Effect of Leaving Group on the Structures of Alkylsilane SAMs

Nov 19, 2014 - Recruiting physisorbed water in surface polymerization for bio-inspired materials of tunable hydrophobicity. S. Oyola-Reynoso , I. D. T...
4 downloads 3 Views 3MB Size
Article pubs.acs.org/Langmuir

Effect of Leaving Group on the Structures of Alkylsilane SAMs Vikrant V. Naik, Roman Stad̈ ler, and Nicholas D. Spencer* Department of Materials, ETH Zurich, Vladimir-Prelog-Weg 5, 8093 Zurich, Switzerland S Supporting Information *

ABSTRACT: Multiple transmission and reflection (MTR) infrared spectroscopy has been used to study the kinetics of the formation of self-assembled monolayers (SAM) of octadecylsilanes with different leaving groups, viz. trichloro, trimethoxy, and triethoxy. It was observed that the chlorosilanes form much denser and crystalline-like SAMs and ethoxysilanes form thin SAMs, while methoxysilanes form extremely thin SAMs. The high sensitivity of the MTR IR technique allows the molecular conformations of the alkyl chains and appearance/disappearance of the silanol groups to be scrutinized in detail. This enables the formulation of models for the structures of the SAMs that are in many ways different than the classical picture of silanes on oxide surfaces. We observe that the structure of SAMs depends on the rate of hydrolysis of the leaving groups and thus their chemical nature. SAMs of chlorosilanes resemble a structure of snow moguls or densely packed umbrellas. SAMs of ethoxysilanes, on the other hand, look like stacks of fallen trees, while the molecules of the ultrathin methoxysilane SAMs are lying nearly parallel to the surface, resembling creepers.



INTRODUCTION SAMs of various functionalities have been used for many industrial purposes, such as hydrophobization, in applications including sensors, MEMs, and NEMs.1−3 A large variety of molecules consisting of end-functionalized alkyl chains have been used to form SAMs, with binding groups including thiols,4,5 phosphates,6 phosphonates,7 catechols,8 and silanes.9−11 However, SAMs formed from silanes offer some distinctive advantages. Silanes are known to functionalize a number of practically important oxide substrates, including silica, titania, and alumina.10,12 Silanes are also known to form covalent bonds with the surfaces and with each other, thereby forming coatings that are stable over long periods of time.10,13,14 Developments in the characterization of thin-film organic coatings on surfaces have closely followed upon the heels of advancements in thin-film fabrication technology. Recently we have applied one such method, multiple transmission− reflection infrared (MTR IR) spectroscopy,15 to study the formation and organization of SAMs of octadecyltrichlorosilane (OTCS) on silica.16 In that study we reported that alkylsilanes assemble on surfaces in such a way that not every alkyl chain is anchored to the surface. This structure differs from what has traditionally been described in the “classical model” of silane SAMs. Instead, we proposed that the alkylsilanes initially oligomerize prior to adsorption, and only a few chains actually bind to the surface, forming a structure that resembles snow moguls or closely packed umbrellas. We further stated that this structure could explain certain discrepancies that the classical model cannot account for particularly the volume constraints of alkyl chains. For a SAM © 2014 American Chemical Society

formed by OTS molecules, considering a typical Si−O bond length of 1.8 Å and a standard Si−O−Si angle of 105°, the maximum available space for the alkyl chains corresponds to a cylinder of diameter 2.9 Åa value much lower than the van der Waals diameter for an all-trans alkyl chain (4.9 Å). In our previous communication we suggested that, in addition to achieving more space by forming moguls, the chains could in principle form a gauche defect at C1−C2 torsion, similar to the behavior of phospholipids, thereby allowing the chains to maintain a parallel configuration while the Si−O−Si plane is distorted. In the present study, we have investigated the kinetics of formation of SAMs for three different leaving groups, viz. chloro, ethoxy, and methoxy. The idea behind this study was to test the validity of the proposed model and to determine whether the rate of hydrolysis of silanes plays a significant role in the structure of the SAM. It is generally accepted that chlorosilanes hydrolyze faster than methoxysilanes, which in turn are faster to react than ethoxysilanes.10 However, in the course of the present investigation, it was observed that octadecyltriethoxysilane forms denser SAMs than octadecyltrimethoxysilane. This will of course depend on experimental conditions, in particular the solventdecalin in this case.



MATERIALS AND METHODS

Octadecyltrichlorosilane (OTCS), octadecyltrimethoxysilane (OTMS), and octadecyltriethoxysilane (OTES) were obtained from Received: September 18, 2014 Revised: November 14, 2014 Published: November 19, 2014 14824

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831

Langmuir

Article

Aldrich (purity >90%) and used without any further purification. Dilute (3 mM) solutions of all three silanes were prepared in freshly distilled decahydronaphthalene (decalin) (cis−trans mixture) (SigmaAldrich) to coat 30 × 18 mm2 ⟨111⟩ double-sided, polished silicon wafers (Si-Mat, Germany). Before surface coating, the wafers were cleaned thoroughly by treating them with piranha (3:1, 98% H2SO4, Sigma-Aldrich, and 30% H2O2, Merck) solution for 20 min. The wafers were then rinsed in deionized water, dried, and oxygen-plasma (high power, Harrick Plasma Cleaner/Sterilizer, Ithaca, NY) treated for 2 min. The silicon wafers were then immersed in the three different silane solutions for different specified periods of time from 10 s to 1 day, in order to obtain the silane-coated surfaces. The wafers were then cleaned by sonicating them in toluene and deionized water and then dried under a stream of dry nitrogen. A major problem for silanes is the reproducibility of film thickness. Our efforts to devise a reproducible protocol are described in the Supporting Information. The formation of a SAM was confirmed by variable-angle spectroscopic ellipsometry (VASE) (M-2000FTM, J. A. Wollam Inc., Lincoln, NE) and static-contact-angle measurements (Model 100, Ramé Hart Inc.). The film thickness was measured as a difference between the optical thickness of a blank silicon wafer and its thickness after coating with the silanes. The data were evaluated using WVASE32 software (WexTech Systems, Inc., New York). For water and hexadecane contact-angle measurements, 3 μL of solvent was used. The AFM measurements were performed on a Dimension 3000 (Veeco Metrology Group, Santa Barbara, CA) using Olympus AC-160 tips in tapping mode (resonant frequency: between 200 and 400 kHz). MTR FTIR measurements were performed on a home-built MTR setup (Prof. Shoujun Xiao, University of Nanjing, China). The doublepolished Si wafer was placed between two gold mirrors, 2 mm apart, such that its distance from the two mirrors was 0.3 and 1.2 mm, and the simplified number of transmission passes was six. The spectra were acquired on a Bruker IFS66v FTIR spectrometer in vacuum at a resolution of 4 cm−1, measured using a mercury cadmium telluride (MCT) detector, cooled by liquid nitrogen. A bare, piranha-cleaned, and oxygen-plasma-treated Si wafer was used as background. Each sample and background measurement consisted of 256 scans. A grid polarizer placed between the sample and the source was used to obtain the polarized spectra.

Figure 1. Plot of the ellipsometric thickness of OTCS-, OTES-, and OTMS-derived layers as a function of immersion time in the respective silane.

Figure 2. 1 × 1 μm2 tapping mode AFM images of SAMs of (i) OTMS and (ii) OTES after 6 h of immersion time.



RESULTS AND DISCUSSION Ellipsometry. Figure 1 shows the thickness of OTCS, OTMS, and OTES SAMs determined by ellipsometry as a function of the immersion time. All the curves show a rapid increase at lower immersion times, followed by saturation. What is striking is that both the alkoxysilanes show a much lower final thickness than the chlorosilane. The formation kinetics of OTES SAMs slow down considerably after around 2 h at 0.8 nm thickness, followed by a slow increase to 0.9 nm at 24 h. OTMS, which is generally expected to react faster than OTES, resulting in a thicker SAM, however, saturated at a thickness of only 0.5 nm after 4 h. SAMs of OTCS, on the other hand, had a final saturation thickness of 2.5 nm after only about 30 min. The OTS SAM thickness of 2.5 nm at full coverage corresponds to the octadecyl chain length.17 The OTCS chains are thus expected to be oriented almost orthogonal to the surface. In contrast, the SAMs of OTES and OTMS are much thinner. This could be due to a greater tilting of the chains or a sparsely packed SAM or a combination of both. It should be noted that ellipsometry averages the measurement over a large area (3 × 10 mm2) and thus does not give a direct orientation of individual chains. Using simple trigonometric calculations, by considering the length, l, of an octadecylsiloxyl (C18−Si−O)

Figure 3. Plot of water and hexadecane contact angles for OTCS, OTES, and OTMS as a function of immersion time.

chain as 2.6 nm and using the saturated ellipsometric thicknesses, d, the tilt angles of these chains from the surface plane, θ, can be evaluated (θ = sin−1(d/l)). The estimated values are, 11°, 18°, and 74° for OTMS, OTES, and OTCS, respectively. AFM. Figure 2 shows tapping-mode atomic force microscope (AFM) images of the saturated SAMs of OTES and OTMS. Both the wafers show smooth surfaces, with RMS roughness of around 0.25−0.3 nm, which is similar to the RMS roughness of the uncoated silicon wafer. This indicates a 14825

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831

Langmuir

Article

prior to hydrolysis, then attaching to the surface, orienting, and oligomerizing, once hydrolysis has occurred. The hexadecane contact angle is sensitive to the orientation of CH3 groups, due to the very different hexadecane wetting properties of methylene and methyl.19 After 4 h, the hexadecane CAs are 10° for OTES and 7° for OTMS, compared to a value of 35° for OTCS.16 These low values indicate that the chains are less crystalline for OTES and OTMS than in the case of OTCS. The results also indicate that the terminal methyl groups do not show a preferential orientation. The small values of hexadecane CA, however, do show that the surface is rendered oleophilic by silanization. Similarly to the water contact angle, the hexadecane contact angle for the OTES SAM shows a slight rise after 15 h, again indicating a slight straightening/orientational change of the alkyl chains, particularly of the terminal methyl group. MTR FTIR Spectroscopy. The MTR IR spectra were recorded as a function of immersion time in the region of 800− 4000 cm−1 for the three silanes. The region below 1400 cm−1 is dominated by the SiO2 of the silicon wafer and hence will be ignored for rest of the discussion. Here we discuss the alkylstretching region (3000−2800 cm−1) and the hydroxylstretching region (3800−3000 cm−1). MTR IR gives a direct measurement of the surface coverage and crystallinity of the alkyl chains. Figure 4 shows a plot of the alkyl-stretching region of the fully developed SAMs (immersion time of 24 h) for the three silanes. It is immediately clear from the three spectra that not only does the spectrum of OTCS show higher intensities, but the lower wavenumber peakpositions of the CH2 symmetric and asymmetric stretches and narrower line widths of the two bands indicate that the OTCS SAMs are better ordered than OTMS and OTES. The OTCS spectrum also shows a well-defined CH3 symmetric band at 2879 cm−1, which indicates the formation of a dense SAM. Another interesting feature is the resolution of the CH3 asymmetric peak at 2960 cm−1. Both in-plane and out-of-

Figure 4. Comparative spectra of OTCS, OTES, and OTMS SAMs after 24 h immersion time.

uniform SAM without agglomerates, similar to previously reported AFM images of OTCS.16 Contact-Angle Measurements. Water contact-angle measurements (Figure 3) show a similar trend to the ellipsometry results. The SAMs show an initial rise in CA, followed by saturation. Water contact angles at 4 h are 78° for OTES and 74° for OTMS. These absolute values are much lower than the 104° for OTCS.18 This trend is, however, in agreement with the lower SAM thickness of the alkoxysilanes. It is interesting to note the changes in the water contact angle for the ethoxysilanes. The water contact angle appears to have saturated at a value of 80° by 12 h. However, after 15 h, the contact angle again shows a jump and rises to a value of 90° by the end of 24 h. This change is not reflected in the film thickness of the OTES SAM, which appears to saturate much earlier (5 h). We believe that this change in contact angle could be due to surface rearrangement of the silane molecules on the surface. This could be a result of slow hydrolysis of the ethoxy group, which would lead to the SAMs arriving at the surface

Figure 5. MTR IR spectra in the alkyl-stretching region on a silica surface as a function of immersion time for (a) OTES and (b) OTMS SAMs. (c) Plot of νs and νas frequencies as a function of immersion time for OTCS, OTES, and OTMS SAMs. 14826

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831

Langmuir

Article

Figure 6. MTR IR spectra as a function of immersion time recorded under (a) s-polarization and (b) p-polarization. (c) Plot of dichroic ratio of the OTCS, OTES, and OTMS SAMs as a function of immersion time.

asymmetric stretches increase with immersion time, providing a quantitative measure of surface coverage and again illustrating the sensitivity of the MTR IR technique. The spectra of OTMS are not so well resolved because of the lower surface coverage. Methylene-stretching peaks could be quantified only for longer immersion times. Nevertheless, saturation is visible as the intensities at 6 and 8 h are similar, but the 24 h spectrum shows a further increase in intensity, as observed for OTES. The methyl (CH3) asymmetric peak at 2960 cm−1 is visible at higher coverages. The less intense symmetric stretch at 2879 cm−1 only appears at higher immersion times for the two SAMs. The methylene symmetric and asymmetric regions are known to be extremely sensitive to the arrangement of the alkyl chains in assemblies. In the case of crystalline n-alkane chains, where the methylene units of the alkyl chains exist in an all-trans conformation, the symmetric and asymmetric stretching modes appear in the range of 2846−2849 and 2916−2918 cm−1, respectively.20 At higher temperatures, as the number of gauche conformations increases, these bands shift to higher wavenumbers. In the case of molten alkanes, for example, the symmetric and antisymmetric stretching modes typically appear around 2856−2858 and 2924−2928 cm−1, respectively.21 The methylene symmetric and asymmetric stretching frequencies as a function of immersion time are plotted in Figure 5c for the three silanes. It can be seen that the observed peak positions for both the symmetric and antisymmetric modes, for all immersion times, fall between those of the molten alkanes and the crystalline alkanes. It is interesting to note how the crystallinity of the alkyl chains changesnot just as a function of immersion time but also as a function of the leaving group. In case of OTCS, for example, the effect is more pronounced at extremely low immersion times, even as short as 10 s, there is a distinct presence of alkyl chains assembled on the surface, albeit in a disordered state. In the case of OTES, the effect is less pronounced and it is least clear for OTMS. While OTCS chains become more rigid by 1 h of immersion, the alkyl chains of OTES and OTMS appear to be very mobile and less crystalline, even after 24 h. The saturation values of the frequency shifts show small variations in the intensities and

Figure 7. MTR IR spectra of the hydroxyl stretching region for (a) OTES and (b) OTMS SAM as a function of immersion time. (c) A plot of hydroxyl peak heights as a function of immersion time.

Scheme 1. Dynamic Equilibrium of Siloxane Linkages As Suggested by Maoz et al.

plane modes are clearly visible, which highlights the sensitivity of MTR IR compared to other methods for silica surfaces. MTR IR spectra of the OTES and OTMS SAMs in the methylene-stretching region are shown in Figure 5 as a function of immersion time. The peak intensities of both symmetric and 14827

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831

Langmuir

Article

Scheme 2. Proposed Mechanism for the Formation of OTMS, OTES, and OTCS SAMs Based on MTR IR Results

increases. However, under s-polarization, they seem to increase quite dramatically after 12 h. In the case of OTCS, these changes were seen at times as low as 30 min. This would indicate that the chains become a bit more orthogonal to the surface. In the case of OTMS, this change is not as drastic. In fact, under s-polarization, only at 24 h does the intensity under s-polarization show any dramatic change. This indicates that the OTMS chains lie almost parallel to the surface. It also appears that, unlike the OTCS chains, OTES and OTMS chains are more disordered. OTES, however, becomes more ordered after about 12 h of immersion while OTMS chains remain disordered throughout. Sagiv and co-workers have indicated that OTCS chains tend to become crystalline over time.9,23

frequency shifts of the CH2 symmetric and asymmetric stretches long after the point at which ellipsometric measurements saturate. One of the advantages of the MTR IR method is the possibility to compare spectra taken under s- and p-polarization. Figure 6 shows changes in the intensity as a function of immersion time when observed under s- and p-polarization, denoted by As and Ap, respectively, for OTES and OTMS. It is interesting to note the changes in intensities of the methylene stretching bands under polarization, which can be used to gather information about the orientation and quantification of chain ordering.22 It can be seen that under p-polarization the intensities seem to increase only slightly as immersion time 14828

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831

Langmuir

Article

result of on-surface oligomerization of two or more chains resulting in more oriented structures. The changes in surface organization of the SAMs can be studied by following the changes in the hydroxyl-stretching region. Figures 7a and 7b show the changes in the OH stretching region of OTES and OTMS chains, respectively, as a function of immersion time. The OH band consists of two peaks: the silanol, C18−Si−OH peak and the surface Sisurf−OH peak.24 These peaks occur at 3150 and 3500 cm−1, respectively. In some cases, a small peak is seen ∼3475 cm−1. This peak, in principle, could arise from surface silanols that are slightly redshifted because of hydrogen bonding with alkylsilanols. In the case of OTMS in Figure 7c, there is a clear increase in the intensity of the silanol peaks as a function of immersion time. This indicates that the alkyl chains of OTMS consist of C18− Si−OH peaks and hence do not directly attach to the surface. This is not very surprising. In their paper, Wen et al. have clearly demonstrated that silane monolayers need not attach covalently to the surface for stability.25 They can do so through hydrogen bonds to the surface. The case of OTES molecules is rather peculiar. The silanol concentration is much higher for times until 6 h, as seen from the higher values of absorbance for these times. At 8 h, the intensity begins to drop considerably (about 40% decrease). At longer times of 12 h and over, this silanol intensity drops considerably to a much lower value, indicating that the number of silanol units has decreased drastically. It should be noted that this time is consistent with the increase in the values of water contact angle and, more importantly, the dichroic ratio. Another important observation is the relative intensity of the silanol peaks for OTMS and OTES. It appears that OTES SAMs at all coverages contain approximately 50% less silanol groups than OTMS SAMs, despite having almost double the dry alkyl chain mass, as indicated by ellipsometric and IR data. This, in turn, could be a result of two possible scenarios: (i) The ethoxysilanes oligomerize faster, or (ii) because of their slower rates of hydrolysis, the chains retain a large number of ethoxy groups, which ultimately translates into a smaller number of silanol groups. It should be recalled that in the case of OTCS the number of silanols was extremely small because of the faster rate of oligomerization of OTCS chains.16 In the first scenario, in which the OTES chains oligomerize faster, this would result in a structure similar to the reported snow-mogullike structure of OTCS chains.16 Such a structure results in extremely dense and crystalline packing of alkyl chains due to the volume constraints inherent in the oligomerized silane structure.16 However, it is evident from the IR measurements that the chains of OTES SAMs are neither crystalline nor ordered. This rules out the possibility of an oligomerized structure. As stated in the Introduction, the rate of hydrolysis is lowest for ethoxysilanes,10 and this is also one of the more important reasons for their major use in industry.26 It would thus appear that these OTES molecules indeed exist in their partially hydrolyzed ethoxy state for initial immersion times of up to 8 h. After this, the chains would probably start to oligomerize and consequently covalently attach to the surface, thereby further decreasing their silanol concentration. This oligomerization and surface attachment would also ensure that the chains become more orthogonal to the surface, as reflected by the increase in their dichroic ratio. This structure would in principle be similar to that described by Sagiv et al., who indicate the possibility of a dynamic equilibrium between the siloxane linkages and the silanols as depicted in Scheme 1.14,25

Here we show that under the conditions of our experiments this is true only for OTCS and not for OTES and OTMS chains. The s- and p-polarized spectra can be further used to calculate the dichroic ratio. The dichroic ratio, defined as the ratio of absorbance in s- and p-polarization (As/Ap), can be effectively used to calculate the tilt angles and orientation of the alkyl chains.15 This quantity is more sensitive to the changes in tilt angles than frequency shifts. A plot of dichroic ratio as a function of immersion time is shown in Figure 6c for the three SAMs. When the chains are oriented orthogonally on the surface, methylene stretching vibrations absorb more strongly in s- than in p-polarized light, leading to an increase in dichroic ratio. Here again all curves show an increase, followed by saturation, but the absolute values for OTES and OTMS are much lower than those of OTCS. The final value for OTMS is about 0.5. It is interesting to note the change in the dichroic ratio of OTES chains. After appearing to saturate by 12 h to a value of about 0.55, the dichroic ratio shows a rapid increase from 12 h until 24 h, by which point it has reached a value of 0.91 for the symmetric stretch and 0.75 for the asymmetric stretch. This trend is not very different from that of the water contact angle, shown in Figure 3. While the water contact angle does not rise from 18 to 24 h, the dichroic ratio continues to rise, indicating that the chains continue their surface rearrangement over a long period of time. This is also clear from the ellipsometry data (Figure 1), which show a saturation much earlier. However, unlike the case of OTCS, which displays a steep jump in the hexadecane contact angle, OTES chains show only a gradual change. This would mean that despite the changes in their orientation with respect to the surface normal, the chains are still disordered and thin. In the case of OTCS, the chains are extremely crystalline, resulting in the terminal CH3 methyl groups being oriented away from the surface and leading to a higher hexadecane contact angle. The molecularlevel information obtained by IR spectroscopy is thus able to explain macroscopic changes in the structure of SAMs. The dichroic ratio can be used to calculate tilt angles of alkyl chains.15 In the present article, tilt angle, θ, is defined as the angle between Si−C18 vector of the alkyl chain and the surface. Since the dichroic ratio depends only on the direction of the dipole moment of the methylene groups, the value of tilt angle obtained is independent of the azimuthal orientation of the molecules. The values of dichroic ratio of OTMS chains indicate a tilt angle of around 20°. Simple geometric calculations, using the ellipsometric thickness and lengths of octadecyl chains, indicate a tilt angle of 11°. These small differences arise due to a number of factors. The dichroic ratio calculates the tilt angle as an ensemble average. IR data indicate that the alkyl chains of OTMS are disordered, and hence, the tilt axis is not well-defined. Ellipsometry, on the other hand, averages the thickness over a larger macroscopic area. Ellipsometry does not consider surface reorganization. Ellipsometry, for example, cannot differentiate between a highly oriented but a thin SAM and a dense but disordered organic film, leading to the difference in the estimated tilt angle. The final SAM of OTMS could therefore have some intermediate structure between random and crystalline states with highly tilted alkyl chains. The values of dichroic ratio for the chains of OTES SAMS indictaes an initial tilt angle of 20°, which then jumps to a value of about 50° at the end of 24 h. This dramatic change, without considerable change in the ellipsometric thickness, could be a 14829

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831

Langmuir

Article

(2) Srinivasan, U.; Houston, M. R.; Howe, R. T.; Maboudian, R. Alkyltrichlorosilane-Based Self-Assembled Monolayer Films for Stiction Reduction in Silicon Micromachines. J. Microelectromech. Syst. 1998, 7, 252−260. (3) Ewers, B. W.; Batteas, J. D. The Role of Substrate Interactions in the Modification of Surface Forces by Self-Assembled Monolayers. RSC Adv. 2014, 4, 16803−16812. (4) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M. Self-Assembled Monolayers of Thiolates on Metals as a Form of Nanotechnology. Chem. Rev. 2005, 105, 1103−1169. (5) Ulman, A. Formation and Structure of Self-Assembled Monolayers. Chem. Rev. 1996, 96, 1533−1554. (6) Liu, H.-B.; Venkataraman, N. V.; Spencer, N. D.; Textor, M.; Xiao, S.-J. Structural Evolution of Self-Assembled Alkanephosphate Monolayers on TiO2. ChemPhysChem 2008, 9, 1979−1981. (7) Pawsey, S.; Yach, K.; Reven, L. Self-Assembly of Carboxyalkylphosphonic Acids on Metal Oxide Powders. Langmuir 2002, 18, 5205−5212. (8) Dalsin, J. L.; Hu, B.-H.; Lee, B. P.; Messersmith, P. B. Mussel Adhesive Protein Mimetic Polymers for the Preparation of Nonfouling Surfaces. J. Am. Chem. Soc. 2003, 125, 4253−4258. (9) Maoz, R.; Sagiv, J. On the Formation and Structure of SelfAssembling Monolayers. I. a Comparative Atr-Wettability Study of LangmuirBlodgett and Adsorbed Films on Flat Substrates and Glass Microbeads. J. Colloid Interface Sci. 1984, 100, 465−496. (10) Arkles, B. Tailoring Surfaces with Silanes. CHEMTECH 1977, 7, 766−778. (11) Subramanian, V.; van Ooij, W. J. Silane Based Metal Pretreatments as Alternatives to Chromating. Surf. Eng. 1999, 15, 168−172. (12) De Graeve, I.; Vereecken, J.; Franquet, A.; Van Schaftinghen, T.; Terryn, H. Silane Coating of Metal Substrates: Complementary Use of Electrochemical, Optical and Thermal Analysis for the Evaluation of Film Properties. Prog. Org. Coat. 2007, 59, 224−229. (13) Eckstein, Y. Role of Silanes in Adhesion. Part I. Dynamic Mechanical Properties of Silane Coatings on Glass Fibers. J. Adhes. Sci. Technol. 1988, 2, 339−348. (14) Maoz, R.; Sagiv, J.; Degenhardt, D.; Möhwald, H.; Quint, P. Hydrogen-Bonded Multilayers of Self-Assembling Silanes: Structure Elucidation by Combined Fourier Transform Infra-Red Spectroscopy and X-Ray Scattering Techniques. Supramol. Sci. 1995, 2, 9−24. (15) Liu, H.-B.; Venkataraman, N. V.; Bauert, T. E.; Textor, M.; Xiao, S.-J. Multiple Transmission−Reflection Infrared Spectroscopy for High-Sensitivity Measurement of Molecular Monolayers on Silicon Surfaces. J. Phys. Chem. A 2008, 112, 12372−12377. (16) Naik, V. V.; Crobu, M.; Venkataraman, N. V.; Spencer, N. D. Multiple Transmission-Reflection IR Spectroscopy Shows That Surface Hydroxyls Play Only a Minor Role in Alkylsilane Monolayer Formation on Silica. J. Phys. Chem. Lett. 2013, 4, 2745−2751. (17) Wasserman, S. R.; Tao, Y. T.; Whitesides, G. M. Structure and Reactivity of Alkylsiloxane Monolayers Formed by Reaction of Alkyltrichlorosilanes on Silicon Substrates. Langmuir 1989, 5, 1074− 1087. (18) Janssen, D.; De Palma, R.; Verlaak, S.; Heremans, P.; Dehaen, W. Static Solvent Contact Angle Measurements, Surface Free Energy and Wettability Determination of Various Self-Assembled Monolayers on Silicon Dioxide. Thin Solid Films 2006, 515, 1433−1438. (19) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M. Self-Assembled Monolayers of Alkanethiols on Gold: Comparisons of Monolayers Containing Mixtures of Short- and Long-Chain Constituents with Methyl and Hydroxymethyl Terminal Groups. Langmuir 1992, 8, 1330−1341. (20) MacPhail, R. A.; Strauss, H. L.; Snyder, R. G.; Elliger, C. A. Carbon-Hydrogen Stretching Modes and the Structure of N-Alkyl Chains. 2. Long, All-Trans Chains. J. Phys. Chem. 1984, 88, 334−341. (21) Snyder, R. G.; Strauss, H. L.; Elliger, C. A. Carbon-Hydrogen Stretching Modes and the Structure of N-Alkyl Chains. 1. Long, Disordered Chains. J. Phys. Chem. 1982, 86, 5145−5150.

They have also indicated that this kind of oligomerization would include 2−4 alkylsilane chains.25 Scheme 2 represents the possible mechanisms for the formation of the OTMS, OTES, and OTCS SAMs.



CONCLUSION We have used MTR IR to probe the effect of leaving group on the formation and structure of self-assembled monolayers of alkylsilanes. For comparison, we used three silanesoctadecyltrimethoxy, octadecyltriethoxy, and octadecyltrichloro silanes which differ only in their leaving groups. We have found that the leaving group has a significant effect on the formation of SAMs. Although the main reason for this is the difference in their rate of hydrolysis, the final structure of the SAMs is considerably different. In the case of OTMS, the SAMs mostly consist of individual hydrolyzed alkylsilane chains. These chains lie almost parallel to the surface and bind to the surface through partial covalent and hydrogen bonds. These SAMs are extremely sparse, lack orientation, and are noncrystalline. These chains could be compared to vines or creepers lying flat on the surface. The OTES SAMs form a structure not very different from that of OTMS at lower immersion times. These chains however are not fully hydrolyzed and still contain ethoxy groups. After long immersion times (12 h and over), they tend to oligomerize (approximately 2−4 chains) and probably attach to the surface through covalent bonds. This results in chains attaining greater orientation and denser packing, rendering the surface more hydrophobic, despite the increase in the number of alkylsilane chains being negligible. The SAMs of OTES can be likened to a stack of fallen trees after a heavy storm. Finally, the OTCS chains, we believe, oligomerize before they attach to the surface. At this stage we cannot comment on the exact degree of their oligomerization. However, we believe that these SAMs, similar to OTES SAMs, have a dynamic equilibrium of siloxane linkages, resulting in making and breaking of Si−O−Si bonds. OTCS SAMs are formed within very short times and are extremely dense and crystalline. This SAM would resemble a stack of densely packed umbrellas or snow moguls.



ASSOCIATED CONTENT

S Supporting Information *

(S1) Reproducibility of experiments. This material is available free of charge via the Internet at http://pubs.acs.org.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected] (N.D.S.). Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS The authors thank Drs. Norman Blank, Andreas Kramer, and Wolf-Ruediger Huck, SIKA Technology AG, for valuable discussions and input and SIKA Technology AG, Zurich, Switzerland, for the financial support for this project via the ETH Zurich Foundation.



REFERENCES

(1) Maboudian, R.; Carraro, C. Surface Chemistry and Tribology of MEMS. Annu. Rev. Phys. Chem. 2004, 55, 35−54. 14830

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831

Langmuir

Article

(22) Picard, F.; Buffeteau, T.; Desbat, B.; Auger, M.; Pézolet, M. Quantitative Orientation Measurements in Thin Lipid Films by Attenuated Total Reflection Infrared Spectroscopy. Biophys. J. 1998, 76, 539−551. (23) Cohen, S. R.; Naaman, R.; Sagiv, J. Thermally Induced Disorder in Organized Organic Monolayers on Solid Substrates. J. Phys. Chem. 1986, 90, 3054−3056. (24) Tripp, C. P.; Hair, M. L. Direct Observation of the Surface Bonds Between Self-Assembled Monolayers of Octadecyltrichlorosilane and Silica Surfaces: a Low-Frequency IR Study at the Solid/ Liquid Interface. Langmuir 1995, 11, 1215−1219. (25) Wen, K.; Maoz, R.; Cohen, H.; Sagiv, J.; Gibaud, A.; Desert, A.; Ocko, B. M. Postassembly Chemical Modification of a Highly Ordered Organosilane Multilayer: New Insights Into the Structure, Bonding, and Dynamics of Self-Assembling Silane Monolayers. ACS Nano 2008, 2, 579−599. (26) Kallury, K. M.; Macdonald, P. M.; Thompson, M. Effect of Surface Water and Base Catalysis on the Silanization of Silica by (Aminopropyl) Alkoxysilanes Studied by X-Ray Photoelectron Spectroscopy and 13C Cross-Polarization/Magic Angle Spinning Nuclear Magnetic Resonance. Langmuir 1994, 10, 492−499.

14831

dx.doi.org/10.1021/la503739j | Langmuir 2014, 30, 14824−14831