Fabrication of Self-Assembled Monolayers (SAMs ... - ACS Publications

3278

Langmuir 2004, 20, 3278-3283

Fabrication of Self-Assembled Monolayers (SAMs) and Inorganic Micropattern on Flexible Polymer Substrate Junhui Xiang, Peixin Zhu, Yoshitake Masuda, and Kunihito Koumoto* Department of Applied Chemistry, Graduate School of Engineering, Nagoya University, Nagoya 464-8603, Japan Received November 6, 2003. In Final Form: January 9, 2004 By grafting (aminopropyl)triethoxysilane (APTES) as the buffer layer on poly(ethylene terephthalate) (PET) surface, the SAMs of octadecyltrichlorosilane (OTS), phenyltrichlorosilane (PTCS), vinyltrichlorosilane (VTCS), and p-tolyltrichlorosilane (TTCS) were fabricated on the flexible polymer substrate. The properties of SAMs were accurately controlled by adjusting the immersing time of substrates in the solutions and the concentration of the solutions. The SAMs acted as templates, and TiO2 micropattern was successfully deposited on OTS, TTCS, and PTCS SAMs.

1. Introduction Nanodevices, especially those integrated on flexible substrates, are expected to be applied in many fields, such as very large scale integration (VLSI) systems, integrated optic systems, automatic systems, robots, micro air vehicles (MAV), and communication systems, as they offer many advantages, including suitability for any complex shape, minimum size, low power, and high performance. Despite these advantages, nanodevices still await practical application. One of the key problems is how to fabricate and assemble the functional materials into nanoscale devices. Other problems include how to improve the reliability of nanosystems and how to reduce the cost.1,2 In recent years, there have been many reports about fabricating nanodevices on rigid substrates such as Si wafer and glass by lithography, sputtering, CVD, etc.3-6 However, these techniques are not fit for flexible substrates such as polymer. To date, there have been no reports about fabricating nanodevices on polymer substrates, which is mainly due to the difficulties in fabricating technology. The self-assembly process is a potential approach to the fabrication of nanodevices. This process makes it possible for nanomaterials to be synthesized and assembled into nanodevices spontaneously onto the polymer substrates. The location of the fabricated devices can be accurately controlled. This process can be applied to both rigid and flexible substrates, and the cost is very low. In the self-assembly process, self-assembled monolayers (SAMs) are fabricated on polymer substrates and modified as templates. Then a micropattern of inorganic functional films can be formed according to the designed shape of the templates. * To whom correspondence should be addressed. Phone: +81-52-789-3329. Fax: +81-52-789-3201. E-mail: g44233a@ nucc.cc.nagoya-u.ac.jp (1) Balzani, V.; Credi, A.; Raymo, F. M.; Stoddart, J. F. Angew. Chem., Int. Ed. 2000, 39, 3348-3391. (2) Borman, S. Chem. Eng. News 2001, 79, 45-55. (3) Kim, G. M.; van den Boogaart, M. A. F.; Brugger, J. Microelectron. Eng. 2003, 67/68, 609-614. (4) Kretz, J.; Dreeskornfeld, L.; Hartwich, J.; Rosner, W. Microelectron. Eng. 2003, 67/68, 763-768. (5) Cayssol, F.; Ravelosona, D.; Wunderlich, J.; Chappert, C.; Mathet, V.; Jamet, J.-P.; Ferre, J. J. Magn. Magn. Mater. 2002, 240, 30-33. (6) Obraztsov, A. N.; Volcov, A. P.; Zakhidov, Al. A.; Lyashenko, D. A.; Petrushenko, Yu. V.; Satanovskaya, O. P. Appl. Surf. Sci. 2003, 215, 214-221.

The fabrication of SAMs is one of the most critical steps in this process. SAMs can be fabricated using trichlorosilanes, alkanethiols, fatty acids, etc. Among the different types of SAMs, trichlorosilane-derived monolayers are the most promising candidates for modifying the surface properties of substrates and are widely used.7-10 However, the reactive functional groups contained in trichlorosilanes make it difficult to form high-quality monolayers.11 To obtain high-quality SAMs from trichlorosilanes, it is necessary to accurately control the operating parameters, such as the amount of water in the solvent, reaction temperature, and the immersing time of substrates in solution.11-18 The most disputed parameter is the immersing time of substrates in the trichlorosilane solution, with suggested times varying from a few minutes to several hours and even more than 24 h.11,17,19 The time largely depends on the reaction conditions, the surface properties of the substrates, the type of trichlorosilanes, etc. In this research, SAMs derived from trichlorosilanes with different functional groups were fabricated on PET substrates to act as templates for forming a micropattern of inorganic films. To ensure the reproducibility of the monolayers, pretreatment of the polymer surfaces, immersing time of substrates in trichlorosilane solutions, and the concentration of the solution were investigated in detail. The contact angle was employed for evaluating the SAMs fabrication since it is the key criterion of the templates. TiO2 film was deposited on these SAMs with different functional groups. (7) Wasserman, S. R.; Tao, Y. T.; Whitesides, J. M. Langmuir 1989, 5, 1074-1087. (8) Du, Y. J.; Brash, J. L. J. Appl. Polym. Sci. 2003, 90, 594-607. (9) Allaraand, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 45-52. (10) Allaraand, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 52-66. (11) Ulman, A. Chem. Rev. 1996, 96, 1533-1554. (12) Le Grange, J. D.; Markham, J. L.; Kurjian, C. R. Langmuir 1993, 9, 1749-1753. (13) Brandriss, S.; Margel, S. Langmuir 1993, 9, 1232-1240. (14) Tripp, C. P.; Hair, M. L. Langmuir 1992, 8, 1120-1126. (15) Angst, D. L.; Simmons, G. W. Langmuir 1991, 7, 2236-2242. (16) McGoven, M. E.; Kallury, K. M. R.; Thompson, M. Langmuir 1994, 10, 3607-3614. (17) Silberzan, P.; Leger, L.; Ausserre, D.; Benattar, J. J. Langmuir 1991, 7, 1647-1651. (18) Gao, W.; Reven, L. Langmuir 1995, 11, 1860-1863. (19) Banga, R.; Yarwood, J.; Morgan, A. M.; Evans, B.; Kells, J. Langmuir 1995, 11, 4393-4399.

10.1021/la036088m CCC: $27.50 © 2004 American Chemical Society Published on Web 03/12/2004

Fabrication of SAMs on Flexible Polymer Substrate

2. Experimental Section

Langmuir, Vol. 20, No. 8, 2004 3279 Table 1. XPS Atomic Ratio (%) of PET Surfaces

2.1. Fabrication of Buffer Layer on PET Surface.20 APTES ((aminopropyl)triethoxysilane) was dissolved in acetone (1 vol %) and aged for 240 h. PET (poly(ethylene terephthalate)) substrates were ultrasonically cleaned for 5 min in distilled water, ethanol, and acetone and then dried at 120 °C for 5 min. The dried substrates were immersed in the acetone solution of APTES for 5 min and baked at 150 °C for 10 min to form a buffer layer. 2.2. Fabrication of SAMs on Polymer Substrates. Trichlorosilanes octadecyltrichlorosilane (OTS), phenyltrichlorosilane (PTCS), vinyltrichlorosilane (VTCS), and p-tolyltrichlorosilane (TTCS) were dissolved in anhydrous toluene with different concentrations. The substrates grafted by buffer layers were immersed in these anhydrous toluene solutions under a nitrogen atmosphere for different periods of time and then rinsed with anhydrous toluene and baked at 120 °C for 5 min to remove residual solvent and promote the chemisorption of SAMs. 2.3. Formation of TiO2 Micropattern on SAMs. The PET substrates fabricated with SAMs were irradiated with UV light in air through a photomask. The UV-irradiated parts became hydrophilic due to the formation of Si-OH groups, while the nonirradiated parts remained hydrophobic. The aqueous solutions of (NH4)2TiF6 (0.05 M) and H3BO3 (0.15 M) were mixed together, and the pH was adjusted to 2.88 by HCl. The modified PET substrates were immersed in the mixture and kept at 50 °C for a certain period of time to allow TiO2 to be deposited. After deposition, the substrates were rinsed in distilled water by sonication to remove the randomly adsorbed and deposited TiO2 particles on the surface. Then the TiO2 micropattern appeared on the substrates. 2.4. Characterization. The crystalline phase of deposited TiO2 film was identified by XRD (model RU200, Rigaku Co., Ltd. Japan, Cu KR at 40 kV, 30 mA). The microstructure was observed by SEM (S-3000N, Hitachi Ltd., Japan). The buffer layer was identified by XPS (ESCALAB 210, VG Scientific Ltd.; 1-3 × 10-7 Pa, X-ray source: Mg KR, 1253.6 eV operated at 15 kV and 18 mA). The contact angle was detected by a FACE CA-D contactangle meter (Kyowa Interface Science Co., Ltd.; medium, distilled water). The TiO2 micropattern was observed by optical microscopy (BX51WI, OLYMPUS Optical Co., Ltd.). UV light was obtained from a low-pressure mercury lamp (NL-UV253, Nippon Laser & Electronics Lab., 184.9 nm).

3. Results and Discussion The fabrication of an inorganic micropattern on polymer substrates by the self-assembly process consisted of three stages: fabricating the buffer layer, forming the SAMs, and depositing the micropattern. The buffer layer was fabricated to obtain a hydrophilic surface, and then SAMs could be formed. The formed SAMs were modified into templates, which hybrid with hydrophilic and hydrophobic surfaces, and the inorganic films could be site-selectively deposited by the difference in wettability of SAMs to create a micropattern. The wettability played the most important role during the whole process. Therefore, contact angle was used to evaluate the surface properties at each stage. 3.1. Fabrication of Buffer Layer on PET Surface. In general, trichlorosilane SAMs cannot be directly formed on a blank polymer surface. A buffer layer containing Si-O bonds is necessary to act as a bridge between the polymer substrate and the SAMs.21 APTES was selected to fabricate the buffer layer since it can form a polymeric silane structure by the formation of Si-O-Si bond on PET substrate.20,22,23 Figure 1a shows the schematic procedure of the buffer layer fabrication.17,22-24 The PET substrate (20) Zhu, P. X.; Teranishi, M.; Xiang, J. H.; Masuda, Y.; Seo, W. S.; Koumoto, K. Thin Solid Films, submitted for publication. (21) Chaudhury, M. K. Biosens. Bioelectron. 1995, 10, 785-788. (22) Kallury, K. M. R.; Krull, U. J.; Thompson, M. Anal. Chem. 1988, 60, 169-172. (23) Bui, L. N.; Thompson, M. Anal. Chem. 1993, 118, 463-474. (24) Sagiv, J. J. Am. Chem. Soc. 1980, 102, 92-98.

C 1s O 1s Si 2p N 1s

buffered PET

buffered PET, UV irradiated

26.85 46.23 12.11 14.79

22.51 56.77 13.38 7.33

Table 2. Effect of Buffer Layer on Substrates contact angle (°) different stage

buffered

nonbuffered

blank substrate (UV irradiated) OTS SAM fabricated modified OTS SAM (by UV irradiation)

4.5 110 5

40 76 23

was dip-coated with acetone solution of APTES. In the acetone solution, the -H in -NH2 groups of APTES were partially substituted by -CH3, as well as the -EtOH by -OH.20 While heating to 150 °C, Si-O-Si bonds were formed and a polymerized silica-like layer was anchored on the surface.20 With UV irradiation, the terminal groups -(CH2CH2CH2)NH2 (where -H in -NH2 was partially substituted by -CH3), abbreviated as -R, were destroyed by high-energy photons and substituted by -OH groups. Thus, the substrate surface was hydrophilic and ready for SAM fabrication. Figure 1b shows the schematic procedure of SAM fabrication. The driving force of trichlorosilane SAM was in-situ formation of polysiloxane, which was connected to surface silanol groups (-SiOH) via Si-O-Si bonds.11 While the hydroxylated substrate immersed in the solution of trichlorosilanes, the trichlorosilane molecules were hydrogen bonded on the surface. With water elimination, they were chemically anchored on the surface to form a monolayer. The procedure of the buffer layer fabrication was identified by XPS. As shown in Figure 2, in the spectra of the blank PET substrate, only the characteristic peaks of carbon and oxygen could be observed. After APTES grafted to the PET surface, the characteristic peaks of silicon and nitrogen appeared, indicating the formation of the buffer layer. UV irradiation depressed the characteristic peak of nitrogen. According to the XPS atomic ratio data shown in Table 1, the ratio of nitrogen to silicon decreased from 1.22 to 0.55. This decrement was due to the fact that the terminal groups -R, containing nitrogen element, were substituted by -OH groups. The surface properties were identified by contact angle. As shown in Figure 3, the contact angle of the blank PET surface was reduced from 75° to 40° by UV irradiation for 5 min or longer. This reduction could be due to the partial destruction of the surface molecules of PET by the highenergy photons and the partial appearance of -OH groups on the surface. However, this surface was not hydrophilic enough for SAM fabrication. With APTES grafted on the surface, the contact angle of the PET surface was 45°. It was reduced to 4.5° by UV irradiation for 5 min. This variation resulted from the fact that the terminal groups -R were substituted by -OH groups. Thus, the surface was hydroxylated. For comparison, OTS SAM was fabricated on both buffered and nonbuffered PET substrates. The properties of these two substrates are listed in Table 2. On the buffered substrate, the contact angle of SAM was 110°, while on the nonbuffered substrate it was 76°. It is because the SAM on buffered substrate is more homogeneous than that on nonbuffered substrate. TiO2 films were deposited after the SAMs were modified (Figure 4). On the buffered substrate a uniform TiO2 film was obtained. However, on the nonbuffered substrate some blank regions were

3280

Langmuir, Vol. 20, No. 8, 2004

Xiang et al.

Figure 1. Schematic procedure of the buffer layer and trichlorosilane SAMs fabrication.

distributed in the TiO2 film. This difference suggests that the buffer layer is efficient for improving the homogeneity of SAMs and the consequent deposition of inorganic films. 3.2. Fabrication of SAMs. With the buffer layer grafted, SAMs of trichlorosilane derivatives can be fabricated on PET substrates. In order to obtain a large contrast in the wettability between the hydrophilic/ hydrophobic surfaces after modification, the SAMs are expected to be hydrophobic having as large a contact angle as possible when fabricated. The contact angle is mainly affected by the terminal functional groups and the coverage of SAMs on substrates.11 For comparison, trichlorosilanes with different terminal functional groups such as OTS, PTCS, VTCS, and TTCS were selected for SAM fabrication. Their terminal functional groups are long methylene chain, phenyl, unsaturated vinyl, and tolyl (Figure 5). The coverage of SAMs on substrates could be controlled by varying the immersing time in trichlorosilane solutions and the concentration of the solutions. To obtain the optimized immersing time, the substrates were immersed in toluene solution of OTS, PTCS, VTCS, and TTCS monomers for different times. The experimental results (Figure 6) showed that the substrate surfaces were changed from hydrophilic to hydrophobic within 5 min. As the immersing time was further prolonged, the contact angle of OTS SAM remained almost unchanged while those of the three others increased gradually. From 5 min to 1 h, the increments in the contact angle of PTCS, TTCS, and VTCS SAMs were about 15°, 15°, and 30°, respectively.

The different behavior of OTS SAM from the three others is attributed to its long chain groups -(CH2)17CH3 in the OTS molecules. While OTS SAM was fabricated, the long chains form an array dotted on the surface of the monolayer. The interchain distance is about 4.4 Å, and the chain length is about 22 Å.11,17 This array causes a steric effect which obstructs the adsorption of more molecules on the substrate. Therefore, the reaction attained equilibrium instantly. As a result, the coverage and contact angle of OTS SAM could not be further improved with the immersing time being prolonged. Consequently, pinholes were formed in the monolayer.25 On the contrary, the three other SAMs do not have long chain groups in the molecules causing steric effect. The fabrication reactions take about 1 h to attain equilibrium, and the coverage and contact angles are gradually increased during the period. Besides immersing time, the concentration of trichlorosilane solutions was also checked. SAMs were fabricated in trichlorosilane solutions with different concentrations ranging from 1 to 5 vol %. Figure 7 shows the effect of concentration on the contact angle of SAMs. It was found that VTCS SAM obtained from 5 vol % solution was 20° higher in contact angle than that of obtained from 1 vol % solution. On the contrary, the contact angles of OTS, PTCS, and TTCS SAMs were little affected by the variation of concentration and were nearly the same at different concentrations. (25) Nakagawa, T.; Ogawa, K. Langmuir 1994, 10, 525-529.


Langmuir, Vol. 20, No. 8, 2004 3281 Table 3. Highest Contact Angle of SAMs Fabricated with Optimized Parameters

Figure 2. XPS spectra of (a) blank PET surface, (b) APTESgrafted PET surface, and (c) UV-irradiated PET/APTES surface.

Figure 3. Effect of buffer layer on polymer surface.

This difference between VTCS SAM and the three others is due to the polymerization of unsaturated -CHdCH2 groups in VTCS molecules. The unsaturated group is more reactive and more easily modified or polymerized than the saturated groups or benzene rings.17 In a high concentration solution the polymerization was promoted. Therefore, the coverage and contact angle of VTCS SAM were largely improved. According to the above results, SAMs were fabricated by adjusting the immersing time and the concentration of the trichlorosilane solution. Table 3 lists the reaction parameters for the highest contact angle of different SAMs. OTS SAM was almost not affected by the variations of immersing time and concentration of solution. It was

SAMs

immerzing time (min)

concentration (vol %)

highest contact angle (°)

OTS PTCS TTCS VTCS

5 60 60 60

2 1 2 5

110 93 120 158

considered that due to the steric effect of the long chain array, the reaction of OTS reaches equilibrium too quickly to be affected by these variations and results in a relatively low coverage and pinholes in OTS SAM. PTCS and TTCS SAMs exhibited similar behaviors to the variation of immersing time as well as not being affected by variation of the solution concentration. This similarity is decided by their similar molecule structures. However, it was noticed that the contact angle of TTCS was nearly 30° higher than that of PTCS. This was presumed to be due to the effect of the terminal methyl group in the TTCS molecule, which is the only difference between TTCS and PTCS molecules. VTCS SAM was sensitive to variations of both immersing time and concentration of solution. Its unsaturated groups are very reactive and can be easily modified or polymerized, and the reaction can be largely affected by the reaction time and the concentration of the solution. Besides, the unsaturated groups are more apolar and more hydrophobic than saturated ones or benzene rings.17 Therefore, the contact angle of VTCS SAM can be much higher than that of other SAMs. It had a contact angle of 158° after 1 h immersion in 5 vol % solution. However, when the immersing time was more than 30 min or the concentration was more than 2 vol %, VTCS SAM gradually became a cloudy film, due to the generation of polymeric siloxanes.7,17 It was noticed that the cloudy film was too thick to combine firmly with the substrate and could be easily peeled off by sonication. 3.3. Formation of TiO2 Micropattern on SAMs. To act as templates, the fabricated SAMs were modified by UV irradiation through a photomask. On the irradiated surface, terminal functional groups of SAMs were destroyed by the high-energy photons and replaced by -OH groups.26-30 As a result, the contact angle was reduced to less than 5°. On the contrary, the unirradiated surface was kept unchanged. Thus, the SAMs were hybrid with highly hydrophilic and hydrophobic surfaces to fit for the site-selective deposition and micropattern formation of inorganic films. The modified SAMs were submerged in the aqueous solution to form TiO2 micropattern by the liquid-phase deposition (LPD) process. The reaction can be expressed as follows27

TiF62- + 2H2O ) TiO2 + 4H+ + 6FBO33- + 4F- + 6H+ f BF4- + 3H2O TiO2 was generated from aqueous molecules by the electrolysis of Ti-containing molecules. On the hydrophilic (26) Koumoto, K.; Seo, S.; Sugiyama, T.; Seo, W. S. Chem. Mater. 1999, 11, 2305-2309. (27) Masuda, Y.; Sugiyama, T.; Koumoto, K. J. Mater. Chem. 2002, 12, 2643-2647. (28) Masuda, Y.; Seo, W. S.; Koumoto, K. Langmuir 2001, 17, 48764880. (29) Masuda, Y.; Jinbo, Y.; Yonezawa, T.; Koumoto, K. Chem. Mater. 2002, 14, 1236-1241. (30) Masuda, Y.; Sugiyama, T.; Lin, H.; Seo, W. S.; Koumoto, K. Thin Solid Films 2001, 382, 153-157.

3282

Langmuir, Vol. 20, No. 8, 2004

Xiang et al.

Figure 4. SEM micrograph of TiO2 film deposited on (a) buffered PET substrate and (b) nonbuffered PET substrate (3000×).

Figure 5. Molecular structures of (a) OTS, (b) PTCS, (c) VTCS, and (d) TTCS.

Figure 6. Effect of immersing time on SAMs fabrication (concentration ) 1 vol %).

surface of modified SAMs, TiO2 was nucleated and combined with SiOH to form Ti-O-Si bonds and resulted in films. TiO2 films showed strong adhesion to the substrates and did not peel off when subjected to sonication. On the contrary, the hydrophobic surface has a repulsion force to aqueous molecules, which suppresses the nucleation of TiO2. Accordingly, TiO2 film was siteselectively deposited on the hydrophilic surface and a micropattern was obtained. The TiO2 micropattern was observed by an optical microscope (Figure 8). TiO2 films deposited on hydrophilic parts appear as white contrast in the photograph. It was noticed that on the OTS template, the micropattern was blurred and difficult to observe whereas it could be distinctly observed on the PTCS and TTCS templates. This difference could be due to the different properties of SAMs. As mentioned above, the coverage of OTS SAM was lower than that of PTCS and TTCS SAMs owing to the steric effect. Pinholes with diameters ranging from

Figure 7. Effect of concentration on SAMs fabrication (immersing time ) 5 min).

several nanometers to 100 nm were distributed in OTS SAM.25 At the pinhole positions, the SiOH of the buffer layer was exposed and tended to form Ti-O-Si bonds with TiO2. Limited by the diameter of the pinholes, these TiO2 particles were randomly deposited on the hydrophobic surface. These contaminants largely weakened the contrast between the hydrophilic surface and hydrophobic surface. Accordingly, the micropattern on the OTS template was not as distinct as that on PTCS and TTCS SAMs. Different from the OTS, PTCS, and TTCS templates, there was no micropattern observed on the VTCS template. As mentioned above, VTCS SAM was a little thick and did not combine to the substrate firmly enough. With TiO2 deposited, VTCS SAM was more easily peeled off by sonication, leaving just a blank substrate. The different appearances of micropatterns on the SAMs showed that the different properties of SAMs have an important effect on the formation of micropatterns. Table 4 lists the effects of different SAMs properties on TiO2 micropatterns. According to the properties of the four types of SAMs, PTCS and TTCS are the best candidates for templates to obtain a high-quality micropattern. 4. Conclusion In this research, PET was selected as the substrate and APTES was grafted onto the surface to form a buffer layer. SAMs of OTS, PTCS, VTCS, and TTCS were successfully fabricated on the flexible PET substrates. The OTS SAM was not sensitive to the immersing time and concentration, whereas the PTCS and TTCS SAMs were greatly affected by the immersing time. Their maximum contact angles were 110°, 93°, and 120°, respectively. The VTCS SAM was affected by both the immersing time and the concentration of the solution, and the maximum contact


Langmuir, Vol. 20, No. 8, 2004 3283

Figure 8. Optical micrographs of TiO2 micropattern deposited on (a) blank PET substrate, (b) OTS SAM, (c) VTCS SAM, (d) PTCS SAM, and (e) TTCS SAM. Table 4. Comparison of TiO2 Micropattern on Different SAMs

angle attained was 158°. However, the VTCS SAM was too thick and could not be combined firmly with the substrate. The fabricated SAMs were modified, and TiO2 micropattern was successfully deposited on the modified SAMs of OTS, PTCS, and TTCS. The micropattern was affected by the properties of SAMs. Among the SAMs, PTCS and TTCS are the best candidates for templates to obtain a high-quality micropattern.

Acknowledgment. This work was partially supported by the 21st Century COE program “Nature-Guided Materials Processing” and by the “Ceramic Integration” Project of the Ministry of Education, Culture, Sports, Science and Technology, Japan. The authors would like to express their appreciation to Yanfeng Gao for his assistance in XPS analysis. LA036088M