pubs.acs.org/langmuir © 2009 American Chemical Society
Growth and Stability of a Self-Assembled Monolayer on Plasma-Treated Mica Ibrahim B. Malham and Lionel Bureau* Institut des Nanosciences de Paris, UMR 7588 CNRS-Universit es Paris 6 and 7, 140 rue de Lourmel, 75015 Paris, France Received December 22, 2008. Revised Manuscript Received February 2, 2009 Increasing the range of surfaces that can be studied using the surface forces apparatus, e.g., for friction measurements, requires chemical modification of the surface of mica, which may be achieved by grafting selfassembled monolayers (SAMs) onto plasma-modified mica. In order to gain a better idea of the grafting conditions leading to good-quality SAMs on plasma-activated mica, we focus, in the present work, on the early stages of octadecyltriethoxysilane (OTE) deposition. We use atomic force microscopy to study the morphologies of incomplete monolayers obtained at different grafting temperatures in the range 10-30 °C, and for different immersion times in dilute solutions of OTE. We observe that OTE molecules deposited on plasma-treated mica form stable and robust layers, the morphology of which are markedly affected by the grafting temperature: submonolayers deposited at temperatures below 18 °C exhibit micrometer-sized islands of well-packed molecules, whereas much smaller ordered domains, coexisting with a “liquid-expanded” phase, are observed at room temperature and above. We use water contact angle measurements to probe the quality of the SAMs, and find in particular that the most hydrophobic SAMs, presenting large contact angles and low hysteresis, are obtained for deposition temperature below 18 °C.
Introduction A widely used strategy for tailoring the surface properties of a solid is based on assembling amphiphilic molecules into an ordered monolayer on the substrate.1-3 Such self-assembled monolayers (SAMs) can form on a variety of substrates (metals, metal oxides, semiconductors, and so forth), and surface passivation or functionalization is achieved by properly choosing the chemical endgroups of the molecules.1,2 SAMs therefore represent attractive systems when control of interfacial properties is required, e.g., in situations involving wetting,4-6 adhesion,7-9 or friction.10-14 Understanding the formation mechanisms and structure of self-assembled monolayers has been the scope of numerous studies, among which three classes of model systems have been extensively scrutinized: (i) alkanethiols on gold,2,15-17 *E-mail:
[email protected]. (1) Ulman, A. Chem. Rev. 1996, 6, 1533. (2) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Chem. Rev. 2005, 105, 1103. (3) Schreiber, F. Prog. Surf. Sci. 2000, 65, 151. (4) Silberzan, P.; Leger, L. Phys. Rev. Lett. 1991, 66, 185. (5) Silberzan, P.; Leger, L.; Ausserre, D.; Benattar, J. J. Langmuir 1991, 7, 1647. (6) Chaudhury, M. K.; Whitesides, G. M. Science 1992, 256, 1539. (7) Noy, A.; Frisbie, C. D.; Rozsnyai, L. F.; Wrighton, M. S.; Lieber, C. M. J. Am. Chem. Soc. 1995, 117, 7943. (8) Chaudhury, M. K.; Whitesides, G. M. Science 1992, 255, 1230. (9) Maboudian, R.; Ashurst, W. R.; Carraro, C. Sens. Actuators, A 2000, 82, 219. (10) Kim, H. I.; Koini, T.; Lee, T. R.; Perry, S. S. Langmuir 1997, 13, 7192. (11) Barrena, E.; Kopta, S.; Ogletree, D. F.; Charych, D. H.; Salmeron, M. Phys. Rev. Lett. 1999, 82, 2880. (12) Xiao, X. D.; Hu, J.; Charych, D. H.; Salmeron, M. Langmuir 1996, 12, 235. (13) Vorvolakos, K; Chaudhury, M. K. Langmuir 2003, 19, 6778. (14) Bureau, L.; Caroli, C.; Baumberger, T. Phys. Rev. Lett. 2006, 97, 225501. (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. (16) Liu, G. Y.; Salmeron, M. Langmuir 1994, 10, 367. (17) Lio, A.; Charych, D. H.; Salmeron, M. J. Phys. Chem. B 1997, 101, 3800.
Langmuir 2009, 25(10), 5631–5636
(ii) alkylsilanes on silicon oxide1,5,18-24 (typically the native oxide layer of silicon wafers), and (iii) alkylsilanes on mica.17,23-30 These three types of substrate offer the advantage of being smooth and planar, which makes them suitable for characterization of the deposited SAM using various techniques (spectroscopy, scanning probe microscopy, contact angle measurements). It appears that the SAM structure and degree of order with such systems result from the subtle interplay of (i) interchain interactions (typically Van der Waals interactions between alkyl chains) and (ii) interactions between the molecular headgroups and the solid surface. The three above-mentioned model systems correspond to distinct types of headgroups/surface interactions: thiols bind to the metal surface through sulfur-gold bridges, chloro- or alkoxysilanes form covalent siloxane (Si-O-Si) bonds with the hydroxylated silicon oxide surface, whereas silane headgroups are only weakly adsorbed on the surface of freshly cleaved mica, which is chemically inert. (18) Brzoska, J. B.; Benazouz, I.; Rondelez, F. Langmuir 1994, 10, 4367. (19) Parikh, A. N.; Allara, D. L.; Azouz, I. B.; Rondelez, F. J. Phys. Chem. 1994, 98, 7577. (20) Davidovits, J. V.; Pho, V.; Silberzan, P.; Goldmann, M. Surf. Sci. 1996, 352, 369. (21) Barrat, A.; Silberzan, P.; Bourdieu, L.; Chatenay, D. Europhys. Lett. 1992, 20, 633. (22) Carraro, C.; Yauw, O. W.; Sung, M. M.; Maboudian, R. J. Phys. Chem. B 1998, 102, 4441. (23) Vallant, T.; Brunner, H.; Mayer, U.; Hoffmann, H.; Leitner, T.; Resch, R.; Friedbacher, G. J. Phys. Chem. B 1998, 102, 7190. (24) Brunner, H.; Vallant, T.; Mayer, U.; Hoffmann, H.; Basnar, B.; Vallant, M.; Friedbacher, G. Langmuir 1999, 15, 1899. (25) Rill, C.; Glaser, A.; Foisner, J.; Hoffmann, H.; Friedbacher, G. Langmuir 2005, 21, 6289. (26) Schwartz, D. K.; Steinberg, S.; Israelachvili, J.; Zasadzinski, J. A. N. Phys. Rev. Lett. 1992, 69, 3354. (27) Britt, W. D.; Hlady, V. J. Colloid Interface Sci. 1996, 178, 775. (28) Xiao, X. D.; Liu, G. Y.; Charych, D. H.; Salmeron, M. Langmuir 1995, 11, 1600. (29) Kessel, C. R.; Granick, S. Langmuir 1991, 7, 532. (30) Wood, J.; Sharma, R. Langmuir 1994, 10, 2307.
Published on Web 2/24/2009
DOI: 10.1021/la804213q
5631
Article
Malham and Bureau
Modification of the mica surface using SAMs or adsorbed molecules is a common and long-standing issue in experiments using the surface forces apparatus (SFA). Indeed, SFA experiments use freshly cleaved mica sheets as atomically smooth substrates, which can be approached down to nanometer distances in order to measure surface interactions through various liquids or gases. Soon after the pioneer studies on surface forces employing bare mica, modification of the mica surfaces for use in the SFA has been envisaged, using adsorption of surfactants from solutions, in order to probe, for instance, the so-called hydrophobic interactions.31 However, in the absence of strong anchoring between the deposited molecules and the mica, interpretation of the measurements was complicated by the question of the robustness and stability of the monolayer against mechanical sollicitation32 or displacement by other molecules.33 Different routes have been proposed in order to treat the mica surface for subsequent covalent attachment of molecules: (i) Ion exchange, i.e., replacement of the native K+ by H+ ions at the mica surface by immersion in hydrochloric acid prior to grafting trichlorosilanes.34 (ii) Adsorption of prehydrolyzed trialkoxysilanes onto freshly cleaved mica followed by postbaking (T J 100 °C for a few hours).29 (iii) Steam treatment, which consists in exposing freshly cleaved mica to water vapor prior to immersion in a trichlorosilane solution.26 (iv) Exposing mica to a plasma created in water vapor under reduced pressure prior to SAM grafting.33 The first three methods are believed to result in sparse siloxane bonds created between the mica surface and the hydrolyzed headgroups of chloro or alkoxysilane. The effect of plasma treatment on mica surface composition has been more systematically characterized using surface analysis techniques (ESCA, XPS, and TOF-SIMS). Very recently, Liberelle et al.36 have performed an extensive study which shows that silanol groups (Si-OH) are abundantly created at the mica surface when it is exposed to an Ar/H2O plasma. These authors further show that alkylsilane layers deposited on plasma-treated mica are well-anchored and that they exhibit good resistance to prolonged soaking in basic and acid aqueous solutions. Plasma treatment therefore appears to be a convenient method for surface activation and covalent grafting of molecules on mica, thus making an instrument like the surface forces apparatus much more versatile with respect to the nature of the surfaces under study. Now, it is well-known that the quality of a SAM (surface coverage, molecular arrangement of the alkyl chains) drastically depends on deposition conditions (e.g., temperature19,22,25 or solution concentration38) and that the growth mechanisms and formation kinetics are also affected by the details of substrate surface composition.24 The SAM quality in turn affects the properties of the surface, e.g., its (31) Israelachvili, J. N.; Pashley, R. Nature 1982, 300, 341. (32) Liberelle, B.; Giasson, S. Langmuir 2008, 24, 1550. (33) Parker, J. L.; Cho, D. L.; Claesson, P. M. J. Phys. Chem. 1989, 93, 6121. (34) Carson, G. A.; Granick, S. J. Mater. Res. 1990, 5, 1745. (35) Kim, S.; Christenson, H. K.; Curry, J. E. J. Phys. Chem. B 2003, 107, 3774. (36) Liberelle, B.; Banquy, X.; Giasson, S. Langmuir 2008, 24, 3280. (37) Wasserman, S. R.; Tao, Y. T.; Whitesides, G. M. Langmuir 1989, 5, 1074. (38) Foisner, J.; Glaser, A.; Kattner, J.; Hoffmann, H.; Friedbacher, G. Langmuir 2003, 19, 3741.
5632
DOI: 10.1021/la804213q
wettability.19 In various studies in which monolayers of octadecylsilane were deposited on plasma-treated mica, it is noteworthy that the reported values for contact angle hysteresis (in the range 20-30 °C30,35,36) lie above those reported for compact layers of the same C18 silane on silica37 or ionexchanged mica.34 This suggests that these C18 SAMs on plasma-modified mica displayed coverage defects or regions in which alkyl chains were not all-trans and well-packed. This prompted us to investigate in more detail the growth mechanisms of an alkylsilane layer on plasma-treated mica, for which no data are available, in order to gain a better idea of the grafting conditions leading to good-quality SAMs on such a substrate. For this purpose, we focus on the early (submonolayer) stages of octadecyltriethoxysilane (OTE) deposition and use atomic force microscopy to perform a qualitative study of the submonolayer morphologies obtained at different grafting temperatures in the range 10-30 °C and for different immersion times of plasma-treated mica in dilute solutions of OTE. We observe that OTE molecules deposited on plasmatreated mica form stable and robust layers, the morphology of which is markedly affected by the grafting temperature: submonolayers exhibit micrometer-sized islands of wellpacked molecules at temperatures below 18 °C, whereas much smaller ordered domains, coexisting with a “liquid-expanded” phase, are observed at room temperature and above. We use water contact angle measurements to probe the quality of the SAMs and find in particular that the most hydrophobic SAMs, presenting large contact angles and low hysteresis, are obtained with deposition temperatures below 18 °C.
Experimental Methods SAM Deposition. Self-assembled monolayers of n-octodecyltriethoxysilane (OTE 94%, from ABCR, Germany, used as received) were deposited on mica following a protocol similar to that proposed initially by Kessel and Granick29 and modified by Xiao et al.28 OTE was prehydrolyzed by dissolving 0.4470 g of silane in 50 mL of tetrahydrofuran (THF puriss. from Riedel-de Haen, Germany) containing hydrochloric acid at a concentration of 0.1 M. The solution was stirred at room temperature for 3-5 days. 2.5 mL of the prehydrolysis solution was then diluted in 50 mL of cyclohexane (puriss. grade, Riedel-de Haen), while continuously stirring. After 30 min of agitation, the solution was filtered through a 0.2 μm PTFE membrane and brought to controlled temperature by immersion of the beaker in a dodecane bath thermoregulated at (0.2 °C using a water-circulating system. OTE layers were deposited on untreated and plasma-treated mica surfaces at different temperatures within the range 10-30 °C. For each temperature, we have typically used four mica substrates, each immersed in the deposition solution for a given amount of time (1, 5, 15, and 30 s). Independently of the deposition conditions (time, temperature, or mica treatment), we observed that samples emerged almost completely dry from the silane solution, and residual droplets on the samples were immediately blown off in a stream of argon. Without further rinsing or post-treatment, OTE layers were then characterized using atomic force microscopy, at most a few hours after their deposition. Note that the shortest (1 s) immersion time used in our study is on the order of the time needed to plunge and withdraw the mica surface from the grafting solution (∼0.2-0.3s). This results in coverage and wettability gradients along one direction on the surface. However, AFM images show that the central part of such samples displays a reasonably homogeneous coverage, over approximately 1 cm2. Moreover, we obtained good reproducibility of the coverage over this central region for three Langmuir 2009, 25(10), 5631–5636
Malham and Bureau grafting experiments performed under the same nominal conditions, which indicates that the immersion/withdrawing conditions were fairly repeatable. In order to avoid as much as possible any bias due to surface gradients, all the results reported below (AFM images and contact angles for various temperatures and immersion times) correspond to measurements made over the central region of the samples. Mica Treatment. Muscovite mica (JBG-Metafix, France) samples of ∼1 cm 2 cm were cut from a large plate and subsequently cleaved on both faces, down to a thickness of a few tens of micrometers. Cleavage was performed in a laminar flow cabinet in order to prevent dust deposition on the surfaces. Mica sheets were then transferred to the glass vessel of a plasma reactor (plasma cleaning system model Femto from Diener Electronic GmbH, Germany). The reactor is composed of a cylindrical glass vessel (10 cm in diameter and 30 cm in length), two gas inlets (one port for argon and one for water vapor), and an RF generator operated at a frequency of 40 kHz and power adjustable between 20 and 100 W. The results presented hereafter were obtained using the following protocol: the vessel containing the mica samples was evacuated down to a pressure of 0.1 mbar, purged with H2O vapor for approximately 1 h at a flow rate of 100 cm3.min-1, pumped down to 0.1 mbar, and then filled with water vapor so as to reach a reduced pressure of 0.4 mbar. The surface of the mica samples were then exposed to the plasma generated at such a H2O pressure. In order to identify the proper conditions for plasma treatment, we have first systematically studied the effect of exposition time and RF power on roughness (measured by AFM) and surface activation. The latter was probed by exposing the plasma-treated mica to the vapor of 1,1,1,3,3,3-hexamethyldisilazane (99.9%, Sigma-Aldrich), at T = 80 °C and ambient pressure for 12 h (this yields a layer of trimethylsilane that can bind to the mica surface only in the presence of reactive groups). This preliminary study showed that exposition of mica for 5 min using a power setting of 80 W provided surfaces exhibiting both low roughness and sufficient chemical reactivity (see results below). All the results given in the article for OTE layers on plasma-activated mica have been obtained using such conditions. For comparison, we have also deposited OTE layers on untreated (freshly cleaved) mica and on steam-treated mica, using the method described by Schwartz et al.26 (mica was placed on top of a beaker containing boiling water, until water condensation occurred on the surface. The sample was then dried in an argon flow). AFM Imaging. The morphology of OTE submonolayers deposited under various conditions was observed by atomic force microscopy, with a commercial system (Dimension 3100, Veeco) operated in tapping mode using silicon tips (radius of curvature ∼10 nm). Image analysis and treatment was performed with the free software WSxM.39 All AFM images presented below have been set to the same height scale of 5 nm (from black [bottom] to white [top]). Contact Angle Measurements. Contact angles of water on OTE SAMs were measured on a home-built instrument. Each surface, placed on a horizontal holder, was brought into contact with a droplet of deionized water hanging from the vertical needle of a 10 μL microsyringe. The droplet volume was then increased (respectively decreased) by 3-5 μL, by pushing (respectively retracting) the plunger of the syringe. Video acquisition was performed, by means of a CCD camera and a long working-distance objective, during such an advancing/receding sequence of the contact line. Advancing (θA) and receding (θR) contact angles were measured using ImageJ software.40 For each sample, θA,R were measured on 3 to 5 different locations. (39) Horcas, I.; Fernandez, R.; Gomez-Rodriguez, J. M.; Colchero, J.; Gomez-Herrero, J.; Baro, A. M. Rev. Sci. Instrum. 2007, 78, 013705. (40) http://rsbweb.nih.gov/ij/.
Langmuir 2009, 25(10), 5631–5636
Article
Surface Forces Apparatus. The thickness of the SAMs deposited on mica can be inferred from the analysis of AFM images of incomplete monolayers. However, in order to obtain an independent determination for the thickness of complete monolayers, we have used the surface forces apparatus technique on two occasions (plasma-treated mica surfaces grafted at T = 18 and 22 °C, with an immersion time of 30 s). Such measurements were performed on an instrument recently developed in our laboratory,41 according to the following protocol: (i) Two micrometer-thick mica sheets were back-silvered and glued onto two cylindrical lenses having a radius of curvature ∼1 cm. (ii) A thin layer of mica was peeled off each sample using adhesive tape, and the cylindrical lenses were mounted, facing with their axis crossed at right angle, into the SFA in order to measure the thickness of each mica sheet. Such a measurement is based, as described in detail in refs 41-43, on the fringes of equal chromatic order (FECO) produced when shining white light on the Fabry-Perot cavity formed by the two reflective back surfaces of the mica sheets. (iii) Both lenses were then immediately transferred to the plasma reactor for surface activation. (iv) After plasma treatment, both lenses were immersed in the OTE solution, dried, and mounted again into the SFA, and the surfaces were brought into contact. (v) The thickness of the “bilayer” (one SAM on each mica surface) was then determined using the spectral shift of the FECO with respect to the situation where bare mica surfaces were in contact.
Results Plasma Activation of Mica. In the spirit of the extensive study performed by Liberelle et al.,36 we have first probed, as mentioned above in the Experimental Methods, the roughness and chemical reactivity of mica induced under various conditions of exposition time and RF-power for the plasma treatment. We find that a satisfactory compromise is obtained with 5 min exposure at 80 W and pH2O = 0.4 mbar. Under these conditions, we find that the root-mean-square (rms) roughness of plasma-exposed mica is 1.3-1.5 A˚ (measured over 1 1 μm2 AFM scans), only slightly above the ∼1 A˚ roughness exhibited by untreated mica. Vapor-phase deposition of trimethylsilane on such activated mica results in surfaces that are partially wetted by water and exhibit an advancing angle θA = 65° and receding angle θR = 30-35°. These values, measured after vapor-phase grafting, are found to be reproducible after sonication of the silanized mica surfaces for period of 30 min in successive solvents (toluene and ethanol). This indicates that the trimethylsilane layer deposited on plasma-activated mica is not washed away by rinsing and sonication, and is most probably covalently attached to the substrate. It shall be noted that the values we report here for θA and θR of water on a trimethylsilane layer are much lower than that obtained by Liberelle et al.,36 who measured advancing and receding angles of ∼95° and ∼70° on a similar surface. This shows that the trimethylsilane layer grafted following our procedure does not densely cover the mica surface. It suggests that the plasma treatment of mica peformed with our equipment produces a density of Si-OH groups at the surface which is lower than what was obtained in the work of Liberelle et al. However, we will show in what follows that (41) Bureau, L. Rev. Sci. Instrum. 2007, 78, 065110. (42) Israelachvili, J. N. J. Colloid Interface Sci. 1973, 44, 259. (43) Heuberger, M. Rev. Sci. Instrum. 2001, 72, 1700.
DOI: 10.1021/la804213q
5633
Article
Malham and Bureau
this is not a limitation for the production of high-quality SAMs of long alkyl chains. Submonolayer Morphologies. We find that submonolayers of OTE deposited on untreated or steam-treated mica exhibit an island-like morphology for short immersion time (Figure 1), and we observe that the surface coverage and the size of the OTE islands increase with increasing time, until a complete monolayer is formed (Figure 1a-c). Over the range of temperature explored, we note only a slight decrease, for a given immersion time, of the island size at the highest temperature (see Figure 1d-f). In contrast with the above situation, we find that varying the deposition temperature of OTE on plasma-treated mica has a strong impact on the morphology of the submonolayers. (i) It can be seen in Figure 2a,b that deposition temperatures below 18 °C result in OTE molecules organized into islands of height ∼25 A˚ (Figure 2f and Figure 3d) with respect to the background. Increasing the temperature from T = 10 °C to T = 15 °C is seen to result in a decrease of both the island size and the distance between them (Figure 2a,b). (ii) At T = 18 °C, the submonolayer exhibits small and almost fully interconnected domains coexisting with regions of lower thickness (Figure 2c). The height difference between such regions and the surrounding domains is ∼10 A˚. (iii) For grafting temperatures above 18 °C, AFM images suggest that OTE molecules form a continuous phase exhibiting “holes” of a few hundred nanometers in size (see Figures 2d,e and 4a). These holes correspond to zones of lower thickness (height difference ∼1 nm) compared to the continuous surrounding phase. Evolution of the submonolayer morphology with immersion time is illustrated in Figures 3 and 4. We observe that, for T < 18 °C, islands grow and merge to yield a complete monolayer (Figure 3a-c) as the immersion time is increased. For T g 18 °C, submonolayers exhibit fewer holes as grafting time increases (Figure 4). SAMs Stability and Wetting Properties. Stable measurements of advancing and receding contact angles of water on OTE SAMs deposited on untreated and steam-treated mica were not possible. Indeed, independent of the immersion time, hence of the surface coverage, we systematically observed that water droplets rapidly spread on such surfaces. Such droplet spreading is illustrated in Figure 5.
Figure 1. (a-c) Topographic AFM images (size 5 5 μm2) show-
ing the morphology of OTE layers deposited at T = 18 °C on untreated mica, for various immersion times: (a) 1 s, (b) 5 s, (c) 20 s. (d-f) Morphology of SAMs obtained for 1 s of immersion at (d) 15 °C, (e) 22 °C, (f) 30 °C.
5634
DOI: 10.1021/la804213q
OTE layers on plasma-treated mica exhibit contact angles with water which are stable in time. Figure 6a reports θA and θR measured on submonolayers deposited at T = 10, 15, 18, 22, and 30 °C using an immersion time of 1 s (SAM
Figure 2. AFM images (5 5 μm2) showing the morphology of OTE submonolayers obtained on plasma-treated mica with an immersion time of 1 s, at (a) T = 10 °C, (b) T = 15 °C, (c) T = 18 °C, (d) T = 22 °C, and (e) T = 30 °C. (f) Height profile measured from (a) along the line indicated on the image.
Figure 3. AFM images (5 5 μm2) of OTE layers deposited
on plasma-treated mica at T = 15 °C for various immersion times: (a) 1 s, (b) 5 s, (c) 30 s. (d) height profile measured along the line drawn on part b.
Langmuir 2009, 25(10), 5631–5636
Malham and Bureau
Article
Figure 4. AFM images (1 1 μm2) of OTE layers grafted on
plasma-treated mica at T = 22 °C for (a) 1 s and (b) 5 s. The height scale for these images is 1 nm from black to white.
Figure 6. Advancing (•) and receding (o) angles of water in contact with SAMs formed (a) for 1 s at various temperature; (b) at 15 °C for various immersion times. Error bars correspond to the dispersion observed on 3-5 locations on the surfaces.
Figure 5. Image sequence showing the spreading of a water droplet on an OTE SAM deposited on untreated mica (immersion time: 1 s). Time is stamped in the upper left corner of each image. Image field is 5.2 4.8 mm2. morphologies shown in Figure 2). We find that, even for such a short deposition time, θA already lies above 100°. Furthermore, it can be seen that both θA and θR go through a maximum at T = 15 °C, while the wetting hysteresis (Δθ = θA - θR) is a minimum at this temperature. Increasing the immersion time is found to result in an increase of θA and θR, as illustrated in Figure 6b. For deposition temperature of 15 and 18 °C, we measure θA = 111 ( 1° and Δθ = 12 ( 1° for immersion times larger than 30 s. These are the best results obtained in our study regarding hydrophobicity of the surfaces. OTE monolayers grafted at T > 18 °C are found to exhibit θA j 106-108° and a hysteresis on the order of 20°, even for immersion times of several tens of minutes. We find that SAMs deposited on plasma-treated mica resist sonication in toluene and ethanol (30 min in each solvent) and immersion in water (90 min): contact angles measured before and after such treatments displayed no discernible difference. This shows that OTE molecules are strongly attached to the substrate, and that robust monolayers can be formed on plasma-treated mica, as found previously.33,36
Discussion and Conclusions As far as the SAM morphology is concerned, our set of observations on untreated or steam-treated mica is fully consistent with previous studies of OTS25-27 (octadecyltrichlorosilane) or OTE28 on the same substrates. They indicate, as proposed in refs 26,27, that formation of SAMs on bare or Langmuir 2009, 25(10), 5631–5636
hydrated mica occurs by nucleation, growth, and coalescence of islands made of all-trans, close-packed molecules with their alkyl chains almost perpendicular to the solid surface. Such a growth of well-packed domains implies that individual OTE molecules, which adsorb onto the mica surface from the solution, can then easily diffuse laterally on the substrate in order to aggregate with already existing islands. It means that the headgroups of OTE molecules are only weakly bound to bare or hydrated mica, and that strong (covalent) links, if any, do not limit the molecular mobility on the surface. This is confirmed by the fact that OTE layers deposited on untreated or steam-treated mica display a rapidly decreasing contact angle with water (Figure 5), which indeed suggests that such SAMs adhere weakly to the substrate and are readily washed away by water. Such an observation is in qualitative agreement with that of Xiao et al.,28 who reported contact angles of water on OTE SAMs on mica which decayed, within 30 min, down to 20-30°. Note that, in contrast with our findings and those of Xiao et al.,28 stable contact angle measurements have been reported for OTS deposited on hydrated mica.26,27 Although we do not have a clear explanation for such a discrepancy, it might be that, in the case of OTS, the hydrolysis step of the highly reactive Si(Cl)3 headgroups, which produces HCl close to the mica surface, favors covalent binding with the substrate by means of an ion-exchange mechanism akin to that proposed by Carson and Granick.34 Now, by comparing the SAM morphologies shown in Figures 1 and 2, it is obvious that the presence of plasmagenerated active chemical groups on the mica surface affects the growth mechanism of OTE layers. Submonolayers obtained at deposition temperatures below 18 °C exhibit micrometer-sized domains of height ∼25 A˚, which is consistent, as observed on bare mica, with islands of well-packed alltrans OTE molecules separated by regions of bare substrate or low-density (gas-phase) silane. The observation of such an island-like structure, as well as its evolution with immersion time (Figure 3), indicates that diffusion of adsorbed molecules on the surface is still efficiently at play on plasma-treated mica for deposition temperature T < 18 °C. DOI: 10.1021/la804213q
5635
Article
Malham and Bureau
Increasing the temperature is observed to result in a decrease of island size and separation, accompanied by a decrease in height difference between high and low regions. For T g 18 °C, there is an uncertainty regarding whether the regions of lower thickness correspond to bare substrate or not. The absolute height of the brighter regions seen in Figure 2c-e therefore cannot be deduced from AFM images alone. Measurements performed using the surface forces apparatus show that SAMs deposited on plasma-treated mica at 18 and 22 °C, with 30 s immersion time, both have a thickness of 27 ( 2A˚. This shows that, at these temperatures, full monolayers are still formed of molecules having their alkyl chains quasi-perpendicular to the substrate. It suggests that, during the early stage of monolayer formation, brighter regions visible on AFM scans for SAMs grafted at T g 18 °C are formed of close-packed all-trans OTE molecules, and that regions of lower thickness correspond to a disordered, less dense phase. Such a picture is further supported by the fact that the height difference (∼1 nm) between higher and lower domains is consistent with the thickness difference between the so-called liquid-condensed (LC) and liquidexpanded (LE) phases reported for C18 silane layers in previous studies.9,20,44 We therefore conclude that the effect of grafting temperature for OTE on plasma-treated mica is qualitatively very similar to what is observed for OTS on silicon oxide: as T increases, one goes from the LC/G (liquid-condensed/gas) coexistence region to the LC/LE coexistence region of the submonolayer “phase diagram”.22 Moreover, the decrease in size of ordered domains as T is increased (Figure 2), which is much less marked on bare mica (Figure 1), suggests that the probability of formation of covalent Si-O-Si bonds between the substrate and the silane headgroups is increased at higher temperatures. Such a bond formation mechanism is expected to strongly reduce the lateral mobility of OTE molecules, thus reducing the size of islands.
Such a temperature effect on submonolayer morphology manifests itself on the wettability of SAMs. Figure 5a shows that, for a given immersion time, there is an optimum in deposition temperature (T ∼ 15-18 °C), which yields high contact angles and low hysteresis. Although it was beyond the scope of this study to perform quantitative analysis of AFM images, it seems that such an optimum does not result from a maximum in total surface coverage, but rather from a maximum in the fractional area of well-ordered domains. In such a picture, in which the higher the temperature, the lower the lateral mobility of adsorbed molecules (hence the fractional area of ordered domains), one would expect that complete monolayers grafted above 18 °C would display more “surface defects” (holes or LE regions) than those elaborated at low temperature. This is indeed consistent with our mesurements of θA = 112° and Δθ = 10° on complete SAMs grafted at T e 18 °C, whereas Δθ stays on the order of 20° for full monolayers deposited at room temperature and above. In conclusion, we have performed a qualitative AFM study of incomplete SAMs of octadecyltriethoxysilane on plasmatreated mica surfaces. We show that the grafting temperature has a marked influence on monolayer morphology and quality. Contact angle measurements have allowed us to identify that high-quality SAMs were obtained at T = 15 °C and indicate that SAMs elaborated at room temperature probably exhibit more defects in the molecular arrangement of the alkyl chains. Our results suggest that silane monolayers grafted at low temperature on mica may be suitable for use in the surface force apparatus and allow for reliable measurements of hydrophobic interactions. Finally, our results show that it is possible, by adjusting the grafting time or temperature, to build incomplete SAMs which are robustly attached to the mica substrate and exhibit coverage defects of various size. Backfilling such partial layers with a second type of molecule having a different chemical endgroup45 might be a route to make surfaces with controlled chemical heterogeneities for studies of friction based on the SFA technique.
(44) Schneider, J.; Dori, Y.; Tirrell, M.; Sharma, R. Thin Solid Films 1998, 327-329, 772.
(45) Kumar, N.; Maldarelli, C.; Steiner, C.; Couzis, A. Langmuir 2001, 17, 7789.
5636
DOI: 10.1021/la804213q
Langmuir 2009, 25(10), 5631–5636
