Langmuir 2007, 23, 7799-7803
7799
Controlling the Hydrophobicity of Submicrometer Silica Spheres via Surface Modification for Nanocomposite Applications Zhijian Wu,†,‡ Hyuk Han,† Woojoo Han,§ Bumsang Kim,| Kyung Hyun Ahn,*,§ and Kangtaek Lee*,† Department of Chemical Engineering, Yonsei UniVersity, Seoul 120-749, Korea, Qinghai Institute of Salt Lakes, Chinese Academy of Sciences, Xining 810008, China, School of Chemical and Biological Engineering, Seoul National UniVersity, Seoul 151-742, Korea, and Department of Chemical Engineering, Hongik UniVersity, Seoul 121-791, Korea ReceiVed February 9, 2007. In Final Form: April 23, 2007 We control the hydrophobicity of submicrometer silica spheres by modifying their surface with -CH3, -CHdCH2, -(CH2)2CH3, -CH2(CH2)4CH2-, -C6H5, -(CH2)7CH3, and -(CH2)11CH3 groups through a modified one-step process. The scanning electron microscopy (SEM), quasi-elastic light scattering (QELS), UV-visible spectra, nitrogen sorption, and water vapor adsorption methods are used to characterize the particles. The SEM micrographs of the particles demonstrate that the modified particles are uniformly spherical, monodisperse, and well-shaped with the particle size ranging from 130 to 149 nm depending on the modified organic groups. In aqueous solution, the particles modified with phenyl groups have an obvious UV absorption peak at around 210 nm, whereas the other modified particles and unmodified particles do not have any UV-visible absorption peaks. There exist obvious differences in the amount of water vapor adsorbed depending on the type of surface functional groups of the modified particles. Compared with the unmodified particles, the modified particles have a lower water vapor adsorption because of the improved hydrophobicity of the particle surface. As a potential application, we prepared polystyrene/SiO2 nanocomposites by blending polystyrene with the synthesized particles. Water contact angle measurements show that the surface of the composite prepared with the modified particles are more hydrophobic. Confocal microcopy demonstrates that the particles are less agglomerated in the nanocomposite as the particles become more hydrophobic. These comprehensive experimental results demonstrate that the hydrophobicity of the particles can be easily controlled by surface modification with different organosilanes through a modified one-step process.
1. Introduction Spherical silica particles with different surface properties are important in many applications including composite materials, adsorbents, pigments, detergents, cosmetics, and pharmaceuticals.1-4 Chemical treatment and functionalization of the silica particles can be used to improve the hydrophilicity or hydrophobocity of the particle surface for specific applications.5 In the fabrication of polymer nanocomposites, it is critical to make a uniform dispersion of inorganic particles in the polymer matrix,6 but physically blending hydrophobic polymers such as polystyrene (PS) or polypropylene (PP) with hydrophilic inorganic particles may lead to the phase separation or agglomeration of particles, resulting in poor mechanical, optical, and electrical properties. Furthermore, particles with high surface energy can easily agglomerate as the size of the particles decreases.7 To * Corresponding authors. (K.H.A.) E-mail:
[email protected]. Tel: +82-2-8808322. Fax: +82-2-8887295. (K.L.) E-mail:
[email protected]. Tel: +82-2-21232760. Fax: +82-2-3126401. † Yonsei University. ‡ Chinese Academy of Sciences. § Seoul National University. | Hongik University. (1) Nozawa, K.; Gailhanou, H.; Raison, L.; Panizza, P.; Ushiki, H.; Sellier, E.; Delville, J. P.; Delville, M. H. Langmuir 2005, 21, 1516. (2) Kim, S.-Y.; Kim, E.; Kim, S.-S.; Kim, W. J. Colloid Interface Sci. 2005, 292, 93. (3) Miller, C. R.; Vogel, R.; Surawski, P. P. T.; Jack, K. S.; Corrie, S. R.; Trau, M. Langmuir 2005, 21, 9733. (4) Marquez, M.; Grady, B. P.; Robb, I. Colloids Surf., A 2005, 266, 18. (5) Arkhireeva, A.; Hay, J. N.; Oware, W. J. Non-Cryst. Solids 2005, 351, 1688. (6) Von Werne, T.; Patten, T. E.; J. Am. Chem. Soc. 2001, 123, 7497. (7) Rong, M. Z.; Zhang, M. Q.; Ruan, W. H. Mater. Sci. Technol. 2006, 22, 787.
circumvent these problems and achieve a uniform dispersion of particles in the polymer matrix, it is necessary to modify the surface of inorganic particles to improve their compatibility with polymers. For example, pretreatment of inorganic silica particles with organic functional groups could lead to a uniform dispersion of the particles in the polymer matrix. Two methods have been commonly used for such purposes: (i) the silanation of pure silica particles with silane-coupling agents by refluxing in toluene8 and (ii) a hybrid method using tetraethoxysilane and an organosilane with functional groups.3,9-12 The silanation method is a two-step process. Water should be removed thoroughly before silanation, during which a poisonous solventstoluenesis used. For the hybrid method, it is not as easy to control the particle size and shape. In this method, a relatively large amount of expensive organosilanes is usually needed, which limits the applications of the hybrid particles. It is very important to preserve the monodispersity of particles after surface modification in applications such as photonic crystals because the monodispersity of the particles is critical in the assembly of particles.13 Thus, it is very important to prepare modified silica particles with a good shape and a narrow size distribution for such applications.14-16 Although there have been (8) Howard, A. G.; Khdary, N. H. Analyst 2005, 130, 1432. (9) Arkhireeva, A.; Hay, J. N.; Lane, J. M.; Manzqno, M.; Masters, H.; Oware, W.; Shaw, S. J. J. Sol-Gel Sci. Technol. 2004, 31, 31. (10) Chevalier, P. M.; Ou, D. L. J. Sol-Gel Sci. Technol. 2003, 26, 597. (11) Li, Y. S.; Li, B.; Han, N. Y.; Xu, B. J. J. Chromatogr., A 2003, 1021, 183. (12) Choi, J. Y.; Kim, C. H.; Kim, D. K. J. Am. Ceram. Soc. 1998, 81, 1184. (13) Wong, S.; Kitaev, V.; Ozin, G. A. J. Am. Chem. Soc. 2003, 125, 15589. (14) Wang, W.; Gu, B.; Liang, L.; Hamilton, W. A. J. Phys. Chem. B 2003, 107, 12113. (15) Wang, W.; Gu, B.; Liang, L.; Hamilton, W. A. J. Phys. Chem. B 2003, 107, 3400.
10.1021/la700386z CCC: $37.00 © 2007 American Chemical Society Published on Web 06/09/2007
7800 Langmuir, Vol. 23, No. 14, 2007
Wu et al.
Table 1. Precursors for the Preparation of Surface-Modified Particles abbreviation
name
formula
density (g cm-3)
FW
TEOS MTES VTES PTES TSH PhTES OTES DTES
tetraethoxysilane methyltriethoxysilane vinyltriethoxysilane n-propyltriethoxysilane bis(trimethoxysilyl)hexane phenyltriethoxysilane n-octyltriethoxysilane n-dodecyltriethoysilane
Si(OC2H5)4 (C2H5O)3SiCH3 (C2H5O)3SiCHdCH2 (C2H5O)3Si(CH2)2CH3 (CH3O)3Si(CH2)6Si(OCH3)3 (C2H5O)3SiC6H5 (C2H5O)3Si(CH2)7CH3 (C2H5O)3Si(CH2)11CH3
0.934 0.895 0.911 0.892 1.010 0.996 0.875 0.884
208.3 178.3 190.3 206.4 326.5 240.4 276.5 332.6
some reports about the preparation of modified or hybrid silica particles, the preparation of particles with a good shape, narrow size distribution, and different surface hydrophobicity still remains a challenge. In addition, the effect of various surface functional groups on the surface hydrophobicity of the particles is unclear. In this article, we prepare a series of surface-modified silica particles via a modified one-step process in which only a small amount of organosilanes is used to control the hydrophobicity. A water vapor adsorption experiment is used to compare the hydrophobicity of the particles. The synthesized particles are also used in the fabrication of PS/SiO2 nanocompositles. The water contact angle measurement and confocal microscopy are used to examine the hydrophobicity of the nanocomposite surface and the particle dispersion in the polymer matrix, respectively. 2. Experimental Section 2.1. Materials. Tetraethoxysilane, methyltriethoxysilane, vinyltriethoxysilane, and n-propyltriethoxysilane were obtained from Aldrich. Bis(trimethoxysilyl)hexane, phenyltriethoxysilane, noctyltriethoxysilane, and n-dodecyltriethoysilane were obtained from Gelest. The abbreviations and formulas of these precursors are listed in Table 1. Absolute ethanol and ammonia solution (28 wt % ammonia) were obtained from Duksan. All chemicals were used as received without further purification. Polystyene (PS) under the grade name of HF2680 was provided by Samsung Cheil Industries Inc. Xylene solution (82 wt %) was obtained from Showa Chemicals Co. 2.2. Preparation of the Particles. Monodisperse SiO2 spheres were prepared using a modified procedure originally described by Sto¨ber et al.17 The typical preparation involves rapidly mixing two solutions of A and B at room temperature. Solution A is a mixture of 2.79 mL TEOS and 22.2 mL EtOH, and solution B is a mixture of 0.638 mL of 28 wt % ammonia, 6.8 mL of water, and 17.6 mL of EtOH. A modified procedure from Wang et al.14,16 without using any chloroform was used to modify the particles: after 3.5 h of reaction, 4.79 × 10-4 mol of an organosilane was injected into the stirred solution to modify the surface of SiO2 spheres. The reaction was allowed to continue for an additional 19 h with stirring after the addition of the organosilane. For the preparation of pure SiO2 spheres, no precursor was added. After reaction, the particles were separated by centrifugation and washed with ethanol and deionized water. For the washing of the particles, ultrasonication was used to redisperse the particles in the desired solvents, and centrifugation was used in every step of the solid-liquid separation. The particles thus isolated were easily redispersed by ultrasonication in water. Particles were dried at 25 °C for 3 days for N2 sorption and water vapor adsorption experiments. 2.3. Characterization of the Particles. Scanning electron microscopy of the particles was carried out using a JEOL 6300 SEM instrument. The sample was prepared by dispensing drops of an aqueous suspension of particles onto a glass plate. This was allowed to dry at room temperature and was then coated with a thin Pt film. (16) Wang, W.; Gu, B. J. Phys. Chem. B 2005, 109, 22175. (17) Sto¨ber, W.; Fink, A.; Bohn, E. J. Colloid Interface Sci. 1968, 26, 62.
The average diameter of particles dispersed in water was determined by the QELS technique using a Nano ZS Zetasizer (model ZEN3600, Malvern Instruments Ltd.). UV-visible spectra of the particle suspension in water were taken with a TU-1810PC spectrophotometer (Beijing Purkinje General Instrument Co., Ltd, China). The adsorption/desorption isotherms of nitrogen at 77 K were measured with a Micromeritics ASAP 2020 apparatus. Prior to the measurements, the particles were degassed for 2 h at 120 °C. The specific surface area was calculated from the BET equation. 2.4. Water Vapor Adsorption Experiments. Water vapor adsorption was used to compare the hydrophobicity of the particles. A material adsorbing a lower amount of water vapor is known to be more hydrophobic.18 After drying at 80 °C to a constant weight, the particles were allowed to adsorb water vapor in a closed container containing liquid water at 25 °C for 85 h. During the experiments, the particles were not in direct contact with the liquid water to ensure that the particles adsorb water vapor and not liquid water. The amount of water vapor adsorbed was expressed in the percent weight increase of the particles estimated from the particle weigh difference before and after water vapor adsorption. 2.5. Preparation and Characterization of PS/SiO2 Nanocomposites. For the preparation of PS/SiO2 nanocomposites, a polystyre/ xylene solution was prepared by dissolving PS pellets in xylene under magnetic stirring for 48 h at room temperature. After the PS pellets were dissolved in xylene, the synthesized particles were added to the PS/xylene solution and dispersed by an ultrasonic homogenizer for 10 min. The concentration of particles was 10 wt %. For water contact angle measurement, a silica particles/PS/xylene solution was spin-coated onto slide glass at 1000 rpm for 60 s and dried at 80 °C. The contact angle was measured by the sessile drop method using a DSA100 instrument (Kruss). Water (3 µL) was dropped onto the spin-coated sample, and the contact angle was determined by analyzing drop images. Each measurement was repeated five times to ensure reproducibility. The surface of the spin-coated nanocomposites was also observed by confocal microscopy (OLS3000, Olympus) to compare the particle dispersion in the composite.
3. Results and Discussion 3.1. Preparation of the Modified Particles. Using only the organosilanes, it was difficult to obtain spherical particles, and in some cases no particles were formed (results not shown). Although it was possible to prepare organic/inorganic hybrid particles using both TEOS and organosilanes, it was not easy to control the shape and size of the particles (results not shown). Thus, we used a modified one-step process similar to that of Wang et al.14,16 First, we prepared pure silica particles by the Sto¨ber process.17 After 3.5 h of reaction, we added a small amount of an organosilane to the reaction mixture to modify the surface of the particles. In this method, the size of the modified particles can be easily controlled by using pure “mother” silica particles of different sizes, and the surface of the particles can be easily modified by adding different types of organosilanes. Using this method, we could easily prepare uniform and well-shaped particles of different sizes and surface properties. 3.2. Characterization of the Particles. The SEM micrographs of the particles are shown in Figure 1, which demonstrates that
Hydrophobicity of Submicrometer Silica Spheres
Langmuir, Vol. 23, No. 14, 2007 7801
Figure 2. UV spectra of the synthesized particles.
Figure 2 shows the effect of surface modification of the particles on the UV-visible absorption spectra. PhTES particles have an obvious absorption at around 210 nm, indicating the successful modification of the silica particles by phenyltriethoxysilane. Whereas the other particles have spectra of the same shape, they do not exhibit such absorption because of the lack of phenyl groups and other UV-responsive functional groups. For the silica particles modified with 2-(4-chlorosulfonylphenyl)ethyltrimethoxysilane, UV absorption at 225 nm was also observed as a result of the presence of phenol groups.14 Nitrogen adsorption/desorption isotherms of the particles are shown in Figure 3. All of the particles show isotherms of type II, typical of nonporous solids, with type H1 hysteresis loops that are often obtained with materials consisting of agglomerates or compacts of approximately uniform spheres.20 Assuming nonporous particles, the specific surface area (m2 g-1) of the particles can be calculated on the basis of the SEM particle size by the following equation
Scal ) Figure 1. SEM micrographs of the synthesized particles: (a) TEOS, (b) MTES, (c) VTES, (d) PTES, (e) TSH, (f) PhTES, (g) OTES, and (h) DTES. Table 2. Comparison of Particle Sizes Determined by QELS and SEM Methods QELS
SEM
typea
average diameter (nm)
average diameter (nm)
TEOS MTES VTES PTES TSH PhTES OTES DTES
164.0 167.0 165.0 162.0 163.0 167.0 165.0 165.0
149.0 148.0 143.0 140.0 136.0 135.0 139.0 140.0
a TEOS stands for the unmodified particles. The other symbols stand for the particles modified with the corresponding organosilanes.
the modified particles are uniformly spherical, monodisperse, and well-shaped. The particle size determined by QELS and SEM is compared in Table 2. Before the addition of the organosilane (Sto¨ber reaction for 3.5 h), the size of the silica particles was 132 nm by QELS. After the addition of organosilanes and continued reaction for 19 h, the size of both pure silica particles (without further addition of any precursor) and modified particles increased. The QELS method always gives a larger size than does the SEM method as reported by Costa et al.19
6000 Fd
(1)
where d is the average particle diameter (nm) from SEM images and F is the density of the particles, which was taken to be 2.08 g cm-3.21 The calculated and determined BET surface areas of the particles are compared in Figure 4. Even though the BET surface areas are always larger than the calculated values, they are of the same order of magnitude. Because nonporous particles usually have a surface area of less than 100 m2 g-1,22 all of the synthesized particles can be considered to be virtually nonporous. This is consistent with hybrid particles prepared from tetraethoxysilane and vinyltriethoxysilane that were also found to be nonporous.11 3.3. Water Vapor Adsorption Results. The amount of water vapor adsorbed on the particles is compared in Figure 5. The lower water vapor adsorption suggests that the surfaces of the particles are more hydrophobic.18 In Figure 5, DTES- and OTESmodified particles show the lowest water vapor adsorption, whereas pure silica particles show the highest. The long carbon chains on the surfaces of DTES- and OTES-modified particles are responsible for the lowest water vapor adsorption, suggesting the most hydrophobic surface. These results demonstrate that the particles exhibit different hydrophobicity after surface (18) Zhao, X. S.; Lu, G. Q.; Hu, X. Microporous Mesoporous Mater. 2000, 41, 37. (19) Costa, C. A. R.; Leite, C. A. P.; Galembeck, F. J. Phys. Chem. B 2003, 107, 4747. (20) Khalil, K. M. S.; Elsamahy, A. A.; Elanany, M. S. J. Colloid Interface Sci. 2002, 249, 359. (21) Kobayashi, M.; Skarba, M.; Galletto, P.; Cakara, D.; Borkovec, M. J. Colloid Interface Sci. 2005, 292, 139. (22) Pramanik, N.; Tarafdar, A.; Pramanik, P. J. Mater. Process. Technol. 2007, 184, 131.
7802 Langmuir, Vol. 23, No. 14, 2007
Wu et al.
Figure 3. N2 adsorption/desorption isotherms of the synthesized particles.
Figure 4. Comparison of BET surface area of the synthesized particles.
Figure 5. Comparison of water vapor adsorption on the synthesized particles.
modification with different organic groups. Our previous work has demonstrated that the use of organosilanes in the preparation of organic-inorganic hybrid gels also produced hydrophobic surfaces.23-25 (23) Wu, Z.; Ahn, I.-S.; Lee, C.-H.; Kim, J.-H.; Shul, Y. G.; Lee, K. Colloids Surf., A 2004, 240, 157. (24) Wu, Z.; Joo, H.; Ahn, I.-S.; Haam, S.; Kim, J.-H.; Lee, K. Chem. Eng. J. 2004, 102, 277. (25) Wu, Z.; Joo, H.; Lee, K. Chem. Eng. J. 2005, 112, 227.
Figure 6. Comparison of water contact angles on PS/SiO2 nanocomposites.
3.4. Characterization of the PS/SiO2 Nanocomposites. PS/ SiO2 nanocomposites were prepared using the synthesized particles. We expect that using particles with different surface functionalities should affect both the surface hydrophobicity and the particle dispersion in nanocomposites. Figure 6 shows the water contact angles of the PS/SiO2 nanocomposites prepared from particles with different functionalities. The DTES- and OTES-modified particles exhibit the highest contact angles (i.e., the most hydrophobic surface), and the unmodified particle (TEOS) show the lowest contact angles (i.e., the least hydrophobic surface). This is consistent with the previous water vapor adsorption results because the use of hydrophobic particles (DTES- and OTES- modified particles) leads to hydrophobic surfaces of nanocomposites. Thus, it is possible to control the surface hydrophobicity of nanocomposites by a simple surface modification of silica particles with different organic groups. Note that the contact angle of the nanocomposite made from pure silica particles (TEOS) is still high (86.7°) as a result of the presence of polystyrene. Confocal microscopy was also used to compare the particle dispersion in nanocomposites. In Figure 7, the dark area and bright area represent the dispersed particles and the PS matrix, respectively; the unmodified particles (TEOS) are agglomerated, but the modified particles are less agglomerated in nanocomposites, probably as a result of their hydrophobic surfaces. The most hydrophobic particles, such as DTES-, OTES-, and PhTES-
Hydrophobicity of Submicrometer Silica Spheres
Langmuir, Vol. 23, No. 14, 2007 7803
particle dispersibility and hence the rheological properties of nanocompsites by a surface modification of the particles.
4. Summary Submicrometer silica spheres modified with different organic functional groups were prepared through a modified one-step process. Various techniques including SEM, QELS, UV-visible spectra, nitrogen sorption, and water vapor adsorption were used to characterize the modified particles comprehensively. The SEM, QELS, and UV-visible results confirm the successful modification of the silica particle surface with no deterioration of the spherical shape. Nitrogen sorption results suggest that the synthesized particles are virtually nonporous. The particles modified with DTES or OTES are found to be the most hydrophobic from the water vapor adsorption experiment. The water contact angle measurement of the PS/SiO2 nanocomposites suggests that the use of hydrophobic particles in the preparation of nanocomposites results in the hydrophobic nanocomposite surface. Confocal microscopy experiments show that hydrophobic particles exhibit a lower degree of agglomeration and a higher dispersibility in nanocomposites. The method used in this article is a convenient and economical tool for preparing uniform and well-shaped particles of different sizes and surface hydrophobicity. Only a very small amount of organosilane is needed for the surface modification. The effectiveness of such a method has been confirmed by the comprehensive comparison of the surface properties of the particles and PS/SiO2 nanocomposites. The knowledge gained from this study will be useful in the design of advanced materials such as nanocomposites with desired surface and rheological properties.
Figure 7. Confocal micrographs of the PS/SiO2 nanocomposites (magnification 5× and 20× (inset)): (a) TEOS, (b) MTES, (c) VTES, (d) PTES, (e) TSH, (f) PhTES, (g) OTES, and (h) DTES.
modified particles, exhibit the least agglomeration and the best dispersibility. These results suggest the possibility of easily tuning
Acknowledgment. K.L. is grateful for financial support from the Korean Energy Management Corporation and the KOSEF through the Basic Research Fund (R01-2004-000-10944-0) and the National Core Research Center for Nanomedical Technology (R15-2004-024-00000-0). K.H.A. thanks the Korea Energy Management Corporation (KEMCO) for financial support through project no. 2005-R-NM01-P-01-2-400-2005. LA700386Z
