pubs.acs.org/Langmuir © 2009 American Chemical Society
Filled Nanoporous Surfaces: Controlled Formation and Wettability ‡ € Eyal Bittoun,† Abraham Marmur,*,† Mattias Ostblom, Thomas Ederth,‡ and Bo Liedberg‡ †
Department of Chemical Engineering, Technion- Israel Institute of Technology, 32000 Haifa, Israel, and ‡ Division of Molecular Physics, Department of Physics, Chemistry and Biology, Link€ oping University, SE-581 83 Link€ oping, Sweden Received May 13, 2009. Revised Manuscript Received July 21, 2009
The controlled filling of hydrophobic nanoporous surfaces with hydrophilic molecules and their wetting properties are described and demonstrated by using thiocholesterol (TC) self-assembled monolayers (SAMs) on gold and mercaptoundecanoic acid (MUA) as the filling agent. A novel procedure was developed for filling the nanopores in the TC SAMs by immersing them into a “cocktail” solution of TC and MUA, with TC in huge excess. This procedure results in an increasing coverage of MUA with increasing immersion time up to an area fraction of ∼23%, while the amount of TC remains almost constant. Our findings strongly support earlier observations where linear ω-substituted alkanethiols selectively fill defects (nanopores) in the TC SAM (Yang et al. Langmuir 1997, 12, 1704-1707). They also support the formation of a homogeneously mixed SAM, given by the distribution of TC on the gold surface, rather than of a phase-segregated overlayer structure with domains of varying size, shape, and composition. The wetting properties of the filled SAMs were investigated by measuring the most stable contact angle as well as contact angle hysteresis. It is shown that the most stable contact angle is very well described by the Cassie equation, since the drops are much larger than the scale of chemical heterogeneity of the SAM surfaces. In addition, it is demonstrated that contact angle hysteresis is sensitive to the chemical heterogeneity of the surface, even at the nanometric scale.
Introduction The preparation of self-assembled monolayers (SAMs) is currently a well-established, convenient technique to produce organic molecular thin films.1-3 Much attention has been given to SAMs of ω-substituted alkanethiols with the chemical structure HS(CH2)n-X that adsorb onto gold surfaces from a solution. The subscript n refers to the number of methylene units in the chain and X refers to different chemical head groups such as -CH3, -OH, -COOH, or poly(ethylene glycol). Once adsorbed, they exhibit highly ordered, oriented, and densely packed structures.4-7 The interfacial properties and, in particular, the wettability of the surface of these SAMs can be varied from hydrophilic to hydrophobic by changing only the functional headgroup. A useful, convenient way to control the wetting properties of such surfaces is by formation of bicomponent SAMs of alkanethiols with different head-groups. This can be done by, e.g., preparing mixed solutions at desired molar ratios of alkanethiols.8-12 The structure and the composition of the SAMs
determine the wettability of the surfaces. The areas of application of such mixed SAM surfaces are widespread: coatings to resist/ promote biomaterials adhesion,13-15 biochemical sensors,16-18 electron transfer,19-21 and many more. Mixed SAMs of thiocholesterol (TC), cholest-5-ene-3β-thiol (Figure 1a), and a fatty acid, 11-mercaptoundecanoic acid (MUA, Figure 1b) are of special interest because they serve as good model systems10,21 for studying the role of cholesterol in biomembranes, cell growth, and gene expression.22,23 The molecular structure of TC is a nearly planar polycyclic steroid ring with a branched alkyl chain. Yang et al. were the first to prepare and characterize TCþMUA SAMs on gold; it was observed that TC SAMs are characterized by defects (nanopores) of 5-8 A˚, which make the covered area fraction only approximately 65% of that of dense alkanethiols SAMs.24 The reason for this low area coverage was attributed to the size and irregular geometric shape of the TC molecule,√which √ prevent the formation of hexagonally closepacked ( 3x 3)R30° overlayer structures as in “conventional” alkanethiols SAMs on Au(111) surfaces. The properties of mixed
*Corresponding author. E-mail:
[email protected]. (1) Poirier, G. E.; Pylant, E. D. Science 1996, 272, 1145–1148. (2) Ulman, A. An Introduction to Ultrathin Organic Films: From Langmuir-Blodgett to Self-Assmbly; Academic Press: Boston, 1991. (3) Schreiber, F. Prog. Surf. Sci. 2000, 65, 151–256. (4) Porter, M. D.; Bright, T. B.; Allara, D. L.; Chidsey, C. E. D. J. Am. Chem. Soc. 1987, 109, 3559–68. (5) Chidsey, C. E. D.; Liu, G. Y.; Rowntree, P.; Scoles, G. J. Chem. Phys. 1989, 91, 4421–3. (6) Strong, L.; Whitesides, G. M. Langmuir 1988, 4, 546–58. (7) Dubois, L. H.; Zegarski, B. R.; Nuzzo, R. G. J. Chem. Phys. 1993, 98, 678– 88. (8) Ulman, A.; Evans, S. D.; Shnidman, Y.; Sharma, R.; Eilers, J. E. Adv. Colloid Interface Sci. 1992, 39, 175–224. (9) Ulman, A. Thin Solid Films 1996, 273(1-2), 48–53. (10) Yang, Z.; Engquist, I.; Wirde, M.; Kauffmann, J.-M.; Gelius, U.; Liedberg, B. Langmuir 1997, 13, 3210–3218. (11) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M.; Deutch, J. J. Phys. Chem. 1994, 98, 563–71. (12) Bertilsson, L.; Liedberg, B. Langmuir 1993, 9, 141–9.
12374 DOI: 10.1021/la9016992
(13) Luk, Y.-Y.; Kato, M.; Mrksich, M. Langmuir 2000, 16, 9604–9608. (14) Hederos, M.; Konradsson, P.; Liedberg, B. Langmuir 2005, 21, 2971–2980. (15) Jon, S.; Seong, J.; Khademhosseini, A.; Tran, T.-N. T.; Laibinis, P. E.; Langer, R. Langmuir 2003, 19, 9989–9993. (16) Guiomar, A. J.; Guthrie, J. T.; Evans, S. D. Langmuir 1999, 15, 1198–1207. (17) Frederix, F.; Bonroy, K.; Laureyn, W.; Reekmans, G.; Campitelli, A.; Dehaen, W.; Maes, G. Langmuir 2003, 19, 4351–4357. (18) Briand, E.; Salmain, M.; Herry, J.-M.; Perrot, H.; Compere, C.; Pradier, C.M. Biosens. Bioelectron. 2006, 22, 440–448. (19) Finklea, H. O.; Avery, S.; Lynch, M.; Furtsch, T. Langmuir 1987, 3, 409–13. (20) Henderson, J. I.; Feng, S.; Ferrence, G. M.; Bein, T.; Kubiak, C. P. Inorg. Chim. Acta 1996, 242, 115–24. (21) Yang, Z.; Engquist, I.; Liedberg, B.; Kauffmann, J.-M. J. Electroanal. Chem. 1997, 430, 189–195. (22) Yeagle, P. L. Biochim. Biophys. Acta 1985, 822, 267–87. (23) McMullen, T. P. W.; Lewis, R. N. A. H.; McElhaney, R. N. Curr. Opin. Colloid Interface Sci. 2004, 8, 459–468. (24) Yang, Z. P.; Engquist, I.; Kauffmann, J. M.; Liedberg, B. Langmuir 1996, 12, 1704–7.
Published on Web 08/21/2009
Langmuir 2009, 25(20), 12374–12379
Bittoun et al.
Article
equations were suggested for this purpose:30,31
Figure 1. Space-filling models and molecular structures of the molecules of (a) TC and (b) MUA.
SAMs of TC and MUA have also been investigated by coadsorption of both molecules from ethanol solution on a gold surface at different molar ratios.10 It was shown for these SAMs that the surface coverage of MUA was always higher than that of the TC molecules, even at a TC molar fraction of 0.9 in the incubation solution. However, the orientation of the rigid TC molecules remains the same in the mixed SAMs as in the pure TC SAM, whereas the structure of MUA molecules in the mixed SAMs was significantly disordered compared with a pure SAM of MUA. The wettability of surfaces is usually characterized by their contact angles (CAs) with various liquids, especially with water. This characterization can be done in two ways: (a) by measuring the most stable apparent CA, and (b) by measuring the CA hysteresis range, i.e., the difference between the advancing and receding CAs.25,26 The most stable apparent CA (to be referred to, for brevity, as the most stable CA) represents the state associated with the lowest Gibbs energy of the system. The Gibbs energy curve of a drop on a chemically heterogeneous surface exhibits multiple minima points at various apparent CAs that represent metastable states.25,27,28 In order to move from one local minimum to another and eventually get to the most stable state, an external energy source (e.g., mechanical vibrations) should be supplied to the system to overcome the energy barriers between the metastable states. The most stable CA is an important wettability measure, since it is related to the Young CA, from which the surface tension of the solid surface can be assessed through the Cassie equation:29
cos θms1 ¼ ðcos θadv þcos θrec Þ=2
ð2Þ
θms2 ¼ ðθadv þθrec Þ=2
ð3Þ
However, the usage of these equations has not been theoretically substantiated so far. The CA hysteresis range characterizes the heterogeneity of a surface in general. For smooth surfaces, it specifically reflects the chemical heterogeneity. High values of the CA hysteresis range imply nonuniformity in the chemical composition of the SAMs. Surfaces with minimal CA hysteresis can be very useful, for example, in the prevention of adhesion of drops to surfaces.32,33 In the present paper, a novel approach is used to fill up the nanopores in a TC SAM on a gold surface by MUA molecules, in a way that avoids replacement of TC. It is suggested that the present approach of using a TC SAM as a template in combination with a tuned filling procedure offer an attractive platform for the preparation of homogeneously mixed SAMs of TC and MUA where phase segregation into nano- or microscale domains of different composition is less likely to occur. Thus, we do not believe that domains similar to those formed by coadsorption of ω-substituted alkanethiols on gold are formed in our TC/MUA SAM (cf. the pioneering scanning tunneling microscopy (STM) work by Stranick et al.34). Further, we believe that neither STM nor atomic force microscopy (AFM) would give an unequivocal answer to the nanoporosity of the TC SAM because of the tilted orientation of the TC molecules in the SAM10,24 and the hydrophobic dangling tail, which most likely will block the defects (pores) in the TC SAM, making them inaccessible to the STM/ AFM tip. The wetting properties of such SAMs are characterized by measuring the most stable CA as well as the advancing and receding CAs using a drop shape analysis system.
Experimental Section Gold Substrate Preparation. Gold substrates were prepared
The subscript C stands for Cassie, ηi is the area fraction that is occupied by a SAM of type i, and θYi is the Young CA of this SAM had it been the only component that covers the surface in a ideal way. In order to accomplish meaningful measurement and interpretation of the most stable CA with the Cassie equation, the size of the drop should be much larger than the scale of the chemical heterogeneity.28 For mixed SAMs, the Cassie equation can always be used, since the size of any macroscopic drop is orders of magnitude larger than the size of the molecular domains in the SAMs. As an alternative to the direct measurement of the most stable CA, θms, it can be roughly estimated by averaging of the measured advancing CA, θadv, and receding CA, θrec. Two different
as follows: silicon wafers (100) were cut into 20 mm 40 mm pieces, and cleaned in TL1, which is a mixture of Milli-Q water (>18 MΩ cm), 30% hydrogen peroxide (Merck), and 25% ammonia (Merck) (5:1:1 by volume), at 80 °C for 10 min, and thoroughly rinsed in Milli-Q water. The cleaned silicon pieces were mounted in an electron-beam evaporation system (Balzers UMS 500P) and were coated first by a Ti layer with a thickness of 10-25 A˚, at an evaporation rate of 1 A˚/s, and subsequently by a gold layer of 2000 A˚, at a rate of 10 A˚/s. The base pressure was always less than 2 10-9 Torr, and the pressure during evaporation was held below 2 10-7 Torr. Self-Assembled Monolayers. The SAMs were prepared from ethanol (Kemetyl, Sweden, 99.5%) solutions with thiols in micromolar concentrations, by incubation in polypropylene beakers (Nalgene) at room temperature. TC and MUA powders (Aldrich) were dissolved in ethanol by ultrasonication for 5 min. Gold surfaces were cleaned twice in TL1 and were rinsed in MilliQ water. The TC SAMs were formed by immersion of the clean gold surface in the solution and incubation overnight. Subsequently, the TC SAM surfaces were rinsed in ethanol and ultrasonicated for 5 min to remove physisorbed thiol molecules, and then incubated again in mixed MUA/TC solutions at differ-
(25) Marmur, A. Soft Matter 2006, 2, 12–17. (26) Meiron, T. S.; Marmur, A.; Saguy, I. S. J. Colloid Interface Sci. 2004, 274, 637–644. (27) Marmur, A. J. Colloid Interface Sci. 1994, 168, 40–6. (28) Marmur, A.; Bittoun, E. Langmuir 2009, 25, 1277–1281. (29) Cassie, A. B. D.; Baxter, S. Tran. Faraday Soc. 1944, 40, 546–51.
(30) Andrieu, C.; Sykes, C.; Brochard, F. Langmuir 1994, 10, 2077–80. (31) Decker, E. L.; Garoff, S. Langmuir 1996, 12, 2100–10. (32) Krasovitski, B.; Marmur, A. Langmuir 2005, 21, 3881–3885. (33) Pierce, E.; Carmona, F. J.; Amirfazli, A. Colloids Surf. 2008, 323, 73–82. (34) Stranick, S. J.; Parikh, A. N.; Tao, Y. T.; Allara, D. L.; Weiss, P. S. J. Phys. Chem. 1994, 98, 7636–46.
cos θC ¼
X
X ηi cos θYi , ð ηi ¼ 1Þ
i
Langmuir 2009, 25(20), 12374–12379
i
ð1Þ
DOI: 10.1021/la9016992
12375
Article
Bittoun et al.
ent concentrations and times. Finally, the surfaces with TC/MUA SAMs were immersed into HCl solution of 0.01 M for 1 min, rinsed in Milli-Q water, ultrasonicated again for 5 min, dried under a stream of nitrogen, and immediately analyzed. The area fraction of the TC SAM on the gold surface was calculated by integration of the peak intensities of the IR spectrum, similarly to the method previously used,24 and then from
ηTC ¼
ATC ATCþMUA
"
ATCþMUA 1AMUA
# ð4Þ
where ATC and ATCþMUA are the integrated intensities between 2950 and 3000 cm-1 for the SAMs of pure TC and TC posttreated with MUA, respectively. AMUA and ATCþMUA are the integrated peaks between 1620 and 1780 cm-1 for the SAMs of pure MUA and TC post-treated with MUA, respectively. Infrared Reflection-Absorption Spectroscopy. Infrared reflection-absorption (IRAS) measurements of the SAMs were performed using a Bruker IFS66 system with a liquid-nitrogencooled MCT detector. The spectra were recorded by averaging 3000 interferograms (10 min) at 2 cm-1 resolution. The measurement chamber was continuously purged with nitrogen gas during the measurements. A spectrum of deuterated hexadecanthiol on gold was used as a reference. Ellipsometry. Single-wavelength ellipsometry was performed on a Rudolph AutoEL ellipsometer. The light source was a He-Ne laser with a wavelength of 632.8 nm at an angle of incidence of 70°. The optical constants of the gold surfaces were determined immediately after cleaning in TL1 twice, rinsing, and drying in N2. The optical constants were inserted in a three-phase model, air/organic film/gold, assuming an isotropic transparent organic layer having n = 1.5. The thickness was calculated by the Rudolph AutoEL ellipsometer software as an average value of five different spots. Contact Angle Measurements. The advancing and receding CAs of distilled water in ambient atmosphere at 23 °C were measured with a drop shape analysis system (DSA100, KRUSS, Germany) in air. Water was continuously added from a blunt-ended needle attached to a syringe at a rate of 7 μL/s up to a drop maximum volume of 100 μL to measure the advancing CA. The receding CA was measured by continuously removing water at a rate of 7 μL/s. Side view images were taken automatically at a rate of 2 frames/s. The CAs were calculated by the system using the tangent method fitting. This procedure gave a reproducibility of (1°. The most stable CAs were measured as follows: a surface was placed horizontally on the center of the DSA100 stage and fastened with two clips. A drop of 80 μL was then gently deposited on the surface by a micrometric syringe. Horizontal vibrations were applied to the system from a DC-micromotor device (FAULHABER, Germany) that was connected to the stage at a maximum 3.5 V and 0.3 A. The amplitude and frequency of the plate were 0.2 mm and 13 Hz, respectively. It was assumed that the vibrations made the drop reach its global minimum in the Gibbs energy when the CA stopped varying. A necessary (but not sufficient) indication for reaching this state is getting an axisymmetric drop. Therefore, a top view image of the drop was taken by a charge-coupled device (CCD) video camera. The image was analyzed using Image-Pro software to calculate the largest crosssection area and the roundness of the drop. Only axisymmetric (round) drops were considered for assessing the most stable CA. The criterion used for acceptable axisymmetry was26
Figure 2. IRAS spectra in (a) the C-H stretching region and (b) the fingerprint region of pure TC, TCþMUA, and MUA monolayers on gold.
Results and Discussion
ð5Þ
Controlled Filling of Nanoporous TC Surfaces with MUA. In order to measure the area fraction of TC, ηTC, for a pure TC SAM on a gold surface, the following procedure was applied.24 Three types of SAMs were prepared: pure SAMs of TC and MUA, and, for comparison, a TC SAM immersed in ethanol solution of 100 μM MUA overnight. The reflectionabsorption spectra of pure TC, pure MUA, and TCþMUA SAMs are shown in Figure 2a for the C-H stretching region and in Figure 2b for the fingerprint region. For a pure TC SAM, all peaks in the regions of 3000-2800 cm-1 and 15001450 cm-1 are in agreement with previous results,24 indicating an ordered monolayer. In the spectra of the MUA SAM, the peaks at 2850 and 2920 cm-1 refer to the symmetric and antisymmetric methylene stretches of the chain, indicative of a densely packed SAM.4,35,36 Integrated peak areas from the IRAS spectra and eq 4 were used to calculate the area fraction of TC: ηTC = 0.77. As mentioned above, this area fraction is significantly lower than that of regular SAMs of alkanethiols, and is the result of both the irregular shape of the TC molecule and a mismatch between the size of TC molecules and the adsorption site spacing on the gold substrate.24 The resulting SAM is a nanoporous surface with pore sizes of about 5-8 A˚. These pores may be filled selectively with alkanethiols, whereas bulkier molecules are prevented from adsorption by size-exclusion. This was demonstrated by Yang in a previous study, where the pores in TC SAMs were filled in a controlled manner with
where S and d are the area and diameter of the drop, respectively. This CA was calculated by a computer program that fits the Young-Laplace equation to the measured radius, volume of the drop, water surface tension, and density.
(35) Smith, E. L.; Alves, C. A.; Anderegg, J. W.; Porter, M. D.; Siperko, L. M. Langmuir 1992, 8, 2707–14. (36) Nuzzo, R. G.; Dubois, L. H.; Allara, D. L. J. Am. Chem. Soc. 1990, 112, 558–69.
Rd ¼ 4S=πd 2 g0:95
12376 DOI: 10.1021/la9016992
Langmuir 2009, 25(20), 12374–12379
Bittoun et al.
MUA molecules in order to yield a variation in the wetting properties of the mixed monolayers.10,21 To obtain well-defined mixed monolayers, the goal is to maintain the area fraction of TC on the surface constant, while the area fraction of MUA increases. This should be done until all available pores are filled, but without replacement and concomitant phase segregation of MUA on the surface. In an attempt to achieve this, gold surfaces with pure TC SAMs were incubated in ethanol solutions of MUA at concentrations ranging from 1 μM to 1 mM, for different times. The IRAS spectra of these TC/MUA SAMs indicated that the area fraction of TC significantly decreased relative to that of the pure TC SAM, even for immersion times shorter than two minutes. These results were confirmed by two methods: (a) by comparing the asymmetric CH3 stretching mode intensities at 2963 cm-1 of the MUA-filled TC SAM with the pure TC SAM, and (b) by calculating the area fraction of MUA on the surface from the spectrum obtained by subtraction of the pure TC spectrum from the MUAþTC spectrum. The phenomenon of replacement of TC by MUA can be explained as follows: When a TC SAM is immersed in an ethanol solution of MUA, it may, at first, be expected that the TC remains adsorbed to the surface because of its low solubility in ethanol. However, as previously discussed,10,21 the cross-sectional area of a MUA molecule is smaller by about a factor of 2 than the cross sectional-area of a TC molecule. This implies that MUA molecules yield about twice the thiolate/gold chemical bonding energy per unit gold area. The energy of forming the chemical bonding is dominant compared with the interaction energy of the TC with ethanol, therefore an exchange between MUA with TC on the gold surface is observed. The pairwise interaction energies MUA-MUA > TC-MUA ∼ TC-TC also favors the formation of a MUA-rich SAM. Thus, the stimulated replacement of TC by MUA must be prevented in order to form a homogeneously mixed SAM that is controlled by the defect distribution in the TC SAM. To avoid replacement of TC by MUA, and to maintain the area fraction of TC approximately constant at its initial value while filling the pores with MUA, a different approach is required. In the used protocol, the TC SAMs were immersed in mixed solutions of TC and MUA at various concentration ratios, where the concentration of TC in the solutions was always higher by about an order of magnitude than the concentration of MUA. With such TC-rich solutions, the exchange between MUA and TC was greatly reduced, because a TC molecule, which for some reason is removed, could readily be replaced by another TC molecule from solution. This effect is due to the weak interaction of TC with ethanol (as opposed to MUA, which forms strong hydrogen bonds with ethanol), which encourages its adsorption to the surface. The optimal way of filling the TC nanopores with MUA was achieved by immersing the TC surfaces in ethanol solution of 100 μM TC and 5 μm MUA for 60 min. The area fraction of MUA on the surface was obtained by subtracting the spectrum of pure TC from the spectrum of the TCþMUA SAM. The area fractions of TC and MUA and the ellipsometric thicknesses, d, for various immersion times are shown in Table 1. It is assumed that the area not occupied by either TC or MUA is simply occupied by air. The SAM thicknesses are between 16 and 17 A˚ for all immersion times, in good agreement with values obtained from molecular modeling, assuming a tilt angle of about 30° with respect to the surface normal.37 (37) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R. G. J. Am. Chem. Soc. 1989, 111, 321–35.
Langmuir 2009, 25(20), 12374–12379
Article Table 1. Area Fractions and Ellipsometric Thicknesses Obtained after Filling of MUA into the Pores of a TC SAM from a Solution of 100 μM TC and 5 μM MUA in Ethanol immersion time, min
ηTC
ηMUA
ηair
d, A˚
0 2 5 10 20 40 60
0.77 0.758 0.750 0.739 0.730 0.726 0.718
0 0.078 0.131 0.189 0.199 0.221 0.223
0.230 0.164 0.119 0.072 0.071 0.053 0.059
17.0 ( 0.1 17.0 ( 0.3 16.6 ( 0.5 16.7 ( 0.3 16.7 ( 0.6 16.6 ( 0.5 16.6 ( 0.4
Figure 3. The filling processes of the nanopores in the TC SAM: (9,0) experimental values of the area fraction of TC, ηTC; (2,Δ) experimental values of the area fraction of MUA, ηMUA. Filled symbols: incubation of TC SAMs in ethanol solution of 5 μM MUA þ 100 μM TC; empty symbols: incubation of TC SAMs in ethanol solution of 100 μM MUA only. The cartoons in the figure represent the cases of a pure TC SAM before the immersion in the solution (up), nanopores of TC SAM being filled with MUA (right), and phase segregation of MUA following replacement of TC by MUA molecules (left).
On the basis of the IRAS experimental results, Figure 3 emphasizes the difference between the formation of a TC/MUA SAM by immersing in an ethanol solution of MUA that is rich in TC and that by immersing in a solution of MUA only. The thick solid lines through the filled symbols (9, 2) represent the filling process where nanopores in the TC SAM are filled by MUA using an ethanol solution of 100 μM TC and 5 μM MUA up to 60 min. The area fraction of MUA increases with time, meaning that nanopores are filled with MUA molecules, while the area fraction of TC on the surface remains almost constant. The dashed line at ηTC = 0.77 represents the ideal situation where the area fraction of TC on the surface remains constant during the filling process. The difference between the experimental and ideal lines of the filling process is relatively small. The thin solid lines in Figure 3 through the empty symbols (0, Δ) demonstrate the replacement process of TC by MUA that occurs when the TC SAM is immersed in an ethanol solution of 100 μM MUA up to 20 min. In this case, the TC molecules desorb from the surface after a relatively short immersion time, and islands of MUA of different size, shape, and composition may form on the surface. DOI: 10.1021/la9016992
12377
Article
Bittoun et al.
Table 2. The Measured Most Stable CAs, θms, Advancing CAs, θadv, Receding CAs, θrec, Calculated Cassie CAs, θC, and Most Stable CAs Based on Eqs 2 and 3 of Water on SAMs for Various TC and MUA Area Fractions on Gold Surfacesa ηMUA
θms, deg
θadv, deg
θrec, deg
θC, deg
(2) θeq ms , deg
(3) θeq ms , deg
hysteresis range, deg
0 103 ( 2 107 ( 1 92 ( 1 103* 99.4 99.5 15 ( 1 0.078 95 ( 1 103 ( 2 88 ( 1 94.8 95.5 95.5 14 ( 1 0.131 89 ( 2 97 ( 1 83 ( 2 89.3 90.0 90.0 14 ( 1 0.189 86 ( 2 95 ( 2 81 ( 2 83.4 88.0 88.0 14 ( 2 0.199 84 ( 1 89 ( 2 77 ( 1 82.7 83.0 83.0 12 ( 1 0.221 81 ( 1 84 ( 1 73 ( 1 80.4 78.6 78.5 9(1 0.223 76 ( 3 78 ( 1 71 ( 2 80.7 74.5 74.5 7(1 1 14 14 ( 1 14 ( 1 14* 14 14.0 0þ1 a The values of θC for pure TC and MUA were marked with an asterisk since they are a priori assumed to be identical to the measured most stable CA.
The filling of the nanopores with MUA, at almost constant TC coverage (77%) and upright orientation is very different from the mechanism proposed by Fuhrhop et al., who suggested that a similar steroid with the thiol group located at the center of the steroid structure assembles flat on gold at low coverage. They used different agents, e.g., n-alkanethiols and diols/sugars as coadsorption/filling agents to study the electrochemical blocking characteristics.38 A flat orientation and low coverage is not consistent with our thickness and IR experiments, and is therefore abandoned as a possible model for the orientation and nanomorphology of TC on gold in our system. Wettability of MUA-Filled Nanoporous TC Surfaces. The wettability of the mixed, TC/MUA SAM surfaces was characterized experimentally by two different methods: 1 Measurement of the most stable CA, θms, using a horizontally vibrating plate.26 This CA is a characteristic of the surface tension of the solid surface, and can be related to the individual CAs of the pure SAMs of TC and MUA through the Cassie equation. 2 Measurement of the advancing CA, θadv, and receding CA, θrec, from which the hysteresis range (the difference between θadv and θrec) was calculated. The hysteresis range is a measure of the surface chemical heterogeneity. In order to use the Cassie equation for the mixed SAMs of TC/MUA, it is required to know their individual Young CAs. To calculate the Young CA of TC, θYTC, (had it been possible to form a monolayer of it without holes), it was assumed that the TC SAM surface is composed of two chemical species: TC and air. The Cassie equation for this surface is then given by cos θms ¼ ηTC cos θYTC -ηair
ð6Þ
Since air is completely hydrophobic, its “contact angle” is considered to be 180°. The values of ηTC and ηAir are given in Table 1, and the value of θms is given in Table 2. Substituting these values in eq 6 leads to a value of θYTC = 89.6°. In the case of a pure SAM of MUA, θadv and θrec are equal; therefore it can be concluded that this surface behaves as an ideal one, namely, completely smooth and chemically homogeneous. This conclusion is supported by the IR data for MUA, which indicate that a MUA SAM is a highly ordered and closely packed monolayer.36 Thus, according to the values in Table 2, θYMUA = 14°. The Cassie equation can now be used to calculate the most stable CA for surfaces of TC SAMs filled with MUA. As explained above, the surface is actually composed of three (38) Fuhrhop, J.-H.; Bedurke, T.; Gnade, M.; Schneider, J.; Doblhofer, K. Langmuir 1997, 13, 455–459.
12378 DOI: 10.1021/la9016992
Figure 4. The ratio of the measured and calculated most stable CAs, θms, to the calculated Cassie CA, θC, versus the area fraction of MUA. (0) θms obtained experimentally by inducing horizontal vibrations; (]) θms calculated from eq 2; (4) θms calculated from eq 3. The dotted line emphasizes the location of equality between the experimental values and the Cassie CA.
chemical species: TC, MUA, and air. Therefore, the Cassie equation for these surfaces is given by cos θC ¼ ηTC cos θYTC þηMUA cos θYMUA -ηair
ð7Þ
The results of these calculations are given in Table 2. It can be seen that most of them are very close to the values of the measured most stable CA, the highest difference being about 6%. This is expected, since the drop size is much larger (more than 3 orders of magnitude) than the length scale of the chemical heterogeneity.28 It is also of interest to compare the most stable CAs obtained from eqs 2 and 3 with the most stable CAs obtained in the direct measurement using the vibrating plate. The measured advancing and receding CAs and the calculated values of the most stable (2) eq (3) CAs using eqs 2 and 3, θeq ms and θms respectively, are also given in Table 2. Figure 4 summarizes the above results by presenting the ratio θms /θC as a function of the area fraction of MUA on the surface, ηMUA. The dashed line in the figure represents the equality line between θms and θC. The error bars refer to the standard deviation in the measurements of the most stable CAs. The results show that the deviation from this line is in most cases smaller when using the direct measurement of the most stable CA by the vibrating plate than when calculating it using eqs 2 and 3. Practically, these results indicate that the measured most stable CA is the same as the calculated Cassie CA. These results confirm the reliability of this method of measurement. The differences between the ratios Langmuir 2009, 25(20), 12374–12379
Bittoun et al.
Article
macroscopic measure) is sufficiently sensitive to detect chemical heterogeneity at the nanometric scale.
Summary and Conclusions
Figure 5. The hysteresis range versus the area fraction of MUA.
θms/θC calculated by using eqs 2 and 3 are very small for the whole range of ηMUA. However, it should be remembered that, so far, these equations do not have a theoretical basis; therefore, it is better to measure θms experimentally, if possible. Figure 5 shows the hysteresis range as a function of the area fraction of MUA on the surface. As the MUA molecules fill the nanopores in the TC SAM, the chemical heterogeneity of the surface decreases because the surface contains more MUA and less air (the difference in surface tension between TC and MUA is smaller than between TC and air). Therefore, as the surface becomes more chemically uniform, the hysteresis range decreases.39 Remembering that the hysteresis range for a MUA monolayer is practically zero, the important conclusion from these measurements is that the CA hysteresis range (which is a (39) Marmur, A. Adv. Colloid Interface Sci. 1994, 50, 121–41.
Langmuir 2009, 25(20), 12374–12379
The following points summarize the achievements and conclusions of the present work: A controlled method of filling the nanoporous, hydrophobic TC SAM surface with hydrophilic MUA molecules was successfully developed. This method produces a nonsegregated mixed SAM on a gold surface. The filling process was optimized in order to maintain the area fraction of TC on the surface almost constant, while the area fraction of MUA increases up to about 23%. The most stable CAs (that are associated with the global minima in the Gibbs energy) of the mixed SAM surfaces were directly measured using the horizontally vibrating plate method. It was shown that these measured most stable CAs are strongly correlated to the CAs predicted by the Cassie equation. Other methods to determine the most stable CA were discussed. It was demonstrated that CA hysteresis measurements are sufficiently sensitive to indicate chemical heterogeneity of the surface at the nanometric scale. Acknowledgment. The authors acknowledge the support from the AMBIO (Advanced Nanostructured Surfaces for the Control of Biofouling) project (NMP-CT-2005-011827) funded by the European Commission’s sixth Framework Programme. Views expressed in this publication reflect only the views of the authors, and the Commission is not liable for any use that may be made of information contained therein. The authors also acknowledge the support from the Swedish Council (VR). The authors also wish to thank Magnus Falk for assistance with the sample preparation, and Dr. Hossam Haick for the use of his laboratory facilities in preparing the SAMs for the CA measurements.
DOI: 10.1021/la9016992
12379
