Langmuir 2007, 23, 3265-3272
3265
Surface Design for Precise Control of Spatial Growth of a Mesostructured Inorganic/Organic Film on a Large-Scale Area Atsushi Hozumi,*,† Satoshi Kojima,‡ Shusaku Nagano,‡ Takahiro Seki,‡ Naoto Shirahata,§ and Tetsuya Kameyama† National Institute of AdVanced Industrial Science & Technology (AIST), 2266-98, Anagahora, Shimo-shidami, Moriyama-ku, Nagoya 463-8560, Japan, Department of Molecular Design and Engineering, Graduate School of Engineering, Nagoya UniVersity, Furo-cho, Chikusa-ku, Nagoya 464-8603, Japan, and National Institute for Materials Science (NIMS), 1-1 Sengen, Tsukuba Ibaraki 305-0047, Japan ReceiVed May 18, 2006. In Final Form: NoVember 26, 2006 A microfabrication technique is presented to fabricate a mesostructured inorganic/organic composite film, i.e., silica/cetyltrimethylammonium chloride (CTAC) film, with near-perfect site-selectivity on a large surface area based on a spatially regulated growth method. To precisely regulate the site-selective growth of this mesocomposite film at the solid/liquid interface, we designed a novel microtemplate consisting of a “dual-component” self-assembled monolayer (SAM) with alternating hydrophobic trifluorocarbon (CF3) and cationic amino (NH2) groups. First, (heptadecafluoro-1,1,2,2-tetrahydrodecyl)trimethoxysilane (FAS)-SAM was formed onto Si substrate covered with native oxide (SiO2/Si) from vapor phase. The substrate was then photolithographically micropatterned using 172 nm vacuum UV light. Finally, the micropatterned FAS-SAM was immersed in a solution of 1 vol % (aminoethylaminomethyl)phenethyltrimethoxysilane (AEAMPS) in absolute toluene. Due to these treatments, a dual-SAM microtemplate with CF3- and NH2-terminated surfaces was fabricated, as evidenced by lateral force microscopy, ellipsometry, and X-ray photoelectron spectroscopy. Using this template, the microfabrication of a mesocomposite film was demonstrated. As a control, the micropatterned hydrophobic FAS-SAM template (composed of CF3- and OH-terminated surfaces) was also treated under the same conditions. Optical microscopy and atomic force microscopy confirmed that the formation of the continuous mesocomposite film proceeded only on the FAS-SAM-covered regions, while the AEAMPSSAM-covered regions remained free of deposits. This shielding effect also remained constant regardless of the pattern’s geometry, i.e., the interval distance between the FAS-SAM-covered areas in the pattern. Through this approach, we were able to obtain well-defined 5-, 10-, and 20-µm wide mesocomposite microlines over the entire 10 × 10 mm2 area with high area-selectivity. On the other hand, when the SiO2 regions were not terminated with the cationic NH2 groups, cluster formation proceeded not only on the hydrophobic CF3 regions but also on the SiO2 regions, particularly with an increase in the pattern interval distance, resulting in lower final pattern resolution.
1. Introduction The microfabrication of electronic components with both high resolution and spatial precision has been attracting increasing attention in the microelectronics and biotechnology industries. Oxides are key materials in microelectronic components, having become widely available for insulators,1 piezoelectrics,2 ferroelectrics,3 and ferromagnetics.4 However, due to their extremely high chemical durability and physical stability against electron beams and lasers, oxides are difficult to micropattern with typical microfabrication processes employing chemical or physical etching. Consequently, their microprocessing currently depends on complicated photolithographic technology consisting of a sequence of many processes. An alternative method requiring neither etching nor additional posttreatments has been strongly demanded. One approach is to employ patterned organosilane self-assembled monolayers (SAMs). SAMs, with their wellordered structures, have been widely applied to artificially prepare * To whom correspondence should be addressed. Telephone: +81-52736-7175. Fax: +81-52-736-7182. E-mail:
[email protected]. † AIST. ‡ Nagoya University. § NIMS. (1) Kingon, A. I.; Maria, J.-P.; Streiffer, S. K. Nature (London), Phys. Sci. 2000, 406, 1032. (2) Martin, C. R.; Aksay, I. A. J. Phys. Chem. B 2003, 107, 4261. (3) Gorbenko, O. Y.; Samoilenkov, S. V.; Graboy, I. E.; Kaul, A. R. Chem. Mater. 2002, 14, 4026. (4) Cassassa, S.; Ferrari, A. M.; Busso, M.; Pisani, C. J. Phys. Chem. B 2002, 106, 12978.
organic surfaces with various chemical functionalities and can be fabricated to micro- or nanometerscale patterns by various techniques, including microcontact printing,5 ultraviolet (UV) light,6 electron beam,7 and scanning probe microscope lithographies.8 Microstructured oxide films have been successfully prepared using such patterned organosilane SAMs as masks or templates.9-17 This spatially regulated growth method using a patterned SAM is a simple low-cost microfabrication process, based on the site(5) (a) Kaumar, A.; Whitesides, G. M. Appl. Phys. Lett. 1993, 63, 2002. (b) Wilbur, J. L.; Kaumar, A.; Kim, E.; Whitesides, G. M. AdV. Mater. 1994, 6, 600. (c) Kaumar, A.; Biebuyck, H. A.; Whitesides, G. M. Langmuir 1994, 10, 1498. (d) Kaumar, A.; Whitesides, G. M. Science 1994, 263, 60. (e) Jeon, N. L.; Clem, P. G.; Payne, D. A.; Nuzzo, R. G. Langmuir 1996, 12, 5350. (f) Xia, Y. N.; Whitesides, G. M. Angew. Chem., Int. Ed. 1998, 37, 551. (6) (a) Dulcey, C. S.; Georger, J. H.; Krauthamer, J. V.; Stenger, D. A.; Fare, T. L.; Calvert, J. M. Science 1991, 252, 551. (b) Dressick, W. J.; Calvert, J. M. Jpn. J. Appl. Phys., Part 1 1993, 32, 5829. (c) Brandow, S. L.; Chen, M.-S.; Aggarwal, R.; Dulcey, C. S.; Calvert, J. M.; Dressick, W. J. Langmuir 1999, 15, 5429. (d) Sugimura, H.; Ushiyama, K.; Hozumi, A.; Takai, O. Langmuir 2000, 16, 885. (7) (a) Lercel, M. J.; Tiberio, R. C.; Chapman, P. F.; Craighead, H. G.; Sheen, C. W.; Parkin, A. N.; Allara, D. L. J. Vac. Sci. Technol., B 1993, 11, 2823. (b) Mino, N.; Ozaki, S.; Ogawa, K.; Hatada, M. Thin Solid Films 1994, 243, 374. (c) Hild, R.; David, C.; Muller, H. U.; Volkel, B.; Kayser, D. R.; Grunze, M. Langmuir 1998, 14, 342. (d) Harnett, C. K.; Satyalakshmi, K. M.; Craighead, H. G. Langmuir 2001, 17, 178. (8) (a) Kim, Y. T.; Bard, A. J. Langmuir 1992, 8, 1096. (b) Lercel, M. J.; Redinbo, G. F.; Craighead, H. G.; Sheen, C. W.; Allara, D. L. Appl. Phys. Lett. 1994, 65, 974. (c) Perkins, F. K.; Dobisz, E. A.; Brandow, S. L.; Koloski, T. S.; Calvert, J. M.; Rhee, K. W.; Kosakowski, J. E.; Marrian, C. R. K. J. Vac. Sci. Technol., B 1994, 12, 3725. (d) Kleineberg, U.; Brechling, A.; Sundermann, M.; Heinzmann, U. AdV. Funct. Mater. 2001, 11, 208.
10.1021/la061405l CCC: $37.00 © 2007 American Chemical Society Published on Web 02/13/2007
3266 Langmuir, Vol. 23, No. 6, 2007
selective deposition of a target material from either a liquid or gas phase containing a precursor, taking advantage of the difference in chemical reactivity between the SAM-covered and -uncovered regions. Using this method, a number of oxide materials, including FeOOH,9 TiO2,10 ZrO2,11 SnO2,12 ZnO,13 Fe3O4,14 SrTiO3,15 Ta2O5,16 and apatite,17 have been successfully formed into micropatterned thin films. In most cases, a patterned template consisting of chemically inert and hydrophobic SAM (i.e., methyl [CH3], trifluorocarbon [CF3-], p-tolyl [CH3-C6H4], and phenyl [C6H5-] groups) regions and hydrophilic oxide (generally SiO2) regions was employed. Although the deposition sites were markedly influenced by both solution conditions and surface functions,18 the target materials mostly deposited on the hydrophilic SiO2 regions. In contrast to such simple microtemplate systems, there have been only a few reports of successful micorpatterning of oxide materials using heterogeneous SAM surfaces, e.g., SO3H/CH39 and SCOCH3/SO3H.10a Using such dual-SAM microtemplates, FeOOH9 and TiO210a have been siteselectively deposited on hydrophilic surface areas other than SiO2 regions by taking advantage of the marked differences in regional surface charge density. Although the spatially regulated growth method using a SAM microtemplate seems to meet the demands for the fabrication of micropatterns of various target materials, it suffers from shortcomings. Pattern resolution generally becomes significantly worse with increasing reaction time,19 since the film grows not only on the predefined sites but also on undesired regions due to physisorption.12c This has been the prime cause of difficulties in fabricating well-defined micropatterns composed of the target materials. To eliminate such unfavorable deposits on the chemically inert SAM-covered regions, a lift-off process by sonication in organic solvent for a relatively long time (∼2 h) has been demonstrated.12d However, such long treatment is inconvenient for practical application and sometimes causes deterioration in the edge acuity of the final micropatterns. Looking beyond these useful oxide materials, mesoporous oxides (e.g., mesoporpous silica20), fabricated by replicating preformed self-organized supramolecular assemblies of surfactants or amphiphilic block copolymers, are the subject of growing interest as catalysts, supports, and next-generation low-k (9) Reike, P. C.; Tarasevich, B. J.; Wood, L. L.; Engelhard, M. H.; Baer, D. R.; Fryxell, G. E.; John, C. M.; Laken, D. A.; Jaehnig, M. C. Langmuir 1994, 10, 619. (10) (a) Collins, R. J.; Shin, H.; De Guire, M. R.; Heuer, A. H.; Sukenik, C. N. Appl. Phys. Lett. 1996, 69, 860. (b) Koumoto, K.; Seo, S.; Sugiyama, T.; Seo, W. S.; Dressick, W. J. Chem. Mater. 1999, 11, 2305. (c) Masuda, Y.; Seo, W. S.; Koumoto, K. langmuir 2001, 17, 4876. (d) Masuda, Y.; Sugiyama, T.; Koumoto, K. J. Mater. Chem. 2002, 12, 2643. (e) Gao, Y.-F.; Masuda, Y.; Koumoto, K. Chem. Mater. 2004, 16, 1062. (11) (a)Lee, J. P.; Sung, M. M. J. Am. Chem. Soc. 2004, 126, 28. (b) Gao, Y. F.; Masuda, Y.; Ohta, H.; Koumoto, K. Chem. Mater. 2004, 16, 2615. (12) (a) Bunker, B. C.; Pieke, P. C.; Tarasevich, B. J.; Campbell, A. A. Fryxell, G. E.; Graff, G. L.; Song, L.; Liu, J.; Virden, J. W.; McVay, G. L. Science 1994, 264, 48. (b) Supothina, S.; De Guire, M. R. Thin Solid Films 2000, 371, 1. (c) Shirahata, N.; Masuda, Y.; Yonezawa, T.; Koumoto, K. Langmuir 2002, 18, 10379. (d) Shirahata, N.; Shin, W.; Murayama, N.; Hozumi, A.; Yokogawa,Y.; Kameyama, T.; Masuda, Y.; Koumoto, K. AdV. Funct. Mater. 2004, 14, 580. (e) Shirahata, N.; Hozumi, A. Chem. Mater. 2005, 17, 20. (13) Saito, N.; Ohashi, N.; Haneda, H. Koumoto, K. AdV. Mater. 2002, 14, 418. (14) Nakanishi, T.; Masuda, Y.; Koumoto, K. Chem. Mater. 2004, 16, 3484. (15) Gao, Y.; Masuda, Y.; Yonezawa, T.; Koumoto, K. Chem. Mater. 2002, 14, 5006. (16) Clem, P. G.; Jeon, N.-L.; Nuzzo, R. G.; Payne, D. A. J. Am. Ceram.Soc. 1997, 80, 2821. (17) Zhu, P. X.; Masuda, Y.; Koumoto, K. J. Colloid Interface Sci. 2001, 243, 31. (18) Gao, Y.; Koumoto, K. Cryst. Growth Des. 2005, 5, 1983. (19) Pizem, H.; Sukenik, C. N.; Sampathkumaran, U.; McIlwain, A. K.; De Guire, M. R. Chem. Mater. 2002, 14, 2476. (20) (a) Yanagisawa, T.; Shimizu, T.; Kuroda, K.; Kato, C. Bull. Chem. Soc. Jpn. 1990, 63, 988. (b) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Vartuli, J. C.; Beck, J. S. Nature 1992, 359, 710.
Hozumi et al.
dielectrics for integrated circuits due to their large pore size, high internal surface area, and thermal stability. In addition, their precursor materials (e.g., silica/surfactant mesostructured composite), which organic templates do not eliminate, have also been studied extensively and have been applied to optical materials by doping various components, including semiconducting nanocrystals21 and polymers,22 and dye molecules.23 Such mesoporous oxides and their precursor mesostructured materials have also been fabricated into micropatterned thin films by micromolding,23c-f,24 micropen lithography,25 ink-jet printing,25 dip-coating,25 and site-selective UV irradiation26 for advanced applications, such as in sensors, actuators, nanoreactors,and fluid or optical devices. One of the authors reported previously that silica/surfactant mesocomposites were formed into microstructured thin films based on the spatially regulated growth method,27b,c first demonstrated by Yang et al.,27a using a template made up of micropatterned hydrophobic SAM/hydrophilic oxides. In this case, deposition proceeded preferentially on the hydrophobic SAM regions due to the hydrophobic interaction between the surfactant molecules and the SAM regions, resulting in the formation of a continuous composite film.27 However, with this template, the undesirable deposition of disk-shaped or ropelike composites was frequently observed on the hydrophilic oxide regions. In addition, site-selectivity on this type of template varies greatly depending on the pattern’s geometry, i.e., the pattern interval distance, and the surface properties of the hydrophilic oxide regions.27c To overcome these shortcomings seen with spatially regulated growth methods using patterned SAMs and to attain clear-cut area-selectivity, particularly over a large area, a suitable template design requiring no additional posttreatment such as etching or lift-off is required. In this paper, we report a novel strategy for the microfabrication of mesostructured composite film with high site-selectivity on a large surface area based on a modified spatially regulated growth method. In selecting among the various surface/interface designs available, we chose to particularly focus on the spatial control of both hydrophobic and electrostatic interactions between the surfactant molecules and the organic headgroups terminating the solid surface, since these two interactions play crucial roles in directing the supramolecular structure of the organized surfactant molecules at the solid/liquid interface.27c,28 We thus (21) Dag, O ¨ .; Ozin, G. A.; Yang, H.; Reber, C.; Bussie`re, G. AdV. Mater. 1999, 11, 474. (22) Nguyen, T. Q.; Wu, J. J.; Doan, V.; Schwartz, B. J.; Tolbert, S. H. Science 2000, 288, 652. (23) (a) Fowler, C. E.; Lebeau, B.; Mann, S. Chem. Commun. (Cambridge) 1998, 1825. (b) Marlow, F.; McGehee, M. D.; Zhao, D.; Chmelka, B. F.; Stucky, G. D. AdV. Mater. 1999, 11, 632. (c) Yang, P.; Wirnsberger, G.; Huang, H. C.; Cordero, S. R.; McGehee, M. D.; Scotto, B.; Deng, T.; Whitesides, G. M.; Chmelka, B. F.; Buratto, S. K.; Stucky, G. D. Science 2000, 287, 465. (d) Scott, B. J.; Wirnsberger, G.; Stucky, G. D. Chem. Mater. 2001, 13, 3140. (e) Scott, B. J.; Wirnsberger, G.; McGehee, M. D.; Chmelka, B. F.; Stucky, G. D. AdV. Mater. 2001, 13, 1231. (f) Wirnsberger, G.; Yang, P.; Huang, H. C.; Scott, B.; Deng, T.; Whitesides, G. M.; Chmelka, B. F.; Stucky, G. D. J. Phys. Chem. B 2001, 105, 6307. (24) (a) Trau, M.; Yao, N.; Kim, E.; Xia, Y.; Whitesides, G. M.; Aksay, I. A. Nature 1997, 390, 674. (b) Huo, Q.; Zhao, D.; Feng, J.; Weston, K.; Buratto, S. K.; Stucky, G. D.; Schacht, S.; Schu¨th, F. AdV. Mater. 1997, 9, 974. (c) Yang, P.; Deng, T.; Zhao, D.; Feng, P.; Pine, D.; Chmelka, B. F.; Whitesides, G. M.; Stucky, G. D. Science 1998, 282, 2244. (d) Yang, P.; Rizvi, A. H.; Messer, B.; Chmelka, B. F.; Whitesides, G. M.; Stucky, G. D. AdV. Mater. 2001, 13, 427. (25) (a) Fan, H.; Lu, Y.; Stump, A.; Reed, S. T.; Baer, T.; Schunk, R.; PerezLuna, V.; Lo´pez, G. P.; Brinker, C. J. Nature 2000, 405, 56. (b) Fan, H.; Reed, S. T.; Baer, T.; Schunk, R.; Lo´pez, G. P.; Brinker, C. J. Microporous Mesoporous Mater. 2001, 44-45, 625. (26) (a) Doshi, D. A.; Huesing, N. K.; Lu, M.; Fan, H.; Lu, Y.; SimmonsPotter, K.; Potter, B. G., Jr.; Hurd, A. J.; Brinker, C. J. Science 2000, 290, 10. (b) Dattelbaum, A. M.; Amweg, M. L.; Ecke, L. E.; Yee, C. K.; Shreve, A. P.; Parikh, A. N. Nano Lett. 2003, 3, 719. (27) (a) Yang, H.; Coombs, N.; Ozin, G. A. AdV. Mater. 1997, 9, 811. (b) Hozumi, A.; Sugimura, H.; Hiraku, K.; Kameyama, T.; Takai, O. Nano Lett. 2001, 1, 395.(c) Sugimura, H.; Hozumi, A.; Kameyama, T.; Takai, O. AdV. Mater. 2001, 13, 667.
Inorganic/Organic Film on a Large-Scale Area
Langmuir, Vol. 23, No. 6, 2007 3267
Scheme 1. Fabrication Process for Microlines of Silica/Surfactant Mesostructured Film: (a) Formation of a FAS-SAM on SiO2/Si Substrate from Vapor Phase; (b) Vacuum UV Irradiation through a Photomask for 30 min at 10 Pa; (c) Formation of an AEAMPS-SAM on the Micropatterned FAS-SAM Surface from a Liquid Phase; (d) Coplanar Microstructures Comprising the FAS/AEAMPS Dual-SAM; (e) Deposition of Silica/Surfactant Mesostructured Film onto Sample d; and (f) Schematic Illustration of an Actual Sample ([A] Patterned and [B] Masked Regions)
2. Experimental Section
prepared a heterogeneous organic microtemplate system consisting of a dual-component SAM with alternating hydrophobic CF3 and cationic amino (NH2) groups through a two-step surface modification, i.e., the local elimination of the CF3-terminated SAM by vacuum ultraviolet (UV) lithography followed by the site-selective chemisorption of aminosilane molecules. By employing these patterned heterogeneous organic sites with different chemical reactivities to cationic surfactant molecules, we attempted to precisely regulate the spatial growth of the mesocomposite film. Although, as mentioned above, there has been extensive research on the fabrication of micropatterns from target materials using patterned SAMs as templates, most of the experiments have focused particularly on the masking effect of the chemically inert SAM. On the other hand, there have been few reports of microfabrication techniques using patterned dualSAMs. In particular, the simultaneous spatial control of two distinct interactions between the precursor molecules and the different regions of the dual-SAM surface has not yet been demonstrated. Such a technique is critical to artificially mimic crystal growth in vivo, i.e., the biomineralization occurring in biological systems, since this growth is governed by particular molecular recognition mechanisms at solid/liquid interfaces. Furthermore, such heterogeneous SAM surfaces are well-suited as basic model surfaces for studying the mechanisms of biomineralization, which generally occurs on non-homogeneous surfaces. We compared the site-selectivity of the composites fabricated on this dual-SAM system with that of composites fabricated on a conventional microtemplate composed of hydrophobic CF3-terminated/hydrophilic SiO2 regions. The areaselectivity in the growth of the composite film was further investigated in terms of pattern geometry by varying the interval distance between the FAS-SAM-covered regions in the template pattern.
2.1. Microtemplate Preparation. A dual-component SAM microtemplate was formed according to the process illustrated in Scheme 1a-d. As illustrated in Scheme 1a, a fluoroalkylsilane (FAS)-SAM was first formed on UV/ozone-cleaned Si (100) substrates (n-type) covered with native oxide (SiO2/Si) by chemical vapor deposition (CVD) for 3 h at 150 °C using (heptadecafluoro-1,1,2,2-tetrahydrodecyl)trimethoxysilane (CF3[CF2]7CH2CH2Si[OCH3]3) as a precursor (Scheme 1a).29 The thickness of the FAS-SAM was estimated by ellipsometry (Philips, PZ2000) to be 1.1 ( 0.1 nm. The FAS-SAM-covered SiO2/Si surface became hydrophobic with a water-contact angle of about 117°. The FAS-SAM was then micropatterned by vacuum UV light lithography.6d Each sample was exposed to vacuum UV light generated from an excimer lamp (Ushio Inc., UER20172V; λ ) 172 nm and 10 mW/cm2) through a photomask in contact with its surface for 30 min at a reduced pressure of 10 Pa (Scheme 1b). This photomask featured lines arranged in a 10 × 10 mm2 area with a width of either 5, 10, or 20 µm and a spacing of 5, 10, or 20 µm, respectively. A 10 mm thick quartz glass plate (Asahi Glass, Synthetic silica glass AQX for Xe2 172 nm excimer lamp) served as a weight on top of the photomask so as to obtain complete contact between the mask and the sample surface. The transparency of the photomask and the quartz plate at 172 nm was about 93 and 90%, respectively. The total light intensity at the FAS-SAM surface was estimated to be about 8.4 mW/ cm2. The dose was about 15.1 J/cm2. Finally, this micropatterned substrate was immersed in a solution of 1 vol % (aminoethylaminomethyl)phenethyltrimethoxysilane (AEAMPS; H2NCH2CH2NHCH2C6H4CH2CH2Si[OCH3]3) in absolute toluene for 5 min at room temperature (Scheme 1c). After this, each sample was sonicated in acetone for 20 min, rinsed with Milli-Q water, and then treated for 10 min at 150 °C. 2.2. Site-Selective Deposition of a Silica/Surfactant Nanocomposite. A silica/surfactant mesocomposite film was syn-
(28) (a) Miyata, H.; Kuroda, K. Chem. Mater. 1999, 11, 1609. (b) Miyata, H.; Kuroda, K. AdV. Mater. 1999, 11, 1448. (c) Miyata, H.; Kuroda, K. Chem. Mater. 2000, 12, 49.
(29) Hozumi, A.; Ushiyama, K.; Sugimura, H.; Takai, O. Langmuir 1999, 15, 7600. (30) Brunner, H.; Vallant, T.; Mayer, U.; Hoffman, H. Langmuir 1996, 12, 4614.
3268 Langmuir, Vol. 23, No. 6, 2007
thesized onto these coplanar microstructures comprising the FAS/ AEAMPS dual-SAM (Scheme 1d) in a solution prepared as follows. As a control, a micropatterned FAS-SAM/SiO2 surface was similarly examined. A mixture of cetyltrimethylammonium chloride (CTAC, Tokyo Kasei Organic Chemicals), hydrochloric acid (HCl), and water (H2O) was stirred for about 20 min until the CTAC powder completely dissolved. Tetraethoxysilane (TEOS, Si[OC2H5]4; Tokyo Kasei) was then added to the solution and mixed while stirring for about 3 min at room temperature. The final reactant molar ratio of the TEOS/CTAC/HCl/H2O solution was 0.1:0.11:7:100. The substrate was immersed upside down in this solution and kept undisturbed for 30 min at 25 °C (Scheme 1e). After immersion, the substrate was washed with Milli-Q water and blown dry with N2 gas (Scheme 1f). 2.3. Characterization. The static water-contact angles were measured on the sample surfaces using a contact angle goniometer (Kyowa Interface Science, CA-X). The microlines fabricated on the FAS-SAM/SiO2 surface were observed at room temperature in ambient air by an atomic force microscope (AFM; Seiko Instruments, Inc., SPA-400+SPI-3800N) in lateral force microscopic (LFM) mode using a Si probe (Seiko, cantilever; force constant ) 0.12 N/m) covered with a native oxide layer. Additionally, surface topographies of the deposits were acquired by the same AFM system using another Si probe (Seiko, cantilever; force constant ) 15 N/m) with a response frequency of 1 kHz. Changes in chemical properties of the sample surface before and after SAM formation were examined by X-ray photoelectron spectroscopy (XPS; JEOL, Ltd., JPS-90MX) using Mg KR (E ) 1253.6 eV) radiation. The binding energy (BE) scale was calibrated to provide Au4f7/2 ) 83.9 eV and Cu2p3/2 ) 932.8 eV. The X-ray source was operated at 15 mA and 8 kV with the analyzer’s constant pass energy at 20 eV. The pressure in the analysis chamber was about 8 × 10-8 Pa during measurements. The core-level signals were obtained at a photoelectron takeoff angle of 90° (with respect to the sample surface). The BE scales were referenced to 285.0 eV as determined by the locations of the maximum peaks on the C1s spectra of hydrocarbon (CHx), associated with adventitious contamination. The accuracy of the BE determined with respect to this standard value was within (0.3 eV. The deposits were characterized by X-ray diffractometry using Cu KR radiation with λ ) 0.154 178 nm (XRD, MAC Science MXP3) and by micro-infrared transmission spectroscopy (IR, JASCO MICRO-20). The ζ-potentials of the SAM-covered SiO2/Si substrates were determined using an electrophoretic light scattering spectrophotometer (ELS600, Otsuka Electronics). In this analysis, a quartz cell was used for measuring the electrophoretic mobility of polystyrene latex reference particles with an average diameter of 520 nm. The cell was filled with an appropriate amount of a solution of a certain pH containing the reference particles. The solution contained 1 mM KCl as a supporting electrolyte. Its was adjusted over the range from pH 3 to 11 by adding HCl or NaOH. An electric field of 80 kV was supplied between two Pt electrodes, one mounted at each end of the cell. The ζ-potential was estimated from the average of 10 value measurements. ζ-potential error was about (5 mV.
3. Results and Discussion 3.1. Formation of the Dual-Component SAM. To fabricate a dual-component SAM, the FAS-SAM-covered SiO2/Si (FASSAM/SiO2/Si) substrate was first photolithographically micropatterned. Figure 1 shows a typical LFM image of the photopatterned FAS-SAM in which bright and dark regions correspond to photoirradiated and unirradiated regions, respec-
Hozumi et al.
Figure 1. Typical LFM image of 10 µm wide FAS-SAM microlines formed on SiO2/Si substrate through step b in Scheme 1.
tively. As can be clearly seen, the 10 µm wide microlines photoprinted on the FAS-SAM surface over the entire 10 × 10 mm2 area (area A in Scheme 1f) were clearly imaged through the difference in friction between the photoirradiated and unirradiated regions. It has been reported that active oxygen species generated during vacuum UV irradiation degraded and eliminated an FAS-SAM.6d Accordingly, in the vacuum UVirradiated regions of the sample in Figure 1, the surface beneath the FAS-SAM, which consisted of a 0.2-0.27 nm thick SiO2 monolayer,27 became exposed, while in the unirradiated regions the FAS-SAM remained intact. This SiO2 monolayer-covered SiO2/Si surface was highly hydrophilic with a water-contact angle of 5° or less. It therefore adhered more strongly to the Si probe surface, which was also covered with a hydrophilic native SiO2 layer. The friction force contrast on the vacuum UV-irradiated regions was thus greater than that on the FAS-SAM-covered areas. In addition, the former regions became once again reactive to other organosilane molecules, while the latter remained unreactive due to their termination with the CF3 groups of the FAS-SAM. Next, by taking advantage of this chemical difference, AEAMPS molecules were adsorbed on the vacuum UV-irradiated regions through a liquid phase. To investigate the chemical changes which occurred on both the vacuum UV-irradiated and unirradiated regions due to this liquid-phase treatment, the surface chemical properties of regions A and B shown in Scheme 1f (i.e., the vacuum UV-irradiated and unirradiated regions, respectively) were studied by XPS. In this study, we focused particularly on N concentration, since this element is not initially present in the FAS-SAM/SiO2/Si substrate. No N1s signal was detectable in the masked region B in Scheme 1f, indicating that, due to sonication in acetone, any physisorbed AEAMPS molecules had been completely eliminated from region B, i.e., the FAS-SAM surface. On the other hand, an N1s signal was detected in the patterned region A, i.e., the micropatterned surface, although its concentration was very low. This clearly demonstrates that the AEAMPS molecules had indisputably adsorbed only onto the vacuum UV-irradiated surface. However, since it was difficult to selectively characterize the actual surface properties of the individual SiO2 microlines before and after AEAMPS treatment, as a model surface we observed an FAS-SAM/SiO2/Si substrate after vacuum UV irradiation without micropatterning (i.e., a SiO2 monolayer-covered SiO2/Si substrate). This sample was treated with AEAMPS under the same conditions and similarly examined. The thickness and water-contact angle of the AEAMPS layer were 0.9 ( 0.1 nm and about 51°, respectively, which are mostly consistent with those reported for an AEAMPS-SAM-covered
Inorganic/Organic Film on a Large-Scale Area
Langmuir, Vol. 23, No. 6, 2007 3269
Figure 3. Typical XRD trace of a silica/surfactant mesostructured film. Figure 2. Optical micrographs (a and c) and AFM images (b and d) of microfabricated silica/surfactant mesostructured films on two different substrates. Panels a and b show a mesostructured film deposited on a micropatterned FAS-SAM surface, while c and d show the same deposited on a micropatterned FAS/AEAMPS-SAM surface. Panel e shows a cross-sectional view of the AFM image in d.
SiO2/Si substrate.31 The AEAMPS molecules had fixed areaselectivity onto the photoirradiated regions (SiO2 regions) so as to form an AEAMPS-SAM. We thus concluded that, through the present process, a microtemplate consisting of a dual-SAM with alternating CF3 and NH2 functionalities was prepared. 3.2. Site-Selective Growth. Typical optical micrographs of deposits grown on both the micropatterned FAS-SAM/SiO2/Si and the coplanar two-component SAM surface (from a 500 × 600 µm2 region on area A in Scheme 1f) are shown in Figure 2a,c, respectively. As can be seen in Figure 2a, deposits showing continuous filmlike features grew preferentially on the FASSAM-covered regions, while on the SiO2 regions the surface was incompletely covered by deposits with disklike features. Such disk-shaped structures have been typically seen in mesostructured composites grown in the same acidic solution.32 On the contrary, with the dual-SAM template no undesirable deposition was seen on the NH2-terminated surface regions, as demonstrated in Figure 2c. We confirmed by micro-IR that the micropatterned films consisted of CTAC molecules and silica matrix. As can be seen in Figure 3, a typical XRD pattern of the deposited film demonstrates that it had well-ordered periodic mesostructures. One intense (100) peak and a weak (200) peak are observed. The absence of (110) and (210) reflections in this XRD pattern indicates that the channel axis of the hexagonal unit cell runs parallel to the substrate surface, as has been demonstrated in several reports on mesostructured silica films.27a,33-35 The d100 spacing was in the range of 3.2-3.6 nm, which is diagnostic of mesostructured composite. Similar to the results shown in Figure 3, periodic mesostructures were detected by XRD for the disk-shaped deposits on the SiO2 surface (31) Dressick, W. J.; Dulcey, C. S.; Chen, M.-S.; Calvert, J. M. Thin Solid Films 1996, 284-285, 568. (32) Yang, H.; Coombs, N.; Ozin, G. A. Science 1997, 386, 692.
(spectrum not shown). Both sample surfaces were further observed by AFM. As clearly seen in a typical AFM image in Figure 2d, well-defined microlines of ca. 10 µm in width and ca. 280 nm in thickness grew site-selectively on the FAS-SAM regions, while the AEAMPS-SAM regions were completely protected from undesirable adsorption of the disk-shaped deposits seen in Figure 2a,b. Furthermore, as can be seen in Figure 2e, the edge regions of the patterned films were sharp. For precise evaluation of the pattern resolution, we estimated the variation in the nominal pattern width based on a previously reported method.10b By employing a two-dimensional AFM image of the microlines shown in Figure 2d (image not shown), we measured the pattern width at 100 equally spaced points on each microline in the image. The average width of each microline in the image was 11.14 µm. The pattern edge roughness, as measured by the standard variation of the line width, was 0.48 µm. Thus, the variation in the nominal pattern width was calculated to be 4.3% (i.e., 0.48/11.14 × 100), which meets the specifications demanded for current electronics design.10b Such well-shaped mesocomposite microlines could be employed in a further step to fabricate mesoporous silica microlines: through “photocalcination” using the same UV light source, it is expected that the mesostructured composite micropatterns could be readily converted photochemically to mesoporous silica micropatterns while preserving their microstructures.27b Highly ordered nanopores could be fabricated in a similar manner.36 Next, we investigated the relation between pattern geometry and site-selectivity. These results are summarized in Figure 4. The nucleation density on the 20 µm wide hydrophilic SiO2 regions seen in Figure 4a was much larger than that on the 10 and 5 µm wide SiO2 regions in Figure 4b,c, respectively. In addition, close inspection of this micrograph shows that there is (33) Yang, H.; Kuperman, A.; Coombs, N.; Mamiche-Afara, S.; Ozin, G. A. Nature 1996, 379, 703. (34) (a) Aksay, I. A.; Trau, M.; Manne, S.; Honma, I.; Yao, N.; Zhou, L.; Fenter, P.; Eisenberger, P. M.; Gruner, S. M. Science 1996, 273, 892. (b) Yang, H.; Coombs, N.; Sokolov, I.; Ozin, G. A. J. Mater. Chem. 1997, 7, 1285. (35) Sneh, O.; George, S. M. J. Phys. Chem. 1995, 99, 4639. (36) (a) Hozumi, A.; Yokogawa, Y.; Kameyama, T.; Hiraku, K.; Sugimura, H.; Takai, O. AdV. Mater. 2000, 12, 985. (b) Hozumi, A.; Sugimura, H.; Hiraku, K.; Kameyama, T.; Takai, O. Chem. Mater. 2000, 12, 3842.
3270 Langmuir, Vol. 23, No. 6, 2007
Hozumi et al.
Figure 4. Optical micrographs of microfabricated silica/surfactant mesostructured films on two different substrates. Panels a-c show mesostructured films deposited on micropatterned FAS-SAM surfaces with pattern intervals of 20, 10, and 5 µm, respectively. Panels d-f show mesostructured films deposited on micropatterned FAS/AEAMPS-SAM surfaces with pattern intervals of 20, 10, and 5 µm, respectively. Scheme 2. Schematic Illustration of Site-Selective Deposition of Silica/Surfactant Mesocomposite on a Micropatterned FAS/ AEAMPS-SAM Surface: (a) Adsorption of CTAC Molecules on the FAS-SAM Regions, (b) Formation of Hemicylindrical Micelle Arrays, (c) Cylindrical Micelle Rod Formation and Silica Polymerization through the Condensation of Silicate Anions, (d) CTAC Molecules on the AEAMPS-SAM Regions, and (e) Area-Selectively Formed Mesostructured Silica Film
almost no nucleation on the SiO2 regions near the boundaries with the surrounding CF3-terminated regions. This suggests that growth of the composite on the SiO2 regions was suppressed by the adjacent hydrophobic SAM surface. Judging from this optical micrograph, the critical dimension for this growth suppression was estimated to be less than 2-3 µm. This is supported by our optical micrograph shown in Figure 4c. The nucleation on the 5 µm wide SiO2 regions in this image was the least among the samples. Based on these results, just as has been found with conventional FAS-SAM/SiO2 microtemplates, the pattern resolution on the micropatterned FAS-SAM/SiO2/Si surface was concluded to be defined by the pattern geometry, i.e., the interval distance between the different regions in the pattern. On the contrary, as can be seen in Figure 4d-f, our new dual-SAM template completely suppressed unfavorable deposition on the
AEAMPS-SAM regions even when the interval distance increased up to 20 µm. Our present results clearly demonstrate that the spatial growth of the mesocomposite film on the patterned dualSAM template was less dependent on surface geometry. By means of our technique presented here, we successfully fabricated wellordered continuous microlines of composite films. This high area-selective growth was similarly observed not only on the small region shown in Figures 4d-f but also over the entire area A in Scheme 1f. 3.3. Mechanism of Site-Selective Growth. This significant difference between the samples in the area-selective growth of the mesostructured composite was further investigated in terms of the chemical properties of the solid surface, including its wettability (hydrophobicity/hydrophilicity) and charge density. The mechanism for the area-selective deposition of the meso-
Inorganic/Organic Film on a Large-Scale Area
Figure 5. Comparison of ζ-potentials as a function of pH for a FAS-SAM-covered SiO2/Si substrate (open triangles), an AEAMPSSAM-covered SiO2/Si substrate (solid circles), and a bare SiO2/Si substrate (open circles).
composites on the dual-SAM surface is illustrated in Scheme 2. It has been reported that continuous mesocomposite films have been formed on hydrophobic surfaces such as polymer,28 graphite,34 and hydrophobic SAM-coated gold or oxides.27 On such hydrophobic surfaces, the surfactant molecules interact primarily through hydrophobic and van der Waals interactions.37 In the present case, due to these attractive interactions, tailgroups of the CTAC molecules interacted with the hydrophobic FASSAM surface (Scheme 2a). Accordingly, hemicylindrical aggregates, at the bottom of which the surfactant tailgroups adsorbed horizontally to the substrate, are considered to have formed (Scheme 2b).34,37 Subsequently, cylindrical micelle rods are thought to have assembled onto these surfactant aggregates so as to be aligned with the hemicylindrical micelle arrays (Scheme 2c).34 Indeed, the channels of the resulting mesocomposite film ran predominantly parallel to the substrate, as shown in Figure 3. It has also been reported that a mesostructured silica composite grown on a hydrophobic alkanethiol SAM surface was oriented parallel to the SAM surface.27a Due to the presence of the TEOS molecules, film formation was initiated by silica polymerization through the condensation of charge-balancing silicate anions around the headgroup regions of such micelle aggregates (Scheme 2c).34b,38 Subsequent deposition and polymerization of silicate/ surfactant micellar precursor species resulted in the growth of the continuous mesostructured composite film (Scheme 2e).34b,38 The surface of the AEAMPS-SAM (Scheme 2d) was charged positively in the pH region below 1, as evidenced by the ζ-potential measurements (indicated by the solid circles) shown in Figure 5. Therefore, due to electrostatic interaction, cationic CTAC molecules or clusters were probably repelled from this positively charged surface, resulting in a decrease in the adsorbate density.39 In addition to this charge effect, since the AEAMPS-SAM surface was less hydrophobic than the FAS-SAM surface, formation of the hemicylindrical micelles could hardly proceed.28a Furthermore, since silanol (SiOH) groups on the SiO2 surface were consumed due to the formation of siloxane bondings with the AEAMPS molecules composing the SAM, and such groups are considered to provide binding sites for the hydrophilic parts of (37) (a) Manne, S.; Cleveland, J. P.; Gaub, H. E.; Stucky, G. D.; Hansma, P. K. Langmuir 1994, 10, 4409. (b) Manne, S.; Gaub, H. E. Science 1995, 270, 1480. (38) Huo, Q.; Margolese, D. I.; Ciesla, U.; Demuth, D. G.; Feng, P.; Gier, T. E.; Sieger P.; Firouzi, A.; Chmelka, B. F.; Schu¨th, F.; Stucky, G. D. Chem. Mater. 1994, 6, 1176. (39) Wa¨ngnerud, P.; Olofsson, G. J. Colloid Interface Sci. 1992, 153, 392.
Langmuir, Vol. 23, No. 6, 2007 3271
the surfactants,34a,40 there would have been very few anchoring sites on the AEAMPS-SAM regions. For these reasons, as demonstrated in Figures 2c,d and 4d-f, the suppression of undesirable composite growth on the AEAMPS-SAM regions was greatly enhanced and was less dependent on the pattern interval distance. On the other hand, the SiO2 surface was markedly distinct from both organic SAM-covered surfaces in that it was hydrophilic and contained reactive SiOH groups. Judging from Figure 5, the SiO2 surface, as indicated by the open circles, is thought to have been less positively charged than the AEAMPS-SAM-covered surface. Thus, adsorbate density of the CTAC molecules on the SiO2 surface is considered to have been larger than that on the AEAMPS-SAM surface.37 It has also been reported that the SiOH groups on a hydrophilic SiO2 surface interact primarily with surfactant headgroups, resulting in the formation of roughly spherical micelles.37 These would serve as nucleation sites, leading finally to the disk-shaped deposits containing the CTAC molecules on the SiO2 regions, as shown in Figure 4a-c.
4. Conclusion We have demonstrated the large-scale microfabrication of mesostructured silica/surfactant composite films with high (
