Journal of Colloid and Interface Science 303 (2006) 250–255 www.elsevier.com/locate/jcis
Deposition of thin mesoporous silica films on glass substrates from basic solution Naoki Shimura a , Makoto Ogawa a,b,∗ a Graduate School of Science and Engineering, Waseda University, Nishiwaseda 1-6-1, Shinjuku-ku, Tokyo 169-8050, Japan b Department of Earth Sciences, Waseda University, Nishiwaseda 1-6-1, Shinjuku-ku, Tokyo 169-8050, Japan
Received 10 May 2006; accepted 21 July 2006 Available online 1 September 2006
Abstract Transparent thin (ca. 100 nm) films of silica–surfactant mesostructured materials were deposited on borosilicate glass plates and soda-lime glass tubes from aqueous solutions containing tetraethoxysilane, alkyltrimethylammonium chloride, ammonia, and methanol. By calcination in air, the films became mesoporous (BET surface area of 700–900 m2 g−1 ) with pore diameter 2.0–2.8 nm. © 2006 Elsevier Inc. All rights reserved. Keywords: Supramolecular template; Mesoporous silica; Deposition; Transparent film; Pore size
1. Introduction The preparation and the application of nanoporous silica films have been extensively investigated after the successful preparation of nanostructured and nanoporous silica films by the solvent evaporation method [1,2]. Nanoporous and nanostructured silica films on flat substrates have potential for such uses as low-dielectric-constant (low-k) material for semiconductor electronics [3], nanoreactors for photochemical reaction [4], matrices for laser dyes [5–7], pH sensors [8], and ionic conductors [9]. Due to the wide range of possible uses, there is a demand for films with higher quality, controlled thickness, and easier and reproducible synthesis, besides the variation of porosity, pore size, chemical composition, stability, and pore surface engineering. For film preparation on substrates, several synthetic options have been reported [10–13] besides the solvent evaporation method. To control the macroscopic shape of the sol–gel derived products, reactions under acidic conditions are favorable, partly due to their low and controllable rate compared with those under basic conditions. Ogawa took advantage of the slow sol–gel reaction to combine surfactant template mechanisms for * Corresponding author. Fax: +81 3 3207 4950.
E-mail address:
[email protected] (M. Ogawa). 0021-9797/$ – see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2006.07.049
the successful preparation of mesostructured silica–surfactant films by the solvent evaporation method [1,2]. The deposition of silica–surfactant films by dipping substrate in an acidic precursor solution is also used extensively. Thus, the preparation of mesoporous silica films was usually conducted from acidic precursor solutions containing alkoxysilane and surfactant. On the other hand, there are few examples of the preparation of silica–surfactant hybrid films without using acidic conditions. The formation of silica–surfactant nanostructures on substrate by the adsorption of a component or silica–surfactant from vapor phase has been reported so far [11–13]. To the best of our knowledge, there are few reports on the preparation of mesoporous silica films using basic conditions [12,14–16]. Martin et al. [14] and Nishiyama et al. [12] prepared mesoporous silica films by ammonia vapor after gel deposition. Nishiyama et al. also reported the deposition of mesoporous silica on a Stainless steel membrane by a hydrothermal treatment under a basic condition [15,16]. Very recently, we have reported a novel approach to depositing thin mesostructured silica–surfactant films from basic solutions containing tetraethoxysilane, alkyltrimethylammonium chloride, methanol, and ammonia [17]. Due to its simple operation and versatility, the method is a promising alternative to deposit thin layers on various substrates. This paper reports a detailed experimental procedure for the preparation of mesostruc-
N. Shimura, M. Ogawa / Journal of Colloid and Interface Science 303 (2006) 250–255
tured silica–surfactant films and mesoporous silica films on glass substrate from a basic solution and the characterization of the resulting films. The present film deposition from basic solution opens up new opportunities to prepare mesoporous silica film on various substrates, including those unstable in acidic solutions, organic polymers, and those with complex geometry. 2. Experimental 2.1. Materials Methanol and 28% aqueous ammonia solution were obtained from Kanto Chemical Co., Inc. Tetraethoxysilane (abbreviated as TEOS) and alkyltrimethylammonium chlorides [(Cn H2n+1 )(CH3 )3 NCl (n = 12, 14, 16, and 18); abbreviated as CnTAC, where n denotes the carbon number in the alkyl chain] were obtained from Tokyo Kasei Kogyo Co., Ltd. All chemicals were used without further purification. 2.2. Sample preparation Following is a typical synthetic procedure: C16TAC (0.211 g), deionized water (17.7 g), methanol (100 ml), and 28% aqueous ammonia solution (7.2 g) were mixed in a poly(propylene) bottle (250 ml, As One Corporation) and the solution was shaken for 15 s at room temperature. To this solution was added TEOS (0.368 ml) and then the mixture was shaken for another 3 s. The molar ratio of TEOS:C16TAC:deionized water:methanol:ammonia was 1:0.4:774:1501:72. A plate (borosilicate glass, 30 × 30 × 0.2 mm, Matsunami Glass Ind., Ltd.) or a capillary tube (soda-lime glass, Hilgenberg GmbH) was soaked into the solution perpendicularly as shown in Fig. 1 and aged at room temperature for 20 h. The products were washed with methanol and were dried at room temperature for 1 day. To remove surfactant, the products were calcined in air at 350 ◦ C for 10 h at a heating rate of 2.5 ◦ C min−1 . The as-synthesized products were allowed to react with ammonia by placing them in a desiccator (ca. 1 L), where a bottle containing 28% aqueous ammonia solution (100 ml) was placed, and were aged at room temperature for 10 or 24 h. The products were dried in air at 60 ◦ C for 1 day and calcined in air at 350 ◦ C for 10 h (2.5 ◦ C min−1 ).
Fig. 1. Schematic drawing of the experimental setup used for film deposition.
251
2.3. Characterization Scanning electron micrographs (SEM) were obtained on a Hitachi S-2380N scanning electron microscope. The nitrogen adsorption isotherms were measured at −196 ◦ C on a Belsorp TCV (BEL Japan, Inc.). Prior to the measurements, the samples were dried at 120 ◦ C under vacuum for 3 h. The film thickness was evaluated using a surface profilometer, Surfcorder SE 1700 (Kosaka Laboratory, Ltd.). X-ray diffraction (XRD) was performed on a RAD IB diffractometer (Rigaku) using monochromatic CuKα radiation, operated at 40 kV and 20 mA. Transmission electron micrographs (TEM) were obtained on a JEOL JEM-100CX transmission electron microscope. Prior to the measurement, the sample was peeled off from the substrate before the calcination. 3. Results and discussion When the reaction was conducted in the absence of the substrate, the solution became turbid after aging for ca. 10 min due to the generation of surfactant–silica spherical particles and the reaction proceeded until the TEOS was completely consumed [18]. In the present experiment, the solution became turbid after aging for ca. 10 min due to the generation of surfactant–silica spherical particles, which grew to be ca. 1.2 µm in diameter until the silica source was completely consumed (20 h in the present condition). The substrates were placed in the solution perpendicularly as shown in Fig. 1, so that the formed spherical particles were sedimented at the bottom of the bottle and only a small amount of spherical particles was deposited on the substrate. The spherical particle on the substrates was removed by washing with methanol, as evidenced by SEM observation of the substrate surface. In the present experiment, the major part of TEOS was used for the spherical particles, the particles were also collected, and the yield and composition were evaluated as reported in our separate paper [18]. The weight of the plate (30 × 30 mm) increased by 0.2 mg after the reaction and the subsequent washing and drying, indicating the deposition of silica on the substrate. These values are thought to include surfactant and silica. During the coating and drying, volatile species are thought to be vaporized and the amount of the volatile species in the films should be very small. The film was transparent, as shown in Fig. 2a. SEM observations indicated that the surfaces of the films were macroscopically smooth and there were no cracks. The thickness of the film deposited on the glass plate was 100 nm, which was determined by the surface profilometer, and the thickness was homogeneous in the films. The thickness was controlled by repeating the reaction using fresh precursor solution; the thickness of the film became 300 nm after deposition for three times. Coating was not achieved when the reaction was conducted in the absence of C16TAC, indicating the important role of the interactions, which is probably as an electrostatic interaction between the cationic group of CTA+ and the silanol group of the substrate surface. The photograph and the SEM image of the calcined film are shown in Figs. 2b and 3. During the calcination, Si–O–Si
252
N. Shimura, M. Ogawa / Journal of Colloid and Interface Science 303 (2006) 250–255
Fig. 2. Photograph of (a, d) as-deposited film and (b, c) calcined film prepared using C16TAC on (a–c) plate and (d) capillary tube with a diameter of ca. 600 µm. (c) is the photograph of the calcined film after ammonia treatment (24 h).
Fig. 3. SEM image of the film surface after calcination. A spherical particle was put on the film deposited on the plate.
linkages are thought to form between the substrate and the coating. This reaction should be substrate-dependent. Therefore, the films were fixed on the glass substrate. Thus, macroscopically smooth surface was evidenced by the SEM observation. The photograph (Fig. 2b) and the SEM image (Fig. 3) indicated that the homogeneity of the film was retained after the calcination. The film density was determined by the weight of the calcined film deposited on the plate (30×30×0.2 mm) to be 1.0 g cm−3 , suggesting that the film was porous. The nitrogen adsorption isotherm of the calcined film is shown in Fig. 4c. Since the sample weight was less than 1 mg, it is difficult to measure the nitrogen adsorption isotherms. We measured the nitrogen adsorption isotherms with an instrument (Belsorp TCV, BEL Japan, Inc.) [4]. The pore diameter [19] (Fig. 5c) and the BET surface area [20] are determined from the isotherm to be 2.4 nm and 930 m2 g−1 , respectively. It is known that surfactant:TEOS ratios in the starting solution are a factor that affects the mesostructures [21,22]. When the C16TAC:TEOS ratio was 1:1, a film with a similar thickness formed. The XRD patterns of the films prepared
Fig. 4. Nitrogen adsorption isotherms of the calcined films prepared using (a) C12TAC, (b) C14TAC, (c) C16TAC, and (d) C18TAC on the plate.
at C16TAC:TEOS ratio 0.4 showed no diffraction peaks (data not shown), suggesting a disordered mesostructure. That of the film prepared at C16TAC:TEOS of 1.0 also showed no diffraction peaks. The pore diameter [19] and BET surface area [20] of the films prepared at C16TAC:TEOS of 1.0 are 2.4 nm and 830 m2 g−1 , respectively. These values were similar to those of the calcined films prepared at C16TAC:TEOS of 0.4. Pore size was controlled by using alkyltrimethylammonium chlorides with different alkyl chain lengths. The nitrogen adsorption isotherms are shown in Fig. 4. The nitrogen adsorption isotherms of the films prepared by using C12TAC, C14TAC, and C16TAC followed type I and that of films prepared by using C18TAC followed type IV. The pore size depended on the alkyl chain length of CnTAC: 2.0 (C12TAC), 2.25 (C14TAC), and 2.8 nm (C18TAC), as shown in Fig. 5 and summarized in Table 1. These results confirmed the mesoporous silica films formed by a surfactant templated mechanism. The XRD pat-
N. Shimura, M. Ogawa / Journal of Colloid and Interface Science 303 (2006) 250–255
253
Fig. 6. TEM image of the calcined film prepared using C18TAC.
Fig. 5. BJH pore size distributions of the calcined films prepared using (a) C12TAC, (b) C14TAC, (c) C16TAC, and (d) C18TAC on the plate. Table 1 Summary of BET surface area and pore diameter of the calcined films Surfactant
Ammonia treatment period (h)
Pore diameter (nm)
BET surface area (m2 g−1 )
C12TAC C14TAC C16TAC C16TAC C16TAC C18TAC
Non Non Non 10 24 Non
2.0 2.25 2.4 2.5 2.8 2.8
690 700 930 380 360 820
terns of all the films showed no diffraction peaks, indicating disordered mesostructures. The TEM image of the calcined film prepared by using C18TAC (Fig. 6) showed a continuous silica framework with a thickness of less than 2 nm, and the presence of finite particles was not seen. These results indicated that the pores of the film were generated by a surfactant template. The TEM image showed disordered mesostructures, which agreed with the XRD results described before. Field emission SEM images
Fig. 7. Nitrogen adsorption isotherms of (a) untreated and (b, c) ammoniatreated films after calcination. The film was prepared using C16TAC. Ammonia treatment periods were (b) 10 and (c) 24 h.
of the mesoporous silica spherical particles deposited from the same solution in the present synthesis also showed a disordered mesostructure [23]. It is possible to enlarge pore size by a postsynthetic ammonia treatment. The nitrogen adsorption isotherms of the films prepared using C16TAC after the ammonia treatment and subsequent surfactant removal are shown in Fig. 7. The nitrogen adsorption isotherm of the films after the ammonia treatment for 24 h (Fig. 7c) followed type IV, though others followed type I. The pore size (2.4 nm) was expanded to 2.5 and 2.8 nm after the ammonia treatment at room temperature for 10 and
254
N. Shimura, M. Ogawa / Journal of Colloid and Interface Science 303 (2006) 250–255
Fig. 8. BJH pore size distributions of (a) untreated and (b, c) ammonia-treated films after calcination. The film was prepared using C16TAC. Ammonia treatment periods were (b) 10 and (c) 24 h.
24 h, respectively, and subsequent calcination (Fig. 8). The film maintained its morphology during the ammonia treatment, as shown by the photograph (Fig. 2c). Thus, mesoporous silica films were successfully deposited on flat glass substrate by the present method. Similarly to those prepared using acidic precursor solution, pore size was controllable as summarized in Table 1. The process seems to be applicable to depositing mesoporous silica films on substrates, even for those with complex geometry, for which the solvent evaporation process is difficult to apply. To show the possibility, glass capillary tubes were used as the substrate. SEM images of a cross-sectional view (Figs. 9b–9d) showed the formation of a silica–surfactant hybrid layer not only outside but also inside the capillary tube. The film surface was macroscopically smooth, as shown in Fig. 9a. The present method is also applicable to depositing silica– surfactant film on plastic substrates. Fig. 10 shows a scanning electron micrograph of the film deposited on the surface of poly(propylene) bottle that was used for the present reaction. The thickness was determined by SEM observation to be ca. 100 nm. The arrow in Fig. 10 indicates the air–water interface (see Fig. 1). A homogeneous film formed below the arrow; in water. The texture below the arrow was made by bending the product to show the presence of film. The interactions between the silica layer and the poly(propylene) surface should be different if compared with those between silica and glasses; consequently, cracks formed by bending as shown in Fig. 10. Spherical particles can be removed by washing. The adhesion of the formed film with substrates is worth further investigating to understand the film formation mechanism and for practical ap-
Fig. 9. SEM images of the films deposited (a–c) inside and (d) outside the capillary tube. SEM images of (b–d) are cross-sectional view. SEM image of (c) is a close-up view of (b).
N. Shimura, M. Ogawa / Journal of Colloid and Interface Science 303 (2006) 250–255
255
tion, Culture, Sports, Science and Technology (MEXT) of the Japanese Government, by CREST, by the Tokuyama Science Foundation, and by a Waseda University Grant for Special Research Projects (2005B-073, 2005B-074). References
Fig. 10. SEM image of poly(propylene) bottle (inside) after the reaction. The arrow indicates the air–water interface.
plications. To the best of our knowledge, this is the first example of silica–surfactant hybrid film deposition on plastic substrates. 4. Conclusions Transparent silica–surfactant films were deposited on a borosilicate glass plate, a soda-lime glass tube, and poly(propylene) from an aqueous solution containing tetraethoxysilane, alkyltrimethylammonium chloride, ammonia, and methanol. The calcination of the products on glasses at 350 ◦ C in air led to the formation of mesoporous films. The pore size was controllable from 2.0 to 2.8 nm using alkyltrimethylammonium chlorides with different alkyl chain lengths. The pore size was enlarged by postsynthetic ammonia treatment from 2.4 to 2.8 nm. This approach enables us to deposit a mesoporous silica layer on a variety of substrates, including those unstable under acidic conditions and organic polymer. Additionally, the films were deposited on substrates with such complex geometries as tubes and fibers. Thus, the present approach is a versatile way to prepare mesoporous silica films on various materials with various geometries. Acknowledgments This work was supported by a Grant-in-Aid for Scientific Research on Priority Areas (417) from the Ministry of Educa-
[1] M. Ogawa, J. Am. Chem. Soc. 116 (1994) 7941. [2] M. Ogawa, Chem. Commun. (1996) 1149. [3] S. Yang, P.A. Mirau, C.-S. Pai, O. Nalamasu, E. Reichmanis, J.C. Pai, Y.S. Obeng, J. Seputro, E.K. Lin, H.-J. Lee, J. Sun, D.W. Gidley, Chem. Mater. 14 (2002) 369. [4] M. Ogawa, K. Kuroda, J. Mori, Langmuir 18 (2002) 744. [5] P. Yang, G. Wirnsberger, H.C. Huang, S.R. Cordero, M.D. McGehee, B. Scott, T. Deng, G.M. Whitesides, B.F. Chmelka, S.K. Buratto, G.D. Stucky, Science 287 (2000) 465. [6] B.J. Scott, G. Wirnsberger, M.D. McGehee, B.F. Chmelka, G.D. Stucky, Adv. Mater. 13 (2001) 1231. [7] J. Loerke, F. Marlow, Adv. Mater. 14 (2002) 1745. [8] G. Wirnsberger, B.J. Scott, G.D. Stucky, Chem. Commun. (2001) 119. [9] H. Li, M. Nogami, Adv. Mater. 14 (2002) 912. [10] H. Yang, A. Kuperman, N. Coombs, S. Mamiche-Afara, G.A. Ozin, Nature 379 (1996) 703. [11] K.J. Balkus, A.S. Scott, M.E. Gimon-Kinsel, J.H. Blanco, Micropor. Mesopor. Mater. 38 (2000) 97. [12] N. Nishiyama, S. Tanaka, Y. Egashira, Y. Oku, K. Ueyama, Chem. Mater. 15 (2003) 1006. [13] S. Tanaka, T. Maruo, N. Nishiyama, Y. Egashira, K. Ueyama, Chem. Lett. 34 (2005) 1148. [14] J.E. Martin, M.T. Anderson, J. Odinek, P. Newcomer, Langmuir 13 (1997) 4133. [15] N. Nishiyama, A. Koide, Y. Egashira, K. Ueyama, Chem. Commun. (1998) 2147. [16] N. Nishiyama, D.H. Park, A. Koide, Y. Egashira, K. Ueyama, J. Membr. Sci. 182 (2001) 235. [17] M. Ogawa, N. Shimura, A. Ayral, Chem. Mater. 18 (2006) 1715. [18] N. Shimura, M. Ogawa, Bull. Chem. Soc. Jpn. 78 (2005) 1154. [19] E.P. Barrett, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951) 373. [20] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309. [21] M. Ogawa, T. Igarashi, K. Kuroda, Bull. Chem. Soc. Jpn. 70 (1997) 2833. [22] M. Ogawa, N. Masukawa, Micropor. Mesopor. Mater. 38 (2000) 35. [23] N. Shimura, M. Ogawa, submitted for publication.