Orientation of mesochannels in continuous mesoporous silica films by a high magnetic field

Abstract

The effect of a high magnetic field on the orientation of mesochannels in continuous mesoporous silica films is demonstrated; the orientation of mesochannels in the film can be induced parallel to the magnetic field, though the effect is not complete.

---

Mesoporous materials prepared through self-assembly of surfactants and inorganic species have attracted enthusiastic interest as one of the key materials in nanotechnology since the first report on mesoporous silica from a layered polysilicate kanemite in 1990.¹ Ordered mesoporous materials with various compositions and morphologies including spheres, rods, fibers, and films have been reported and such control is vital for practical applications. The macroscopic alignment of mesochannels in films and its control are quite significant for advanced nanomaterials with controlled functions.² Many techniques for the alignment have been reported, including the use of anisotropic surfaces,³ photo-induced polymer orientation,⁴ shear flows,⁵etc. In particular, the use of a rubbing-treated polyimide film⁶ and silicon (110) surface⁷ is quite effective for uniaxial alignment of mesochannels in mesoporous silica films.

However, the direction of mesochannels in almost all mesoporous/mesostructured films is parallel to the substrates. If the direction of mesochannels is perpendicular to a substrate, we can expect huge potential applications including high-sensitive chemical sensors, highly selective separations, ultra-high-density recording media, etc. There have been several reports on the perpendicular alignment of mesochannels.^8–14 Although the use of a combination process of eutectic decomposition of a SiO₂–Fe₂O₃ system and chemical etching can produce a unique mesoporous silica film,⁸ the TEM images of the obtained mesochannels are unclear. Nanoporous films with cylindrical pores perpendicular to the substrate have also been fabricated by using the nanophase separation of a eutectic Al–Si system.⁹ It is difficult to create uniform mesopores by using such a eutectic system, and these systems can not be extended to general synthetic methodology, if compared with surfactant-templated systems. The use of an electric field¹⁰ is another way, but the orientation, arrangement, and pore size of mesochannels in the films have not been evidenced. The use of ternary surfactant systems is useful to create perpendicular orientation, though the system is presently limited to the formation of flake-like materials.¹¹ Perpendicularly arranged mesochannels were prepared inside the columnar pores of anodic porous alumina.¹² However, this approach can not be used to obtain continuous films with homogeneous compositions. Research on anodic porous alumina with shorter scale periodicity is also currently being carried out by controlling the anodization conditions or applying a retexturing process using a mold.¹³ A recent report shows that much more diverse orientations of mesochannels are created in anodic porous alumina, depending on the pore size.¹⁴

One of the most convenient and versatile methods for the orientation is the application of magnetic field. By utilizing the magnetic transition of rod-like macromolecules formed by self-assembly of surfactants, a mesoporous silica monolith with the alignment of mesochannels over a macroscopic length scale under a magnetic field has been reported.¹⁵ We extended this method for the preparation of mesoporous silica films. Although Ogura et al. have very recently reported the partial alignment of mesochannels perpendicular to a substrate by using a magnetic field in a conference abstract,¹⁶ the structures have not been characterized in detail, and direct evidence on the perpendicular alignment in the film has not been presented.

Here we demonstrate the mesochannel alignment in continuous mesoporous silica films, induced by the application of a high magnetic field. The spatial orientation of magnetically aligned mesochannels was elucidated by conventional θ–2θ scanning, in-plane, and two dimensional (2D) X-ray diffractions (XRD), cross-sectional TEM, and HR-SEM.

In this study, continuous SBA-15 type films with 2D-hexagonal structure, possessing a large periodic porous structure, were prepared on glass substrates according to the procedure reported previously.¹⁷ A precursor solution (TEOS: surfactant (P123) : H₂O : HCl : C₂H₅OH = 0.158 : 16.6 : 100 : 0.018 : 145) was cast onto a glass substrate, and then dried at 25 °C for 1 day without the magnetic field (Film A) and under an applied magnetic field of 12 Tesla (T) parallel (Film B) and perpendicular (Film C) to the substrate surface (Fig. 1). The thicknesses of all the films were ca. 8 µm by casting the same amount of the precursor solution onto the substrates.

Fig. 1 Schematic presentation for the preparation of continuous mesoporous silica films under an applied high magnetic field.

The conventional θ–2θ scanning XRD profiles of the as-prepared films prepared without the magnetic field (Film A) and under the magnetic field parallel to the substrate (Film B) show two peaks (Film A; d = 8.0, 4.1 nm, Film B; d = 9.1, 4.6 nm) corresponding to (100) and (200) diffractions of the 2D-hexagonal structure (Fig. 2I-1). Moreover, the in-plane orientation of the mesochannels in Film B calcined at 400 °C for 4 h was evaluated by ϕ–2θχ scanning profiles (Fig. 2I-2). The in-plane diffraction profiles show an intense peak (2θχ = 0.84°, d = 11.74 nm). This peak has already been discussed in the in-plane X-ray diffraction studies on mesoporous silica films.^6,7 To estimate the alignment distribution of the mesochannels, the scan of the ϕ rotation was measured, fixing the detector position at the diffraction peak maximum. The incident X-rays at ϕ = ±90° were parallel to the magnetic field, and then two peaks were clearly observed at every 180° (Fig. 2I-2, inset). On the other hand, in the case of calcined Film A without the magnetic field, the alignments of mesochannels were not confirmed at all. From these results, we confirmed that the 2D-hexagonal mesochannels can be induced parallel to the magnetic field in the film. The distribution of the alignment direction in calcined Film B was estimated to be about 40° from the full-width-at-half-maximum of the ϕ scanning profiles. This distribution is much wider than those observed for films prepared on a rubbing coating substrate⁶ and a Si (110) substrate,⁷ and is similar to that observed for a film prepared through a photo-induced approach.⁴ The reason for the wider distribution of the alignment of mesochannels will be described later.

Fig. 2 I-1) XRD profiles (θ–2θ scanning) of the as-prepared films. a) Film A as-prepared without magnetic field, b) Film B as-prepared under the magnetic field parallel to the substrate, c) Film C as-prepared under the magnetic field perpendicular to the substrate. I-2) In-plane XRD measurements of calcined Film B prepared under the parallel magnetic field to the substrate. The magnetic field is set parallel (A) and perpendicular (B) to the incident X-rays at ϕ = 0°. Inset: Angular dependence of maximum diffraction intensity at 2θχ = 0.84° as a function of incident angle of probing X-rays (ϕ). The incident X-rays at ϕ = ±90° are parallel to the magnetic field. II) TEM images of as-prepared Film C under the magnetic field perpendicular to the substrate. II-1) TEM images recorded along the [110] and [001] zone axes of a hexagonal (p6mm) mesostructure, II-2 and II-3) the cross-sectional TEM image of the film. III) HR-SEM images of calcined Film C. III-1) The entire morphology of the film, III-2) the interface image between the top-surface and the cross-section.

The mechanism of the alignment of lyotropic liquid crystals (LLC) formed through the self-assembly of surfactants under a high magnetic field is basically explained by the interactions between the field and the anisotropic diamagnetic susceptibility Δχ^m = Δχ^m_|| − Δχ^m_⊥, where Δχ^m_|| and Δχ^m_⊥ are anisotropic diamagnetic susceptibilities along the directions parallel and perpendicular to the molecular axis, respectively.¹⁸ Because the magnetic interaction energy related to Δχ^m of an individual surfactant (EO₂₀–PO₇₀–EO₂₀) is very small (less than thermal energy in a 12 T field at room temperature), an individual surfactant can not be induced by a magnetic force. However, when surfactants form lyotropic liquid crystals including additive species like HCl and Si sources, the additive magnetic interaction energy given by a domain of lyotropic liquid crystals can overcome the thermal disordering energy. Therefore, LLC have been used to create macroscopic alignments, with the result that the rod-like macromolecules formed through self-assembly of surfactants align along the direction of the magnetic field.¹⁵ In this system, during the magnetic transition of LLC, polymerization of silica species occurs simultaneously because of the acidic conditions in the precursor solution, hindering the orientation of the mesochannels. In other words, as the silicate condensation proceeds by evaporation of ethanol, the viscosity of the precursor solution becomes higher, and then the magnetic transition of rod-like macromolecules (LLC) could be decelerated. Careful control of both the polymerization of silicate species and the viscosity of the solution are necessary for smooth magnetic transition.

The as-prepared Film C synthesized under the perpendicular magnetic field showed only a small and broad XRD peak (Fig. 2-I). Considering that conventional θ–2θ scanning profiles provide only structural information parallel to a substrate, profile c) suggests that the orientation of the mesochannels in the film is induced along the direction perpendicular to the substrate by applying a high magnetic field.¹⁹ In fact, the intensity of the first peak assigned to (100) in the θ–2θ scanning XRD profiles was progressively decreased with the applied magnetic field (not shown). The interactions among the rod-like macromolecules in LLC and the magnetic field should affect the preferred orientation, though the details remain to be identified.

The TEM images of the powdery samples of as-prepared Film C detached from the substrate show a typical hexagonal (p6mm) mesostructure (Fig. 2II-1). The cross-sectional TEM image shows that, in some domains, the mesochannels are aligned almost perpendicularly to the substrate (Fig. 2II-2). The repeat distance between the mesochannels was ca. 10 nm. We observed the mesochannels running perpendicularly to the substrate in all the images in this domain when the sample holder was tilted over an angular range of ±15°, indicating the formation of mesochannels aligned perpendicularly in the domain. However, all mesochannels formed over the substrate did not align perpendicularly. In other domains, mesochannels are inclined at about 45–60° to the substrate (Fig. 2II-3). The orientation of mesochannels might be hindered by simultaneous polymerization of silica species during the magnetic transition of LLC. Although we attempted to measure the as-prepared and calcined films by in-plane XRD, we could not obtain evidence for the perpendicular alignment. We consider that this in-plane result is caused by the fact that Film C has no single crystalline mesostructure and has not only perpendicular mesochannels but also mesochannels inclined to some extent to the substrate, as is confirmed from the TEM images.

For further structural characterization, the cross-sectional images of calcined Film C were directly observed by HR-SEM (Fig. 2III). The continuous and smooth surface morphology was observed in the low magnification image (Fig. 2III-1). Stripes with a 10 nm interval along the direction perpendicular to the substrate were observed (Fig. 2III-2). Compared to the cross-sectional TEM images of as-prepared Film C, the repeat distance did not change during calcination. This is probably due to the major shrinkage of the silica network in the direction normal to the substrate in the film, which is currently under investigation. These top-surface and cross-sectional TEM and HR-SEM images are direct evidence of the magnetically induced alignment of mesochannels in the film.

We also applied a rapid solvent evaporation method by using a spin coating under a high magnetic field in order to prepare crack-free thin films for 2D-XRD measurements. A precursor solution was spin-coated onto a glass substrate without the magnetic field (Film D) and under a magnetic field of 12 T (Film E) perpendicular to the substrate surface. The thicknesses of two films were ca. 1.5 µm. By this method, the solvent was rapidly evaporated, and the viscosity of the solution on the glass substrate increased immediately. Compared to the casting method described above, these may not be suitable conditions for alignment of mesochannels, because this method does not afford gelation times long enough to align the mesochannels. Although films with higher perpendicular alignments could not be obtained by this method, compared to those prepared by a casting method, the intensity of the first peak in the low angle range in the conventional θ–2θ profile was greatly decreased by applying the magnetic field (not shown).

The 2D-XRD pattern of Film D as-prepared without the magnetic field shows a 6-fold hexagonal spot, and the presence of the intense (100) spot normal to the film plane indicates that the mesochannels are oriented parallel to the substrate surface (Fig. 3I-1). On the other hand, the 2D-XRD pattern of Film E as-prepared under a perpendicular magnetic field was completely different from that of Film D. If the mesochannels are completely aligned perpendicular to the substrate, the main spot assigned to (100) in Fig. 3I-1 should disappear. But, from this 2D-XRD pattern, we cannot identify the formation of a 2D-hexagonal mesostructure with orientations, and the arc of the main spot may be derived from the formation of a mesostructure with a lower ordering.²⁰ However, in the HR-SEM image of the uneven part produced by cracking formed during the calcination at 400 °C for 4 h, the co-existence of both the tubular mesochannels and the honeycomb-like arrangements of mesopores shows the formation of highly ordered 2D-hexagonal structure in the film (see ESI† Fig. A). Therefore, the presence of the arc in the main spot and the slight deviation of the locations of several faint extra spots observed near the main spot suggest that the orientation of the 2D-hexagonally ordered mesochannels is affected by the high magnetic field. It should be noted that the HR-SEM image shows that pores less than 10 nm in size are indeed formed over the entire area of the top-surface in Film E, meaning that almost all the mesochannels in the film are aligned not parallel to the substrate under the perpendicular magnetic field (Fig. 3II). Therefore, it is proved that the application of a high magnetic field is quite effective even in the rapid evaporation method.

Fig. 3 I) 2D-XRD patterns measured under grazing incidence conditions: I-1) Film D as-prepared without magnetic field, I-2) Film E as-prepared under the magnetic field perpendicular to the substrate. II) HR-SEM image of the top-surface of calcined Film E.

In conclusion, we have demonstrated the effect of high magnetic field for the orientation of mesochannels through detailed characterization by XRD, TEM, and HR-SEM for the first time. This approach, utilizing the magnetic transition of LLC containing inorganic species, can be introduced for systems of not only silica but also many inorganic oxides including tin oxide and titania, and it is a conventional and effective methodology for the preparation of continuous mesoporous films with mesochannels aligned in a controlled manner. By further improvements, this process should lead to the production of mesoporous films with mesochannels aligned completely perpendicular to a substrate and with controlled uniform pore sizes and wall thicknesses.

The authors greatly appreciate the reviewers' helpful comments. Advice from Dr. H. Miyata (Canon Inc.) is also greatly appreciated. The authors acknowledge Prof. A. Iida (Photon Factory, KEK) for helpful discussion. The authors also acknowledge Messrs. M. Fuziwara, K. Takahashi (Kagami Memorial Lab., Waseda Univ.) and T. Suzuki (Waseda Univ.) for cross-sectional TEM and in-plane X-ray diffraction measurements. This work is supported in part by a Grant-in-Aid for Center of Excellence (COE) Research “Molecular Nano-Engineering”, the 21st Century COE Program “Practical Nano-Chemistry”, and Encouraging Development Strategic Research Centers Program “Establishment of Consolidated Research Institute for Advanced Science and Medical Care” from the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japanese government.

Notes and references

1. T. Yanagisawa, T. Shimizu, K. Kuroda and C. Kato, Bull. Chem. Soc. Jpn., 1990, 63, 988.
2. A. Fukuoka, H. Miyata and K. Kuroda, Chem. Commun., 2003, 284; W. C. Molenkamp, M. Watanabe, H. Miyata and S. H. Tolbert, J. Am. Chem. Soc., 2004, 126, 4476.
3. H. Yang, A. Kuperman, N. Coombs, S. Mamiche-Afara and G. A. Ozin, Nature, 1996, 379, 703; I. A. Aksay, M. Trau, S. Manne, I. Honma, N. Yao, L. Zhou, P. Fenter, P. M. Eisen-berger and S. M. Gruner, Science, 1996, 273, 892.
4. Y. Kawashima, M. Nakagawa, T. Seki and K. Ichimura, Chem. Mater., 2002, 14, 2842; Y. Kawashima, M. Nakagawa, T. Seki and K. Ichimura, J. Mater. Chem., 2004, 14, 328.
5. H. W. Hillhouse, T. Okubo, J. W. van Egmond and M. Tsapatsis, Chem. Mater., 1997, 9, 1505; M. Trau, N. Yao, E. Kim, Y. Xia, G. M. Whitesides and I. A. Aksay, Nature, 1997, 390, 674.
6. H. Miyata and K. Kuroda, Chem. Mater., 1999, 11, 1609; H. Miyata and K. Kuroda, Chem. Mater., 2000, 12, 49; H. Miyata, T. Noma, M. Watanabe and K. Kuroda, Chem. Mater., 2002, 14, 766.
7. H. Miyata and K. Kuroda, J. Am. Chem. Soc., 1999, 121, 7618.
8. S. Kondoh and Y. Iwamoto, J. Am. Ceram. Soc., 1999, 82, 209.
9. K. Fukutani, K. Tanji, T. Motoi and T. Den, Adv. Mater., 2004, 16, 1456.
10. K. Kuraoka, Y. Tanaka, M. Yamashita and T. Yazawa, Chem. Commun., 2004, 1198.
11. B. C. Chen, H. P. Lin, M. C. Chao, C. Y. Mou and C. Y. Tang, Adv. Mater., 2004, 16, 1657.
12. A. Yamaguchi, F. Uejo, T. Yoda, T. Uchida, Y. Tanamura, T. Yamashita and N. Teramae, Nature Mater., 2004, 3, 337; Q. Li, F. Gao, S. Komarneni and T. E. Mallouk, J. Am. Chem. Soc., 2004, 126, 8650.
13. H. Masuda, K. Takenaka, T. Owaki, M. Nagae and K. Nishio, Self-ordering in Cell Configuration in Anodic Porous Alumina, presented at 70th Meeting of the Electrochemical Society of Japan, Tokyo, Japan, April 1–3, 2003; H. Masuda, H. Yamada, M. Satoh, H. Asoh, M. Nakao and T. Tamamura, Appl. Phys. Lett., 1997, 71, 2770; H. Masuda, M. Yotsuya, M. Asano, K. Nishio, M. Nakao, A. Yokoo and T. Tamamura, Appl. Phys. Lett., 2001, 78, 826; H. Masuda, F. Matsumoto and K. Nishio, Electrochemistry, 2004, 72, 389.
14. Y. Wu, G. Cheng, K. Katsov, S. W. Sides, J. Wang, J. Tang, G. H. Fredrickson, M. Moskovits and G. D. Stucky, Nature Mater., 2004, 3, 816.
15. A. Firouzi, F. Atef, A. G. Oertli, G. D. Stucky and B. F. Chmelka, J. Am. Chem. Soc., 1997, 119, 3596; A. Firouzi, D. J. Schaefer, S. H. Tolbert, G. D. Stucky and B. F. Chmelka, J. Am. Chem. Soc., 1997, 119, 9466; S. H. Tolbert, A. Firouzi, G. D. Stucky and B. F. Chmelka, Science, 1997, 278, 264; S. Ozeki, K. Iguchi, Y. Kojima, K. Nakamoto, Y. Imai and K. Tajima, RIKEN Rev., 2002, 153.
16. M. Ogura, S. Iwai, Y. Zhang, J. Otomo, H. Takahashi, N. Hirota and T. Okubo, Toward the preparation of aligned mesopores perpendicular to the substrate using magnetic field, presented at 4th International Mesostructured Materials Symposium (IMMS), Cape Town, South Africa, May 1–4, 2004, Book of abstracts p.386.
17. P. C. A. Alberius, K. L. Frindell, R. C. Hayward, E. J. Kramer, G. D. Stucky and B. F. Chmelka, Chem. Mater., 2002, 14, 3284.
18. A. Yamagishi, T. Takeuchi, T. Higashi and M. Date, Physica B, 1990, 164, 222.
19. H. W. Hillhouse, J. W. vanEgmond, M. Tsapatsis, J. C. Hanson and J. Z. Larese, Microporous Mesoporous Mater., 2001, 44, 639.
20. D. A. Doshi, A. Gibaud, V. Goletto, M. Lu, H. Gerung, B. Ocko, S. M. Han and C. J. Brinker, J. Am. Chem. Soc., 2003, 125, 11646.

---

Footnote

† Electronic supplementary information (ESI) available: extra data for the films prepared under a high magnetic field and characterizations. See http://www.rsc.org/suppdata/jm/b4/b418478e/

---