triconstituent co-assembly method by using a low-molecular-weight phenolic resin (resols) as an organic precursor; prehydrolyzed TEOS as an inorganic precursor, and triblock copolymer F127 as a template. All of organic components including F127 and phenolic resins are removed by the microwave digestion (MWD) method from mesostructured polymer–silica composites. While the removal of triblock copolymer F127 generates main pore channels, the phenolic resins can also be torn down from the pore walls, yielding mesotunnels between the channels. The resulting silica products exhibit ordered 2-D hexagonal mesostructure, large pore volume (up to 1.92 cm3 /g), and very large pore size (up to 22.9 nm), which is even larger than their mesostructural cell parameter (14.2 nm). TEM images confirm the existence of mesotunnels on the silica pore walls. FT-IR and 29 Si solid-state NMR results reveal that these silica products have a large number of silanol groups. © 2008 Elsevier Inc. All rights reserved.

Article history: Received 28 May 2008 Accepted 7 September 2008 Available online 23 September 2008 Keywords: Mesoporous Silica Mesotunnels Microwave digestion (MWD)

1. Introduction Ordered mesoporous molecular sieves with high surface area, large pore size and pore volume, have attracted much attention because of their potential applications in catalysis, separation, adsorption, and biotechnologies [1–6]. Up to now, a large number of highly ordered mesoporous silica materials (M41S [1,7], SBA [8,9], MSU [10], FDU [5,11,12], HMS [13,14], and KIT [15], etc.) have been successfully synthesized. The control of the pore structure [5] and size [16] is a key issue for the functionalities of mesoporous silicas with variable applications. With the discovery of large pore mesoporous materials [8,9], amphiphilic block copolymers have turned out to be a kind of valuable templates for their facile structuredirecting ability, low cost, and easy to be removed. Adding organic swelling agent such as 1,3,5-trimethylbenzene (TMB) can further enlarge the mesopore size [17]. However, two-dimensional (2-D) hexagonal mesoporous silicas SBA-15 have 1-D pore channels with disadvantages in the diffusion and transport of large molecules, which may limit their applications in catalysis and separation. In addition, 3-D cubic mesoporous silica materials with large pore


Corresponding author. Fax: +86 21 5566 4192. E-mail address: [email protected] (D. Zhao).

0021-9797/$ – see front matter doi:10.1016/j.jcis.2008.09.043

© 2008

Elsevier Inc. All rights reserved.

sizes (SBA-16, FDU-1, and FDU-12, etc.) have cage-like pore structures with small-size windows which block the large molecules into the cages. Therefore, it is challenging to synthesize ordered mesoporous silicas with 3-D opened large-pore channels. It is known that there are a plenty of micropores on the silica walls of SBA-15, which is generated from partial occlusion of poly(ethylene oxide) (PEO) chains of the triblock polymers [18–22]. By a high-temperature hydrothermal process assisted with adding TMB as a swelling agent [23], the micropores on the silica walls can be transformed into mesotunnels, so SBA-15 with 3-D connected mesopores can be obtained. However, the addition of swelling agents leads to remarkable degradation of the mesostructure, and the size of mesotunnels is further difficult to control. Here, we report a “teardown” method to prepare 3-D pore connected mesoporous silica with large mesotunnels on the walls by digesting the triblock copolymer F127 template and phenolic resins from the polymer–silica nanocomposites. The obtained silica products have an ordered mesostructure with average symmetry of hexagonal p6mm (space group). The large mesopore tunnels generated from removal of resin components are as large as 9 nm. The teardown method also yields mesoporous silica with a large average pore size (up to 22.9 nm), which is even larger than the cell parameter a0 (14.2 nm), and large pore volume (up to 1.92 cm3 /g), which may make them much useful in proteolysis [24].

D. Gu et al. / Journal of Colloid and Interface Science 328 (2008) 338–343

Table 1 Preparation conditions of polymer/silica triconstituent nanocomposites by the EISA method. Sample name

TEOS (g)

Resola (g)

F127 (g)

0.1 mol L−1 HCl (g)

EtOH (g)

S-0 S-0.08 S-0.16 S-0.50 S-1.00

2.08 2.08 2.08 2.08 2.08

0 0.08 0.16 0.50 1.00

1.0 1.0 1.0 1.0 1.6

1.0 1.0 1.0 1.0 1.0

12.0 12.0 12.0 12.0 12.0

a The weight of the resol is calculated from the addition of original material phenol and formaldehyde.

2. Materials and methods


of 30 wt% H2 O2 in the autoclave, and then preformed at the microwave with working frequency of 2450 MHz and the voltage of 220 V for about 11 min. The samples were designated as S-x-MWD, where x represented the amount of resol used for per 0.01 mol of TEOS, “MWD” denoted that the method for removal of the organic constituents was microwave digestion. For comparison, the as-made samples were also calcined at 550 ◦ C in air for 5 h to burn out template F127 and phenolic resins completely. They were designated as S- y-Cal, where y represented the same meaning as MWD samples, and “Cal” denoted that the method for removal of the organic constituents was calcination. 2.5. Characterization

2.1. Chemicals Triblock copolymer poly(ethylene oxide)-b-poly(propylene oxide) Pluronic F127 (M w = 12,600, EO106 –PO70 –EO106 ) was purchased from Acros Corp. Other chemicals were purchased from Shanghai Chemical Corp. All chemicals were used as received without any further purification. Millipore water was used in all experiments. 2.2. Synthesis of resol precursors Resol, a low-molecular-weight, soluble phenolic resin, was prepared from phenol and formaldehyde by a base-catalyzed process according to our previous report [25]. In a typical procedure, 0.61 g of phenol was melted at 40–42 ◦ C in a flask and mixed with 0.13 g of 20 wt% sodium hydroxide (NaOH) aqueous solution under stirring. After 10 min, 1.05 g of formalin (37 wt% formaldehyde) was added dropwise below 50 ◦ C. Upon further stirring for 1 h at 75–80 ◦ C, the mixture was cooled to room temperature. The pH value was adjusted with HCl solution until it reached a value of 6–7, and water was removed by vacuum evaporation below 50 ◦ C. The final product was dissolved in ethanol as the concentration of phenol was adjusted to 12.1 wt%. The molar ratio of phenol/formaldehyde/NaOH was 1:2:0.1. The average molecular weight of resol precursors used in this paper was smaller than 500, which was determined by gel permeation chromatography (GPC). 2.3. Synthesis of mesostructured polymer–silica nanocomposites Mesostructured polymer–silica nanocomposites were prepared by the triconstituent co-assembly of resols, silicate oligomers from tetraethyl orthosilicate (TEOS) and triblock copolymer F127 template (Table 1). In a typical preparation, 1.0 g of copolymer F127 was dissolved in 8.0 g of ethanol with 1.0 g of 0.2 M HCl and stirred for 1 h at 40 ◦ C to afford a clear solution. 2.08 g of TEOS and 5.0 g of 12.1 wt% phenol–ethanol solution were added in sequence. After further stirring at 40 ◦ C for 5 h, the mixture was transferred into dishes. It took 5–8 h at room temperature to evaporate ethanol and 24 h at 100 ◦ C in an oven for the thermopolymerization. The as-made nanocomposites were scraped from the dishes and ground into fine powders. 2.4. Preparation of large tunnel mesoporous silicas The large tunnel mesoporous silicas were prepared by the microwave digestion (MWD) of the as-made polymer–silica nanocomposites to remove all of organic components. The microwave digestion was performed with a microwave sample preparation system model (MK-III, Shin-Cor. Institute of Microwave Decomposition and Test Technology) in a Teflon autoclave operated at approximately 1200 W. The pressure and temperature were controlled to be lower than 1.3 MPa and 200 ◦ C, respectively. 0.3 g of the sample was approximately dissolved with 4.5 ml of 65–68 wt% HNO3 and 1.5 ml

Small-angle X-ray scattering (SAXS) measurements were taken on a Nanostar U small-angle X-ray scattering system (Bruker, Germany) using CuK α radiation (40 kV, 35 mA). The d-spacing values were calculated by the formula d = 2π /q and the √ unit cell parameters were calculated from the formula a = 2d10 / 3. Nitrogen sorption isotherms were measured at 77 K with a Quantachrome’s Quadrasorb SI analyzer. Before the measurements, the sample was degassed at 150 ◦ C for more than 6 h in vacuum. The Brumauer– Emmett–Teller (BET) surface areas were calculated using experimental points at a relative pressure ( P / P 0 ) of 0.05–0.25. The total pore volume was calculated from the N2 amount adsorbed at the P / P 0 of 0.99 for each sample. Transmission electron microscopy (TEM) experiments were conduced on a JEOL 2011 microscope (Japan) operated at 200 kV. The samples for TEM measurements were suspended in ethanol and supported onto a holey carbon film on a Cu grid. Fourier transform infrared (FTIR) spectra were collected on Nicolet Fourier spectrophotometer (USA), using KBr pellets of the solid samples. 29 Si solid-state NMR (nuclear magnetic resonance) spectra were performed on a Bruker DSX300 spectrometer with a frequency intensity of 62.5 kHz, and the reference sample of Q8M8 ([(CH3 )3 SiO]8 Si8 O12 ). Thermogravimetric analysis (TGA) was monitored using a Mettler Toledo TGA/SDTA851 analyzer (Switzerland) from 25 to 900 ◦ C under air with a heating rate of 5 ◦ C/min. 3. Results and discussion Mesostructured polymer/silica nanocomposites can be prepared by the triconstituent co-assembly of the resols, silicate oligomers, and triblock copolymers via an EISA approach [26]. The as-made nanocomposite membranes are light flaxen in color and crackingfree in morphology. The SAXS pattern of as-made nanocomposites (Fig. 1a) shows three well-resolved diffraction peaks, which are associated with 10, 20, and 21 reflections of 2-D hexagonal symmetry with the space group of p6mm. After calcination at 550 ◦ C in air for 5 h, thermogravimetric analysis curve (Supporting information, Fig. 1s) shows that the weight remains stable after about 250 ◦ C and the elemental analysis reveals that the carbon and hydrogen contents are 1.4 and 1.4 wt% respectively. This reveals that both template F127 and phenolic resins can be removed completely, and the pure silica composition is obtained (denoted as S-0.5-Cal). The SAXS pattern (Fig. 1b) also shows three-resolved diffraction peaks, which can be indexed to the 10, 11, 20 reflections of p6mm hexagonal mesostructure, suggesting a thermal stability of the silica structure. The structural shrinkage calculated from X-ray data is as large as about 25.5% (Table 2). Similarly, after the microwave digestion treatment of H2 O2 and HNO3 at 1.0 MPa for 11 min to remove organic components, the mesoporous nanocomposites (assigned to S-0.5-MWD) also show at lest three resolved scattering peaks in the SAXS pattern (Fig. 1c), indicating a highly ordered hexagonal mesostructure. The cell parameter is calculated


D. Gu et al. / Journal of Colloid and Interface Science 328 (2008) 338–343

Fig. 1. SAXS patterns of (a) S-0.5 (as-made), (b) S-0.5-Cal, (c) S-0.5-MWD. Table 2 Physicochemical properties of mesoporous silica prepared with different resol amount via triconstituent ELISA method. Sample name

Treatment Unit cell parameter a0 (nm)a

BET surface Pore area size (m2 /g)b (nm)c

Pore volume (cm3 /g)d

Wall thickness (nm)e

S-0 S-0-MWD S-0-Cal S-0.08 S-0.08-MWD S-0.08-Cal S-0.16 S-0.16-MWD S-0.16-Cal S-0.50 S-0.50-MWD S-0.50-Cal S-1.00 S-1.00-MWD S-1.00-Cal

As-made MWD Cal As-made MWD Cal As-made MWD Cal As-made MWD Cal As-made MWD Cal

– 455 282 – 781 400 – 747 406 – 518 283 – 747 355

– 0.54 0.29 – 1.03 0.46 – 0.90 0.46 – 1.92 0.47 – 0.69 0.46

– 5.1 6.5 – 3.2 4.9 – 4.2 4.8 –

15.2 13.4 11.8 14.8 14.2 12.2 15.2 14.2 11.5 15.7 14.2 11.7 15.1 11.3 9.5

– 8.3 5.3 – 11.0 7.3 – 10.0 6.7 – 22.9 8.8 – 6.1 5.3

2.9 – 5.2 4.2


Calculated from SAXS data. Calculated by the BJH model from sorption data in a relative pressure range from 0.05–0.25. c Calculated by the BJH model from the adsorption branches of isotherms. d Calculated from N2 amount adsorbed at a relative pressure P / P 0 of 0.99. e Calculated by the formula h = a0 − D, where a0 represents the unit cell parameter and D represents the pore diameter. b

to be as large as 14.2 nm, which is much larger than that of calcined sample S-0.5-Cal, clearly indicating that the MWD treatment causes much less structural shrinkage (9.6%) (Table 2) compared with calcination in air. It is further confirmed that MWD is a mild template-remove method from mesoporous silica [27]. With shortening operation time (2–11 min) and lowering temperature (bellow 200 ◦ C) of MWD treatments, all the organic components can also be removed by the oxidation (see below data), while SAXS patterns show that the ordered hexagonal silica mesostructures are retained

Fig. 2. Nitrogen sorption isotherms and pore size distributions (inset) of mesoporous silica after the template removal by calcination (a) and by MWD (b): (a) S-0.5-Cal, and (b) S-0.5-MWD.

and their cell parameters are similar to that of S-0.5-MWD, implying that the time and temperature are not key factors for the treatment. N2 sorption isotherms show that both S-0.5-Cal and S-0.5-MWD samples are typical type-IV curves (Figs. 2a and 2b), suggesting uniform mesopores. The calcined sample S-0.5-Cal shows capillary condensation steps at a relative pressure of about 0.60–0.75 (Fig. 2a), the mean pore size is calculated to be 8.8 nm. It is interesting that the sample S-0.5-MWD obtained after MWD treatment shows much large condensation steps at a high relative pressure of about 0.8–0.98 (Fig. 2b), clearly suggesting a very large uniform mesopore. The mean pore size and pore volume are calculated to be 22.9 nm and 1.92 cm3 /g, which are much larger than that from the calcined samples. It should note that the pore size of the sample S-0.5-MWD is even larger than its cell parameter (14.2 nm), implying large interconnected mesochannels and mesoporosity. Roughly evaluated from the data, the size of mesochannels may be larger than 9.0 nm. TEM images reveal that the S-0.5-MWD sample has ordered hexagonal pore arrangements (Fig. 3). The pore size is roughly evaluated to be about 9.0 nm, and the wall thickness is between 5–6 nm. The center-to-center distances of adjacent channels are measured to be ∼11.7 nm, in accordance with that from the SAXS data. TEM image taken along the [110] direction (Fig. 3B) clearly shows some mesotunnels randomly distributed on the silica walls (see arrowhead). TEM image along the [001] direction (Fig. 3D) also shows some randomly distributed mesotunnels between the pores. The FT-IR spectrum (Fig. 4) of as-made sample S-0.5 shows several absorption bands at around 2800–3000 and 1350–1500 cm−1 which can be assigned to the C–O and C–H stretching of triblock copolymer F127 and phenolic resin, respectively. While S-0.5-Cal

D. Gu et al. / Journal of Colloid and Interface Science 328 (2008) 338–343

Fig. 3. TEM images of mesoporous silica S-0.50-MWD after the template removal by MWD, view from [110] (A, B) and [001] (C, D) directions.


Fig. 5. 29 Si solid-state NMR spectra of (a) S-0.5 (as-made), (b) S-0.5-Cal, and (c) S-0.5-MWD.

Fig. 4. FT-IR spectra of (a) S-0.5 (as-made), (b) S-0.5-Cal, and (c) S-0.5-MWD.

Fig. 6. SAXS patterns of mesoporous silica after the template removal by calcination (A) and MWD (B). In (A) and (B): (a) S-0, (b) S-0.08, (c) S-0.16, (d) S-0.50, and (e) S-1.00.

and S-0.5-MWD samples show that these bands are nearly indiscernible (Fig. 4), suggesting the efficient removal of the copolymer templates by calcination and MWD treatment. For S-0.5-MWD, the bending bands of Si–OH at around 950 cm−1 is almost similar to that for the as-made sample S-0.5. Compared S-0.5-MWD sample, the calcination sample shows much weaker intensity for the bending bands, suggesting that MWD method can lead to the retention of a large amount of Si–OH groups on the pore walls while the calcination inevitably reduces the numbers. 29 Si solid-state NMR spectra of the as-made sample S-0.5 show three resolved resonance peaks due to the different Si environments of Q4 [(SiO)4 Si] (δ = −108 ppm), Q3 [(SiO)3 Si(OR)] (R = H, Et) (δ = −100 ppm), and Q2 [(SiO)2 Si(OR)2 ] (δ = −90 ppm) (Fig. 5a). The Q3 /Q4 value of the as-made sample is as high as 99.8%, even much higher than that for as-made SBA-15 [27], providing a further evidence for a high abundance of silanol groups. This phenomenon may be related with the presence of phenolic

resins on the pore walls. During the triconstituent co-assembly, a microphase separation occurs, and the silicates in the composites space out by the phenolic resin species, leading a low level polymerization and condensation of the silicates. When triblock copolymer F127 templates and phenolic resins are removed by calcination at 550 ◦ C for 5 h, the intensity of the Q2 and Q3 resonance signals reduces remarkably (Fig. 5b), suggesting the further crosslinking and condensation. While the organic compositions are removed by MWD treatment, S-0.5-MWD sample shows strong Q2 , and Q3 resonance peaks (Fig. 5c). The Q3 /Q4 value of S-0.5-MWD sample is about 38.7%, which is much higher than that (15.3%) for the calcination sample S-0.5-Cal. These results clearly indicate that MWD method can reserve a large number of silanol groups on the pore walls. In order to understand the formation of the mesotunnels, a serial of samples with different resol/TEOS ratios were synthesized (Table 1). The as-made sample with pure silica composition (S-0)


D. Gu et al. / Journal of Colloid and Interface Science 328 (2008) 338–343

Fig. 7. Nitrogen sorption isotherms (A, C) and pore size distributions (B, D) of mesoporous silica after the template removal by calcination (A, B) and MWD (C, D). In all figures, (a) S-0, (b) S-0.08, (c) S-0.16, (d) S-0.50, and (e) S-1.00. The isotherms in (A) are offset vertically by 150 (b), 400 (c), 650 (d), and 1000 cm3 /g (e), respectively, while the isotherms in (C) are offset vertically by 250 (b), 820 (c), 1350 (d), and 2500 cm3 /g (e), respectively.

shows only one relatively weak scattering peak in the SAXS patterns (Supporting information, Fig. 2s). It may be explained that the silica oligomers grow too large in size, which results in a little poor organization of silica–F127 composite mesostructure. With the resol/TEOS ratio increases, the SAXS patterns of the as-made samples become more resolvable. At least three resolved scattering peaks from 2-D hexagonal symmetry with the space group of p6mm can be observed, suggesting that ordered mesostructure is formed. After calcination at 550 ◦ C, SAXS patterns of the calcined mesoporous silicas show three resolved scattering peaks (Fig. 6A), suggesting that ordered 2-D hexagonal mesostructure is retained. The shrinkage calculated from the cell parameters increases from 17.6 to 37.1% with the increase of polymer contents (Table 2). Similarly, the samples with different resol/TEOS ratio after the MWD treatment to remove all the organic species also show three resolved scattering peaks in SAXS patterns (Fig. 6B). Compared with that for the calcined samples, the mesostructural shrinkage of the

MWD samples is much smaller (9.6%), further suggesting a mild template removal. N2 sorption isotherms show that all of the samples obtained both from calcination and MWD treatment have uniform mesopores (Fig. 7 and Table 2). The pore size for the calcined samples increases from 5.3 to 8.8 nm with the ratio of resol/TEOS (Table 2, Fig. 7B), and the wall thickness reduces from 6.5 to 2.9 nm accordingly (Table 2). It is mainly caused by the presence of small mesotunnels on the pore walls. The size of the mesotunnels increases with the resol/TEOS ratio, and so the average pore size also increases and calculated wall thickness decreases. N2 sorption isotherms of MWD samples show capillary condensation steps at a narrow relative pressure which is higher than that for the calcined samples (Fig. 7C). It indicates that the pore sizes of the formers are larger than that of the latter (Supporting information, Fig. 3s). This is because the mesostructural shrinkage of the MWD samples is smaller, mesotunnels on the pore walls are larger, in the case,

D. Gu et al. / Journal of Colloid and Interface Science 328 (2008) 338–343


the pore wall would facilitate diffusion and transport of molecules between the channels. Acknowledgments This work was supported by the NSF of China (20721063, 20641001, 20871030, and 20521140450), the State Key Basic Research Program of the PRC (2006CB932302 and 2006CB202502), Shanghai Sci. & Tech. Committee (06DJ14006), Shanghai Nanotech Promotion Center (0652nm024), Fudan Graduate Innovation Fund, and Shanghai Leading Academic Discipline Project (B108). We thank Dr. S.H. Xie and Y. Chen for experimental assistance. Supporting information

Scheme 1. Schematic representation of the procedure for preparation mesoporous silica with mesotunnels on the pore walls.

and the average pore size of MWD samples is higher than that of the calcined samples. The main pore size of S-0.5-MWD is about 9.0 nm from the TEM measurements, and the pore wall thickness is about 5–6 nm, then 2r + W = ∼22 nm, which clearly conforms the mean pore size calculated from the adsorption branch of the N2 sorption data. It means that the pore walls are run through by the mesotunnels. The average pore size calculated from the desorption branch of the N2 sorption data is about 9.0 nm, accordant to the TEM data, suggesting that the mesotunnel is larger than the window size of 9.0 nm. We speculate on the formation of the large mesotunnels in the pore walls through a “teardown” process. The triblock copolymer F127, resol precursors and silica oligomers can assembly to nanocomposites with 2-D hexagonal (p6mm) mesostructure based on a triconstituent co-assembly process (Scheme 1). The resols serve as the template for the formation of the mesotunnels. First, the resols involve the assembly of the mesostructure through hydrogen-interaction with triblock copolymer F127 to construct the pore wall frameworks. With the cross-linking and condensation of the silicates themselves, the microphase separation occurs between the silicas species and resol species. The interpenetrating frameworks with “reinforced-concrete”-like structure is formed [26]. At the MWD process, the silicate frameworks are rigid and continuous, may not cross-link and condense further, which makes the mesostructural shrinkage be smaller. While the organic species, either triblock copolymer F127 or the phenolic resins can be removed during the MWD treatment. A plenty of large voids is left on the pore walls of silicas, resulting in an abundant of large mesotunnels. On the other hand, during the calcination for the burning out of organic components, the mesostructure greatly shrinks, and the silica species can be further cross-linked and condensed. A lot of voids from phenolic resins are re-filled, resulting in small-sized mesotunnels. 4. Summary A new type of mesoporous silica with 2-D hexagonal structure and large mesotunnels on the pore walls has been synthesized by using a MWD “teardown” method to remove the organic compositions. The resultant mesoporous silicas have a large number of Si–OH surface groups, a high pore volume (up to 1.92 cm3 /g), and very large average pore size (up to 22.9 nm). The mesotunnels on

Supporting information for this article may be found on Science Direct, in online version. Please visit DOI: 10.1016/j.jcis.2008.09.043. SAXS patterns of as-made triconstituent samples. Cell parameters and pore sizes of polymer/silica triconstituent nanocompositions and mesoporous silica varied with the amount of resol. References [1] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vartuli, J.S. Beck, Nature 359 (1992) 710–712. [2] S.W. Kim, S.U. Son, S.I. Lee, T. Hyeon, Y.K. Chung, J. Am. Chem. Soc. 122 (2000) 1550–1551. [3] K. Kageyama, J. Tamazawa, T. Aida, Science 285 (1999) 2113–2115. [4] M.E. Davis, Nature 417 (2002) 813–821. [5] J. Fan, C.Z. Yu, T. Gao, J. Lei, B.Z. Tian, L.M. Wang, Q. Luo, B. Tu, W.Z. Zhou, D.Y. Zhao, Angew. Chem. Int. Ed. 42 (2003) 3146–3150. [6] X.X. Yan, C.Z. Yu, X.F. Zhou, J.W. Tang, D.Y. Zhao, Angew. Chem. Int. Ed. 43 (2004) 5980–5984. [7] J.S. Beck, J.C. Vartuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Schmitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. Mccullen, J.B. Higgins, J.L. Schlenker, J. Am. Chem. Soc. 114 (1992) 10834–10843. [8] D.Y. Zhao, J.L. Feng, Q.S. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Science 279 (1998) 548–552. [9] D.Y. Zhao, Q.S. Huo, J.L. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem. Soc. 120 (1998) 6024–6036. [10] S.A. Bagshaw, E. Prouzet, T.J. Pinnavaia, Science 269 (1995) 1242–1244. [11] X.Y. Liu, B.Z. Tian, C.Z. Yu, F. Gao, S.H. Xie, B. Tu, R.C. Che, L.M. Peng, D.Y. Zhao, Angew. Chem. Int. Ed. 41 (2002) 3876–3878. [12] C.Z. Yu, B.Z. Tian, B. Fan, G.D. Stucky, D.Y. Zhao, Chem. Commun. (2001) 2726– 2727. [13] P.T. Tanev, M. Chibwe, T.J. Pinnavaia, Nature 368 (1994) 321–323. [14] P.T. Tanev, T.J. Pinnavaia, Science 267 (1995) 865–867. [15] R. Ryoo, J.M. Kim, C.H. Ko, C.H. Shin, J. Phys. Chem. 100 (1996) 17718–17721. [16] J. Fan, C.Z. Yu, J. Lei, Q. Zhang, T.C. Li, B. Tu, W.Z. Zhou, D.Y. Zhao, J. Am. Chem. Soc. 127 (2005) 10794–10795. [17] D. Zhao, P. Yang, N. Melosh, J. Feng, B.F. Chmelka, G.D. Stucky, Adv. Mater. 10 (1998) 1380–1385. [18] M. Imperor-Clerc, P. Davidson, A. Davidson, J. Am. Chem. Soc. 122 (2000) 11925–11933. [19] M. Kruk, M. Jaroniec, C.H. Ko, R. Ryoo, Chem. Mater. 12 (2000) 1961–1968. [20] K. Miyazawa, S. Inagaki, Chem. Commun. (2000) 2121–2122. [21] R. Ryoo, C.H. Ko, M. Kruk, V. Antochshuk, M. Jaroniec, J. Phys. Chem. B 104 (2000) 11465–11471. [22] Z. Liu, O. Terasaki, T. Ohsuna, K. Hiraga, H.J. Shin, R. Ryoo, Chem. Phys. Chem. 2 (2001) 229–231. [23] J. Fan, C.Z. Yu, L.M. Wang, B. Tu, D.Y. Zhao, Y. Sakamoto, O. Terasaki, J. Am. Chem. Soc. 123 (2001) 12113–12114. [24] W.Q. Shui, J. Fan, P.Y. Yang, C.L. Liu, J.J. Zhai, J. Lei, Y. Yan, D.Y. Zhao, X. Chen, Anal. Chem. 78 (2006) 4811–4819. [25] Y. Meng, D. Gu, F.Q. Zhang, Y.F. Shi, H.F. Yang, Z. Li, C.Z. Yu, B. Tu, D.Y. Zhao, Angew. Chem. Int. Ed. 44 (2005) 7053–7059. [26] R.L. Liu, Y.F. Shi, Y. Wan, Y. Meng, F.Q. Zhang, D. Gu, Z.X. Chen, B. Tu, D.Y. Zhao, J. Am. Chem. Soc. 128 (2006) 11652–11662. [27] B.Z. Tian, X.Y. Liu, C.Z. Yu, F. Gao, Q. Luo, S.H. Xie, B. Tu, D.Y. Zhao, Chem. Commun. (2002) 1186–1187.