pubs.acs.org/Langmuir © 2010 American Chemical Society
Solvent Effects on Molecular Packing and Tribological Properties of Octadecyltrichlorosilane Films on Silicon Yue-an Cheng,† Bin Zheng,‡ Po-hsiang Chuang,† and Shuchen Hsieh*,† †
Department of Chemistry and Center for Nanoscience and Nanotechnology, National Sun Yat-sen University, Kaohsiung 804, Taiwan, and ‡School of Materials Science and Engineering, Dalian University of Technology, Dalian, Liaoning 116024, People’s Republic of China Received December 10, 2009. Revised Manuscript Received January 11, 2010
Self-assembled monolayer films of octadecyltrichlorosilane were prepared on silicon substrates using hexadecane, toluene, chloroform, and dichloromethane to determine the effects of solvent on molecular packing and tribological properties. Topographical atomic force microscopy images were used to evaluate the film quality and determine surface roughness, and tribological measurements, including friction, adhesion, and elasticity, provided additional information on the local nanoscale packing of the films. Our results showed that solvent viscosity and polarity affected the tribological properties of the films, with films prepared using hexadecane exhibiting superior properties. LangmuirBlodgett experiments indicated that intermolecular interactions were stronger between octadecyltrichlorosilane and hexadecane molecules than for any other solvent in this study. These results demonstrate that solvent properties are an important consideration in monolayer film preparation and optimization for friction controlled surfaces.
Introduction In recent years, self-assembled monolayers (SAMs) have attracted increased interest due to their potential application in lubrication films, biological sensors, and etching techniques.1-3 In particular, SAM films are now used in micro-electromechanical (MEMs) and nano-electromechanical systems (NEMs) to reduce nanoscale surface friction. Recently, precise control over surface tribological properties of SAMs, such as reducing stiction in MEMS/NEMS devices, has become an important topic of research. The friction and adhesion properties of SAM films depend largely on the molecular structure of the adsorbing species. For example, chain length and functional group selection can be used to tune the resulting surface properties and to optimize the molecular packing of the films.4-8 Many parametric strategies have been used to improve the molecular packing, such as concentration, immersion time, solution temperature, oxygen *To whom correspondence should be addressed: Fax þ88675253908; Tel þ88675252000 ext 3931; e-mail
[email protected].
(1) Maboudian, R. Surf. Sci. Rep. 1998, 30, 207. (2) Maboudian, R.; Ashurst, W. R.; Carraro, C. Tribol. Lett. 2002, 12, 95. (3) Kim, G. M.; Kim, B.; Liebau, M.; Huskens, J.; Reinhoudt, D. N.; Brugger, J. J. Microelectromech. Syst. 2002, 11, 175. (4) Brewer, N. J.; Foster, T. T.; Leggett, G. J.; Alexander, M. R.; McAlpine, E. J. Phys. Chem. B 2004, 108, 4723. (5) Barrena, E.; Kopta, S.; Ogletree, D. F.; Charych, D. H.; Salmeron, M. Phys. Rev. Lett. 1999, 82, 2880. (6) Song, S.; Chu, R.; Zhou, J.; Yang, S.; Zhang, J. J. Phys. Chem. C 2008, 112, 3805. (7) Duwez, A. S.; Jonas, U.; Klein, H. ChemPhysChem 2003, 4, 1107. (8) Mikulski, P. T.; Harrison, J. A. J. Am. Chem. Soc. 2001, 123, 6873. (9) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103. (10) Kawasaki, M.; Sato, T.; Tanaka, T.; Takao, K. Langmuir 2000, 16, 1719. (11) Bensebaa, F.; Voicu, R.; Huron, L.; Ellis, T. H. Langmuir 1997, 13, 5335. (12) Yamada, R.; Wano, H.; Uosaki, K. Langmuir 2000, 16, 5523. (13) Glaser, A.; Foisner, J.; Hoffmann, H.; Friedbacher, G. Langmuir 2004, 20, 5599. (14) Love, J. C.; Wolfe, D. B.; Chabinyc, M. L.; Paul, K. E.; Whitesides, G. M. J. Am. Chem. Soc. 2002, 124, 1576. (15) Laibinis, P. E.; Whitesides, G. M.; Allara, D. L.; Tao, Y. T.; Parikh, A. N.; Nuzzo, R. G. J. Am. Chem. Soc. 1991, 113, 7152.
8256 DOI: 10.1021/la904656y
concentration, and substrate.9-15 However, few studies to date have addressed the issue of solvent effects, which may exhibit an equally important influence on film growth and the resulting tribological properties.16,17 In this study, we have examined solvent effects on topography and tribological properties of octadecyltrichlorosilane (OTS) SAM films on silicon. Films were prepared using OTS dissolved in hexadecane, toluene, chloroform, and dichloromethane, and the resulting film quality was characterized by FTIR and AFM. Friction measurements were performed using AFM under increasing normal loads to determine the friction coefficient. Adhesive and elastic properties were determined by collecting and fitting AFM force-displacement data on each type of SAM film sample. Langmuir-Blodgett experiments were used to examine the intermolecular interaction between OTS and each solvent tested. A correlation between film tribological properties and solvent properties (viscosity and polarity) was revealed and discussed.
Experimental Section Sample Preparation. Silicon (100) wafer substrates (TSR Technology Inc.; P-type/boron dopant; resistivity 1-10 ohm 3 cm) were cleaned using an Extran neutral surfactant (Merck), then rinsed sequentially with Milli-Q reagent-grade (type I) water (18.2 MΩ 3 cm at 25 °C) and ethanol (ACS HPLC grade), and dried under a stream of nitrogen gas. The clean silicon substrates were then placed in a plasma cleaner (Harrick Scientific Products, Inc., model PDC-32G) for 2 min to remove residual organic contamination and to increase the OH concentration at the surface. All glassware was cleaned and rinsed using the same procedure as for the silicon substrates, and then baked overnight at 150 °C prior to use to remove residual adsorbed water. Self-assembled monolayers of n-octadecyltrichlorosilane (CH3(CH2)17SiCl3, Gelest Inc.) were formed by immediately immersing the plasma-cleaned silicon substrates into vials containing a solution of 1 mM OTS dissolved in hexadecane (ACROS), (16) Dannenberger, O.; Wolff, J. J.; Buck, M. Langmuir 1998, 14, 4679. (17) Rozlosnik, N.; Gerstenberg, M. C.; Larsen, N. B. Langmuir 2003, 19, 1182.
Published on Web 01/28/2010
Langmuir 2010, 26(11), 8256–8261
Cheng et al.
Article
Figure 1. (a) Cartoon illustrating an AFM tip contacting a surface during a force-distance measurement. The equation shown corresponds to Hooke’s law, where R, a, δ represent the radius of curvature of the tip, radius of the projected contact area, and indented deformation change of the surface, respectively. (b) Two springs in series representing the cantilever and film and the spring constant equation that relates them. toluene (TEDIA Company, Inc.), chloroform (TEDIA Company, Inc.), or dichloromethane (Merck). The vials were immediately sealed to prevent water absorption due to ambient humidity and allowed to sit for 24 h at room temperature and then rinsed sequentially for 10 min each, in chloroform, isopropanol, and deionized water.18 Samples were then annealed in an oven for 10 min at 115 °C. In this report, OTS films prepared with hexadecane, toluene, chloroform, and dichloromethane solvents are denoted as C18-H, C18-T, C18-C, and C18-D films, respectively. Instrumentation. Transmission Fourier transform infrared (FTIR) spectra were acquired (8 cm-1 resolution, 1000 scans) using a Perkin-Elmer FTIR spectrometer (Spectrum 100) equipped with a liquid nitrogen-cooled mercury-cadmium-telluride (MCT) detector.18,19 Topographic images, surface roughness, adhesion, and friction measurements were performed using an atomic force microscope (MFP-3D, Asylum Research, Santa Barbara, CA) under ambient conditions. Silicon cantilevers (Olympus AC240) with a normal spring constant of 0.74 ( 0.07 N/m (thermal method)20,21 were used for all experiments. Data Analysis. Friction force values for each sample were determined by performing friction loops (average of 30 measurements) and calculating the relative friction force (trace - retrace)/ 2 at various loads (10-55 nN). A sliding speed of 12.5 μm/s and a scan length of 5 μm were used for all measurements. The friction force may be represented by the following equation: Ff ¼ F0 þ μFn
ð1Þ
where μ is the kinetic coefficient of friction, Fn is the normal force, and F0 is the residual force. The residual force (F0) is not accounted for by the normal load but can be estimated by extrapolating friction force versus normal load plots to zero load. Residual forces for one-component and two-component monolayers are typically negligible in tribological tests.22 The elastic properties of a sample can be extracted from AFM force curves, where the tip is driven toward the sample until contact is made and a trigger setpoint force value reached (10 nN in this study). Thirty force curve measurements were made on each sample and average values obtained. The DMT23 and JKR24 contact models were used in this study. Both include molecular adhesive forces in the contact response of a spherical tip against a flat surface (Figure 1a), but the DMT model considers deformation from the adhesion force. The two models (18) Hsieh, S. J. Nanosci. Nanotechnol. 2008, 8, 5839. (19) Sambasivan, S.; Hsieh, S.; Fischer, D. A.; Hsu, S. M. J. Vac. Sci. Technol. A 2008, 24, 1484. (20) Hutter, J. L.; Bechhoefer, J. Rev. Sci. Instrum. 1993, 64, 1868. (21) Proksch, R.; Sch€affer, T. E.; Cleveland, J. P.; Callahan, R. C.; Viani, M. B. Nanotechnology 2004, 15, 1344. (22) Vilt, S. G.; Leng, Z.; Booth, B. D.; McCabe, C.; Jennings, G. K. J. Phys. Chem. C 2009, 113, 14972. (23) Derjaguin, B. V.; Muller, V. M.; Toporov, Y. P. J. Colloid Interface Sci. 1975, 53, 314. (24) Johnson, K. L.; Kendall, K.; Roberts, A. D. Proc. R. Soc. London 1971, 324, 301.
Langmuir 2010, 26(11), 8256–8261
Figure 2. Schematic representation of a surface pressure vs area isothermal curve using the Wilhelmy plate method. give the following relations between the contact radius a and the loading force F. " DMT : JKR :
a ¼ a ¼
3ð1 - υ2 ÞR ðF þ 2πwRÞ 4E
#1=3
3ð1 - υ2 ÞR ½F þ ð3πwR=2Þ þ 4E
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi !1=3 6πwRF þ ð3πwRÞ2
ð2Þ Here, E is the Young’s modulus, υ is the Poisson ratio of the sample, and w is the work of adhesion. A value of υ = 0.44 was used for all film samples. To find w, eq 2 was solved at a = 0, which results in w = Fad/2πR (DMT) and w = 2Fad/3πR (JKR), where Fad is the adhesion force from each F-d data set. To calculate the Young’s modulus, it is necessary to determine the contact area of the tip. We used Hooke’s law as shown in Figure 1a to relate the displacement of the cantilever to the contact radius via the spring constant k, as shown in eq 3 (geometry method): qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi a ¼ R2 - ½R - ðd - F=kÞ2 ð3Þ In addition, if the contact radius can be assumed to be Hertzian, eq 4 applies: pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi a ¼ ðd - F=kÞR ð4Þ
Langmuir-Blodgett Method. A Langmuir-Blodgett trough (NIMA-312D) was used to perform isothermal compression experiments to characterize the OTS-solvent interaction. Measurements were made immediately following dispersion of the OTS/solvent solutions (10 μL aliquots of 0.5, 1, 2, and 5 mM) on the water surface of the LB trough to minimize solvent evaporation prior to the measurement, so that the interaction between the solvent and OTS molecules could be characterized. This deformation can change the surface tension of OTS films. The Wilhelmy balance method was employed to determine the correlation between surface tension and surface area of the films. Under isothermal conditions πA ¼ kT
ð5Þ
where π is the surface pressure, which denotes the surface tension, A is the surface area, and k is the Boltzmann constant. Equation 5 shows that surface pressure and area are inversely proportional. Therefore, the inflection point of surface pressure can be used to identify a phase transition, as shown in Figure 2. For solid monolayer films, further compression results in the collapse of surface molecular DOI: 10.1021/la904656y
8257
Article
Cheng et al.
Figure 3. AFM topographical images of OTS-SAM films prepared with different solvents: (a) bare silicon; (b) C18-H film; (c) C18-T film; (d) C18-C film; (e) C18-D film. All images were acquired in AC tapping mode with a scan area of 5 5 μm2; rms surface roughness was calculated from the entire 5 5 μm2 area of the height images. Table 1. Solvent Viscosity, Polarity, and SAM Film Surface Roughness solvent viscosity ηc (mPa 3 s) hexadecane toluene chloroform dichloromethane
3.34 0.59 0.56 0.39
solvent polarity c SAM rms (D) (pm) 0 0.36 1.08 1.14
69.8 73.1 246.7 1296
structure, which corresponds to the collapse point of the surface pressure. A large surface pressure at the collapse point indicates a strong interaction between the OTS and solvent molecules.
Results and Discussion In Figure 3, we show the surface quality of SAM films on silicon prepared using hexadecane (b), toluene (c), chloroform (d), and dichloromethane (e) solvents. A bare-silicon sample is included for comparison (Figure 3a). The insets in each figure display water contact angle profiles, which increased from very hydrophilic (contact angle ∼100°). Literature results have shown that the contact angle for a complete monolayer of long-chain SAM molecules on silicon saturates at 100°.25,26 All of the reported samples in this study had water contact angles >100°. (25) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321. (26) Creager, S. E.; Clarke, J. Langmuir 1994, 10, 3675.
8258 DOI: 10.1021/la904656y
All of the films were very smooth, with rms surface roughness of toluene-OTS>chloroform-OTS=dichloromethane-OTS.
Conclusions We prepared OTS SAM films on silicon surfaces using hexadecane, toluene, chloroform, and dichloromethane. AFM was used to probe the local nanoscale forces (adhesion, friction, and indentation) to identify the molecular packing and quality of the films, and Langmuir-Blodgett experiments provided information on solvent-solute intermolecular interactions. Experimental results showed that the compatibility/intermixing between the solvent and the SAM molecules enabled ultrathin molecular films to form and assemble orderly onto the silicon surface. The OTS SAM films prepared in hexadecane solvent had the lowest surface roughness and exhibited dense molecular packing. We attribute this to the solvent properties of high viscosity and low polarity. A Langmuir 2010, 26(11), 8256–8261
Cheng et al.
solvent property relationship was developed for the optimization of SAMs for MEMS, magnetic storage, and other applications. Acknowledgment. The authors thank the National Science Council (NSC 97-2113-M-110-007) of Taiwan and the National Sun Yat-sen University Center for Nanoscience and Nanotechnology for financial support of this work.
Langmuir 2010, 26(11), 8256–8261
Article
Supporting Information Available: FTIR experimental details and spectra; FTIR was used to confirm the OTS self-assembled monolayer formation on silicon and to provide additional support for molecular packing arguments based on AFM results. This material is available free of charge via the Internet at http://pubs. acs.org.
DOI: 10.1021/la904656y
8261
