Langmuir 2001, 17, 4329-4335 4329
A Study of Alkyl Chain Conformational Changes in
Self-Assembled n-Octadecyltrichlorosilane Monolayers on
Fused Silica Surfaces
Yi Liu, Lauren K. Wolf,† and Marie C. Messmer*
Department of Chemistry and Zettlemoyer Center for Surface Studies, Lehigh University,
7 Asa Drive, Bethlehem, Pennsylvania 18015
Received January 24, 2001. In Final Form: April 30, 2001
The adsorption and conformational changes of n-octadecyltrichlorosilane (OTS) self-assembled mono-
layers on fused silica surfaces are monitored by a nonlinear optical technique, sum-frequency generation
(SFG) spectroscopy, and lateral force microscopy (LFM). The effect of a small amount of water in the OTS
deposition solution on the alkyl chain conformation within the resulting OTS monolayers is also investigated.
Results show that the alkyl chain conformational changes of OTS films on the surface of fused silica occur
in three stages. The initial stage involves OTS adsorption from solution and the beginning of island forma-
tion. During this stage, the alkyl chains within the film are almost completely disordered, though a significant
increase in surface coverage occurs. The second stage shows dramatic changes in alkyl chain conformation
within the film from disordered gauche conformation to an ordered mainly all-trans conformation, while
only a small increase in surface coverage occurs. The final stage is a much slower adsorption process.
Although the surface coverage increases from 90% to a complete monolayer, only a slight increase in the
SFG band intensities is observed during this stage. The early stages of OTS adsorption and the kinetic
effect of water in the solvent on the resulting OTS monolayers are also studied by LFM. Island structures
are observed, and the correlation of SFG, LFM, and contact angle measurements suggests that both
uniform growth (monomer deposition) and island growth (aggregate deposition) occur during the OTS
adsorption. The effect that water in the deposition solution has on each of these mechanisms is discussed.
Introduction remains unclear. It is known that trace amounts of water
1-3
present either on the substrate surfaces or in the deposition
Alkylsiloxane self-assembled monolayers (SAMs) are solution can hydrolyze the -Si-Cl groups to -Si-OH
of great importance both fundamentally as model self- groups, which then undergo condensation with the surface
assembly systems4 and practically as surface modifiers in or with adjacent monomers.3 When the alkyl chains are
chromatographic stationary phases,5,6 antibody immobi- sufficiently long, the van der Waal’s interactions between
lization for biosensors,7 adhesion promoters for polymer alkyl chains combined with the lateral condensation of
films,8 boundary lubricants,9,10 semiconductor coatings,11 adjacent silanol groups allow formation of high-quality
and nanolithography.12,13 The trifunctional alkyltrichlo- films from these compounds on a variety of substrates
rosilanes such as OTS are unique in that they undergo such as silicon, mica, glass, and gold.9,16-18
lateral cross-linking as well as direct bonding to the To form a highly ordered organosilane SAM, the reaction
substrate.14,15 Although these films achieve a high degree conditions need to be carefully controlled.14,19 The influence
of coverage and stability due to this lateral cross-linking, of precursor concentration,20 solvent,21 surface hydra-
the mechanism of film formation with these compounds tion,11,22 substrate cleanliness,23 relative humidity,24 and
water content25-27 has been extensively studied. What
* To whom correspondence should be addressed. E-mail: mcm6@ emerges from these studies is the critical role that reaction
lehigh.edu. conditions used in film preparation seem to play in
† Current address: Department of Chemistry, Boston University,
determining not only film quality but also the growth
590 Commonwealth Ave., Boston, MA 02215.
(1) Sagiv, J. J. Am. Chem. Soc. 1980, 102, 92. mechanism. A critical temperature of ∼28 °C, above which
(2) Ulman, A. Chem. Rev. 1996, 96, 1533.
(3) Ulman, A. Introduction to Ultra-thin Organic Films: From (16) Parikh, A. N.; Schivley, M. A.; Koo, E.; Seshadri, K.; Aurentz,
Langmuir Blodgett to Self-Assembly; Academic Press: San Diego, CA, D.; Mueller, K.; Allara, D. L. J. Am. Chem. Soc. 1997, 119, 3135.
1991. (17) Allara, D. L.; Parikh, A. N.; Rondelez, F. Langmuir 1995, 11,
(4) Stevens, M. J. Langmuir 1999, 15, 2773. 2357.
(5) Wirth, M. J.; Fatunmbi, H. O. Anal. Chem. 1993, 65, 822. (18) Kessel, C. R.; Granick, S. Langmuir 1991, 7, 532.
(6) Sander, L. S.; Wise, S. A. Anal. Chem. 1984, 56, 504. (19) Silberzan, P.; Leger, L.; Ausserre, D.; Benattar, J. J. Langmuir
(7) Willner, I.; Schlittner, A.; Doron, A.; Joselevich, E. Langmuir 1991, 7, 1647.
1999, 15, 2766. (20) Flinn, D. H.; Guzonas, D. A.; Yoon, R.-H. Colloids Surf. A 1994,
(8) Kurth, D. G.; Bein, T. Langmuir 1993, 9, 2965. 87, 163.
(9) Xiao, X.; Hu, J.; Charych, D. H.; Salmeron, M. Langmuir 1996, (21) McGovern, M. E.; Kallury, K. M. R.; Thompson, M. Langmuir
12, 235. 1994, 10, 3607.
(10) de Gennes, P. G. Rev. Mod. Phys. 1985, 57, 827. (22) Le Grange, J. D.; Markham, J. L.; Kurkjian, C. R. Langmuir
(11) Angst, D. L.; Simmons, G. W. Langmuir 1991, 7, 2236. 1993, 9, 1749.
(12) Komeda, T.; Namba, K.; Nishioka, Y. J. Vac. Sci. Technol. A (23) Bierbaum, K.; Grunze, M.; Baski, A. A.; Chi, L. F.; Schrepp, W.;
1998, 3, 1680. Fuchs, H. Langmuir 1995, 11, 2143.
(13) Jeon, N. L.; Finnie, K.; Branshaw, K.; Nuzzo, R. G. Langmuir (24) Fairbank, R. W. P.; Wirth, M. J. J. Chromatogr. A 1999, 2, 285.
1997, 13, 3382. (25) Vallant, T.; Kattner, J.; Brunner, H.; Mayer, U.; Hoffmann, H.
(14) Parikh, A. N.; Allara, D. L.; Azouz, I. B.; Rondelez, F. J. Phys. Langmuir 1999, 15, 5339.
Chem. 1994, 98, 7577. (26) Vallant, T.; Brunner, H.; Mayer, U.; Hoffmann, H.; Leitner, T.;
(15) Wasserman, S. R.; Tao, Y. T.; Whitesides, G. M. Langmuir 1989, Resch, R.; Friedbacher, G. J. Phys. Chem. B 1998, 102, 7190.
5, 1074. (27) Tripp, C. P.; Hair, M. L. Langmuir 1992, 8, 1120.
10.1021/la010123c CCC: $20.00 © 2001 American Chemical Society
Published on Web 06/14/2001
4330 Langmuir, Vol. 17, No. 14, 2001 Liu et al.
complete OTS coverage cannot be formed, has been process in which two pulsed laser beams, one in the IR,
observed.14 The critical role of water content in the ωIR, and the other in the visible, ωvis, are overlapped on
deposition solution has been demonstrated in the forma- a sample and generate an output at the sum of these two
tion of complete OTS monolayers.19,27 Generally, a thin incident frequencies, ωsum ) ωIR + ωvis. The resulting sum-
layer of water on the solid substrate is required to form frequency intensity from an interface is resonantly
a complete monolayer.22 It has been postulated that this enhanced if the IR frequency is in resonance with a
thin water layer on the substrate acts as a water-phase vibrational transition at the sample surface. Thus, with
analogue of Langmuir film formation,16 but water in the the help of a tunable IR laser, vibrational sum-frequency
bulk solvent also plays an important role on the OTS film spectra of molecules at the interface can be obtained. As
formation.26,28 It has long been noticed that a drop of water a result of its second-order dependence on the input field,
in the solvent will facilitate the OTS film formation, and SFG is forbidden in media possessing inversion symmetry
the mechanism might be explained in terms of larger within the electric-dipole approximation but allowed at
clusters of OTS molecules formed in solvent with higher an interface where the inversion symmetry is broken.
water content before reaching the substrate surface. In this paper, we examine the adsorption and the alkyl
Although the mechanism of OTS monolayer formation chain conformational changes of an OTS monolayer on
has been the focus of many studies, debate about it still fused silica by SFG. The kinetic effects of water content
exists in the literature. Some studies support the uniform of the adsorbate solution have also been probed. The
growth model which describes the alkyl chains of an growth mechanism of OTS monolayers on silica surfaces
incomplete OTS monolayer as uniformly distributing over is inferred from SFG spectroscopy, contact angle mea-
the substrate and having a disordered structure,15,29,30 surements, and lateral force microscopy (LFM) measure-
whereas other work suggests a heterogeneous mechanism ments.
involving aggregate deposition and island growth.31,32 Most
AFM studies of partial OTS monolayers on mica and silicon Experimental Section
wafers suggested an islandlike structure of the films.23,33-35 Materials. Octadecyltrichlorosilane (95%), hexadecane (99+%),
Although AFM provides information on the macroscopic chloroform (99.9%), carbon tetrachloride (99.9+%), squalane (2,6,-
structure of OTS film growth, it does not reveal molecular 10,15,19,23-hexamethyltetracosane, C30H62, 99%), and HYDRA-
structure and conformation changes within the film. The NAL-Composite 5 (for Karl Fischer titration) were purchased
molecular structure of complete OTS monolayers has been from Aldrich and were used without further purification. Sulfuric
studied by infrared techniques such as transmission acid and hydrogen peroxide were analytical grade and obtained
from EM Science. Acetone (99.6%) and methanol (99.9%) were
infrared spectroscopy (TIRS)14,17,20,36 and attenuated total HPLC grade and purchased from Fisher Scientific. Deionized
reflection infrared spectroscopy (ATR-IR),11,25,26 and the (DI) water was obtained with Barnstead NANOpure system until
conformation of the alkyl chains is qualitatively inferred final resistivity >15.8 mΩ/cm was reached. Fused silica slides
from these spectra.37,38 However, determining molecular were purchased from ChemGlass.
conformation from FTIR spectra involves extensive simu- Sample Preparation. Fused silica plates were cut into 15 ×
lation with careful selection of models and initial param- 30 mm slides and were soaked in chloroform for >3 h followed
eters.25,26 Therefore, the results may have large uncer- by rinsing with acetone, methanol, and a copious amount of DI
tainties and do not provide complete details on the water. The slides were then dried under high-purity nitrogen
conformational changes. and immersed in piranha solution (7:3 concentrated H2SO4/30%
H2O2) for 1 h at ∼110 °C. Caution: piranha solution is a highly
To circumvent this problem, we use a nonlinear reactive mixture and severely exothermic during reaction. It should
technique, sum-frequency generation (SFG), to investigate be kept out of contact with oxidizable organic material. The
the growth mechanism of OTS monolayers. SFG was substrates were then rinsed with a copious amount of water,
chosen mainly because it is highly surface selective, dried with nitrogen, and heated at 80 °C for a short time. The
resulting from its second-order dependence on the input clean substrates were completely wetted by water. The cleaning
field. The submonolayer sensitivity of this technique process was performed less than 3 h before monolayer preparation
makes it possible to determine the relative proportions of to minimize contamination. Prior to modification, the substrates
trans or gauche configurations in the alkyl chains of OTS. were stored in a chamber with a controlled relative humidity
(RH) of ∼55%. The glass vials used for OTS adsorption were also
The theory of SFG has been described in detail else-
cleaned by piranha treatment and modified by OTS. The
where.39-44 Briefly, SFG is a second-order nonlinear optical concentration of the OTS solution is 1.0 mM, and the solvent is
a mixture of hexadecane and carbon tetrachloride (4:1 v/v). The
(28) Bunker, B. C.; Carpick, R. W.; Assink, R. A.; Thomas, M. L.; solvent mixture of anhydrous hexadecane and carbon tetrachlo-
Hankins, M. G.; Voigt, J. A.; Sipola, D.; de Boer, M. P.; Gulley, G. L. ride is referred to as “low water content solvent”. The water
Langmuir 2000, 16, 7742. content of this dry solvent was determined to be 2.6 mM by Karl
(29) Mathauer, K.; Frank, C. W. Langmuir 1993, 9, 3446. Fischer titration. In experiments to examine the effect of water,
(30) Wasserman, S. R.; Whitesides, G. M.; Tidswell, I. M.; Ocko, B.
M.; Pershan, P. S.; Axe, J. D. J. Am. Chem. Soc. 1989, 111, 5852. a drop of water was added to 50 mL solvent mixture and shaken
(31) Cohen, S. R.; Naaman, R.; Sagiv, J. J. Phys. Chem. 1986, 90, well and decanted after sitting overnight. Solvent mixtures thus
3054. prepared are referred to as “high water content solvent” and
(32) Maoz, R.; Sagiv, J. J. Colloid Interface Sci. 1984, 100, 465. have a water content of 4.4 mM. The substrates were immersed
(33) Carraro, C.; Yauw, O. W.; Sung, M. M.; Maboudian, R. J. Phys. into the OTS solution for a predetermined amount of time. For
Chem. B 1998, 102, 4441. adsorption times longer than 10 min, the solution was kept in
(34) Britt, D. W.; Hlady, V. J. Colloid Interface Sci. 1996, 178, 775.
(35) Schwartz, D. K.; Steinberg, S.; Israelachvili, J.; Zasadzinski, J.
a humidity-controlled chamber (RH ) 55%); for adsorption times
A. N. Phys. Rev. Lett. 1992, 69, 3354. shorter than 10 min, the process was carried out in ambient
(36) Tripp, C. P.; Hair, M. L. Langmuir 1995, 11, 1215. conditions (RH ) 52%, T ) 22 °C). After immersion the surface
(37) Wood, K. A.; Snyder, R. G.; Strauss, H. L. J. Chem. Phys. 1989, was rinsed in the following sequence: chloroform, acetone,
91, 5255. methanol, water, methanol, acetone, and chloroform. The modi-
(38) Snyder, R. G.; Strauss, H. L.; Elliger, C. A. J. Phys. Chem. 1982, fied substrates were then sonicated for ∼1 min in chloroform,
86, 5145.
(39) Hunt, J. H.; Guyot-Sionnest, P.; Shen, Y. R. Chem. Phys. Lett.
followed by rinsing with acetone, methanol, and water to remove
1987, 133, 189.
(40) Guyot-Sionnest, P.; Superfine, R.; Hunt, J. H.; Shen, Y. R. Chem. (43) Huang, Y. J.; Shen, Y. R. In Laser Spectroscopy and Photo-
Phys. Lett. 1988, 144, 1. chemistry on Metal Surfaces; Dai, H. L., Ho, W., Eds.; World Scientific:
(41) Shen, Y. R. Nature (London) 1989, 337, 519. Singapore, 1995; p 5.
(42) Bain, C. D. J. Chem. Soc., Faraday Trans. 1995, 91, 1281. (44) Eisenthal, K. B. Chem. Rev. 1996, 96, 1343.
Alkyl Chain Conformation Changes Langmuir, Vol. 17, No. 14, 2001 4331
unreacted precursor molecules. The samples were then dried Lateral Force Microscopy. LFM measurements were
under a stream of pure nitrogen and heated in an oven at 80 °C performed with a NanoScope III microscope (Digital Instruments)
for 3 h. The adsorption times considered in this study vary from using a silicon nitride tip with a nominal force constant of 0.12
1 min to 6 h. Allowing the substrate adsorption times longer N m-1. The samples were rinsed with acetone and dried under
than 6 h does not show further changes in the monolayer as a stream of nitrogen before imaging. All images were obtained
shown by contact angle measurements and SFG spectra. under the same ambient conditions, T ) 23 °C and RH ) 14-
Contact Angle Measurements. Contact angles were mea- 18%. No damage of the sample surfaces was observed after
sured using a Rame-Hart model 100 goniometer. The liquids scanning for 30 min with an applied force of 8 nN. In obtaining
used for contact angle measurements were water, hexadecane a LFM image, the applied force was controlled to be ∼2 nN and
(HD), and squalane (SQ). Hexadecane was purified by being the scan rate to be 1 Hz. The scan angle was set at 90° so that
passed twice through a column of active alumina. The advancing the scan direction is perpendicular to the long axis of the
contact angles of water (θaH2O) and HD (θaHD) were measured in cantilever.
a standard way, whereas static contact angles of squalane (θsSQ)
were obtained by depositing a squalane droplet (∼8 mm3) with Results and Discussion
a syringe and waiting for at least 3 min before the first reading.
The reported values are averages of at least three drops on each SFG of OTS Monolayers on Fused Silica. In these
sample surface. Errors indicate 95% confidence level. The SFG experiments, the polarization for the light is s for the
maximum θaH2O (114 ( 2.3°), θaHD (46.6 ( 0.7°), and θsSQ sum-frequency, s for visible, and p for infrared. Under
(50.7 ( 1.1°) values are in good agreement with those obtained this polarization combination, referred to as ssp, the
by others.14,15,30,32,45 relative orientation of the CH3 groups within the OTS
For a simple two-component heterogeneous surface, wetting alkyl chains can be deduced from the SFG spectrum. A
is often related to the composition of heterogeneous surfaces by closely packed OTS monolayer on fused silica will give
Cassie’s law:46 only two strong bands in the C-H vibrational region: one
centered at ∼ 2878 cm-1, corresponding to the CH3
cos θ ) f1 cos θ1 + f2 cos θ2 (1) symmetric stretch, and the other at ∼ 2946 cm-1, which
arises from the Fermi resonance of CH3 symmetric stretch
where f1 and f2 are the fractional areas occupied by components with possibly a small contribution from CH2 asymmetric
1 and 2, and θ1 and θ2 are the contact angles of pure surfaces of stretch. These two peaks in SFG spectra indicate that the
1 and 2. Assuming the OTS-modified silica surface is composed
alkyl chains are in a predominantly all-trans configuration
of only two components, pure OTS and bare silica, surface
coverage of OTS can be computed from and are oriented normal to the surface. The absence of
CH2 symmetric stretch at ∼2850 cm-1 is due to the fact
fOTS ) (1 - cos θ)/(1 - cos θmax) (2) that local ordering of transition dipoles on adjacent carbons
prevents the formation of appreciable nonlinear polariza-
tion, because the C-H symmetric stretch of the all-trans-
using the fact that bare silica surface is completely wetted
ethylene (-CH2-CH2-) unit in the chain is symmetric
(θsilica ) 0). The maximum contact angle obtained on a complete
OTS monolayer (θmax) is used as the value for the pure OTS and is therefore not infrared-active. This feature of all-
component. trans systems has been observed previously for several
Sum-Frequency Generation. The experimental setup for systems.49,52-54 A different observation in the SFG spec-
SFG is described previously.47 Briefly, a lithium niobate (LiNbO3) trum will be made if the alkyl chain is disordered and has
optical parametric oscillator (OPO) is pumped with a Surelite I some gauche defects within the chain. A peak due to the
(Continuum) Nd3+:YAG laser using relay imaging. Tunable CH2 symmetric stretch at ∼2850 cm-1 will appear in the
infrared light is generated between 2600 and 3200 cm-1 (3.1- SFG spectrum because of the change in local symmetry
3.8 µm) with a pulse width of 7 ns. The wavelength is calibrated within the alkyl chain that allows the nonlinear polariza-
with a polystyrene standard. The visible beam at 532 nm used
tion to develop. If the alkyl chains are loosely packed on
in the sum-frequency experiment is generated with a potassium
dihydrogen phosphate (KDP) crystal. Sum-frequency (SF) spectra the surface, the orientations of methyl and methylene
are obtained by using a total internal reflection geometry.48-51 groups would be randomized, and the SFG signal would
The infrared and visible beams are combined through a coupling become much smaller.
prism to the fused silica substrate with an index-matching liquid. The SFG spectra of OTS films prepared with varying
The generated sum-frequency light is collected and passed deposition times with low water and high water content
through several collimating optics, absorptive, interference and solvent are shown in Figure 1. The spectra have been
holographic filters, and a Glan-Taylor polarizer. Signal is detected offset for clarity. The spectrum of the bare fused silica
with a photomultiplier tube and then passed to a preamplifier surface is featureless, indicating that the contribution from
and gated electronics. Data were collected at 2 cm-1 increments surface contaminants is minimal. By examining Figure
in the region from 2800 to 3000 cm-1, and each point was the
average of 300 laser shots. Spectra were corrected for the
1A, the changes in alkyl chain conformation of OTS
wavelength dependence of the Fresnel factors of the input infrared monolayers can be categorized into approximately three
beam. A Voigt line shape was then used to fit each of the spectra. stages. First, when the adsorption time is less than ∼10
The sampling size on the surface was determined by the laser min, the alkyl chain has a high degree of gauche
spot size which was approximately 200 µm in diameter. Each conformation as evidenced by the relatively strong CH2
spectrum is representative of spectra taken from several places symmetric stretch at ∼2850 cm-1. Second, when the
on the sample surface. adsorption time is in the range 15-60 min, the molecular
conformation of OTS monolayers experiences a dramatic
(45) Semal, S.; Voué, M.; de Ruijter, M. J.; Dehuit, J.; De Coninck, change from disorder to order as shown by the rapid
J. J. Phys. Chem. B 1999, 103, 4854.
(46) Cassie, A. B. D. Discuss. Faraday Soc. 1948, 3, 11. increase of the relative intensity of the CH3 symmetric
(47) Yang, Y. J.; Pizzolatto, R. L.; Messmer, M. C. J. Opt. Soc. Am. stretch. The disappearance of the CH2 symmetric stretch
B 2001, 17, 638. peak marks the end of the second stage. Third, when the
(48) Conboy, J. C.; Messmer, M. C.; Richmond, G. L. J. Phys. Chem.
1996, 100, 7617.
adsorption time is longer than 60 min, the adsorption
(49) Messmer, M. C.; Conboy, J. C.; Richmond, G. L. J. Am. Chem.
Soc. 1995, 117, 8039. (52) Guyot-Sionnest, P.; Hunt, J. H.; Shen, Y. R. Phys. Rev. Lett.
(50) Hatch, S. R.; Polizzotti, R. S.; Dougal, S.; Rabinowitz, P. J. Vac. 1987, 59, 1597.
Sci. Technol. A 1993, 11, 2232. (53) Bain, C. D.; Davies, P. B.; Ward, R. N. Langmuir 1994, 10, 2060.
(51) Hatch, S. R.; Polizzotti, R. S.; Dougal, S.; Rabinowitz, P. Chem. (54) Conboy, J. C.; Messmer, M. C.; Richmond, G. L. J. Phys. Chem.
Phys. Lett. 1992, 196, 97. B 1997, 101, 6724.
4332 Langmuir, Vol. 17, No. 14, 2001 Liu et al.
Figure 1. Sum-frequency spectra of OTS layers on fused silica surfaces: (A) low water content solvent; (B) high water content
solvent. Spectra are taken with ssp polarizations (s-SF, s-visible, p-infrared). The solid line is a fit to the data using a Voigt line
shape.
process is much slower. The increase in adsorption time surface coverages of partial OTS monolayers calculated
from 1 to 6 h results in only a slight increase in the SFG from squalane contact angles correlate well the relative
band intensities, suggesting the conformation of the OTS film thickness obtained from ellipsometry and the atomic
chains has already been established. The increase in the ratio of C(1s)/Si(2p) measured by X-ray photoelectron
SF intensity is primarily due to the increased number of spectroscopy (XPS).57 Therefore, in this study we use static
surface species. squalane contact angles to estimate the surface coverages
The effect of water in the deposition solvent on the OTS of OTS monolayers. The effect of high water content solvent
film formation was also examined by SFG. As shown in on the kinetics of film formation is clearly illustrated by
Figure 1B, the CH3 symmetric stretch of films prepared squalane contact angles shown in Figure 2. Water and
with high water content solvent increases much faster hexadecane contact angle measurements were also per-
than low water content solvent, indicating their alkyl formed (data not shown), and the results showed similar
chains are more ordered. It is reasonable to assume that trends as found with squalane. Generally, for shorter
the increase in chain order is due to the increase of surface adsorption times, high water solvent samples give higher
coverage due to faster kinetics with more water present. contact angles. For longer adsorption times, contact angles
In fact, studies of partial OTS monolayers on silicon wafers of both low water and high water solvent samples approach
have shown that higher coverages are obtained for high the same value, confirming that the quality of resultant
water-containing solvent.26 Although the humidity level OTS SAMs are similar for long adsorption times. This
under which our experiments were conducted was chosen observation is consistent with SFG results that indicate
to provide optimal film formation,24 it is obvious that water well-ordered monolayers can be formed using either
in the bulk solvent also plays an important role on the solvent (vide supra). It is interesting to note that even
OTS film formation. Eventually, both solvents produce though there is an obvious difference between contact
well-ordered films, as shown by the similarity of the angle measurements of low water and high water solvent
spectra at 6 h. samples at the first 2 min of film evolution, SF spectra of
Contact Angles and Surface Coverages. Although these two sets are only slightly different (Figure 1). This
water and hexadecane are commonly used to test the difference may suggest that contact angle measurement
quality of OTS monolayers, these liquids do not provide is more sensitive to changes in surface composition, while
quantitative information about surface coverage of partial SFG is more sensitive to the conformation of surface
OTS layers. It has been reported that hexadecane may molecules.
penetrate into monolayers of long hydrocarbon chain and As discussed previously, it is well-known that SAMs of
induce reconstruction of surface molecules.55,56 Recently, organosilanes are very sensitive to many of the reaction
De Coninck and co-workers have shown that static conditions during formation. The adsorption time needed
squalane contact angles of partial OTS monolayers on a to reach a certain coverage can easily be influenced by
silicon wafer surface follow Cassie’s law behavior.45 slight day-to-day and batch-to-batch variations in reaction
Previous work in this lab has demonstrated that the conditions. Therefore, to examine the molecular confor-
mation of separately prepared samples, the use of surface
(55) Allara, D. L.; Parikh, A. N.; Judge, E. J. Chem. Phys. 1994, 100,
1761. (57) Liu, Y.; Messmer, M. C. In Thin Films: Preparation, Charac-
(56) Miranda, P. B.; Pflumio, V.; Saijo, H.; Shen, Y. R. J. Am. Chem. terization, Applications; Stickney, J., Soriaga, M. P., Eds.; Kluwer
Soc. 1998, 120, 12092. Academic/Plenum: New York, 2001; submitted.
Alkyl Chain Conformation Changes Langmuir, Vol. 17, No. 14, 2001 4333
Figure 2. Squalane contact angles of OTS on fused silica Figure 3. Sum-frequency intensity ratios of low ([) and high
prepared with low ([) and high (O) water content solvents. The (O) water samples as a function of surface coverage, as estimated
lines are used as a guide to the eye. by squalane contact angles.
coverage is preferred to that of adsorption time. In doing lead to a faster evolution of order on the surface, once the
so, it is easier to compare our results with those obtained aggregate deposits.
by others, as well as those of different reaction conditions. LFM of OTS on Fused Silica. Complementary
Unlike SFG studies of alkanethiol SAMs on gold substrates information about OTS growth is also obtained from LFM,
in which surface coverage can be estimated from the which measures twisting of the cantilever that arises from
nonresonant SFG substrate signal,58 surface coverage of the lateral forces on the cantilever parallel to the plane
OTS on fused silica is not readily available from SFG of sample surface.60 Factors that affect image contrast in
because the nonresonant substrate signal is very weak.59 LFM include tip-sample contact area,61 chemical func-
We therefore calculate the OTS coverage on fused silica tionality,62 and the presence of a water layer on the
from eq 2 using static squalane contact angles. The order- surface.63 LFM is used here because the AFM images of
disorder transition of the OTS film can be represented by partial OTS monolayers on fused silica surfaces are
the ratio of the methylene symmetric stretch intensity to complicated by the roughness of the fused silica substrates.
methyl symmetric stretch intensity peak, ICH2(SS)/ICH3(SS), It is difficult to distinguish the topography of the partial
referred to as the SF intensity ratio. A higher SF intensity OTS monolayers from the AFM images. Conversely, the
ratio indicates a more disordered molecular conformation source of contrast within a LFM image is based primarily
in the film. The SF intensity ratio as a function of coverage on the fact that the friction between a hydrophilic AFM
is shown in Figure 3. The three stages of the alkyl chain tip and the hydrophobic, closely packed OTS monolayers
conformational changes as discussed earlier can be clearly is lower than between the tip and the hydrophilic silica
seen from the curves of the low water content solvent substrate.64 In fact, LFM images of patterned SAMs often
samples in Figure 3. Initially, when the OTS film is less have better contrast and sharper boundaries than AFM
than 65% coverage, the SF intensity ratio is high and images.65
remains nearly constant as the surface coverage increases The LFM images of partial OTS monolayers on fused
rapidly. This indicates the alkyl chains in the OTS layers silica surfaces are shown in Figure 4 and Figure 5. The
remain in a highly disordered state. A dramatic confor- dark areas (low frictional force) in the images are
mational change occurs in the second stage, where the attributed to OTS molecules on the surface. These images
surface coverage increases at a much slower rate. The indicate that an island growth mechanism does take place
OTS modified surfaces become very well-ordered when for both low water and high water content deposition
surface coverage reaches 90%. The effect of water in the solutions. However, the areas of these islands change with
deposition solution is also apparent in Figure 3. For the time and with the presence of bulk water in the deposition
same surface coverages, the films formed with high water solution. Initially, the low water content deposition
content solvent are more ordered than ones formed in the solution creates islands with sizes in the range 0.1-0.3
low water content solvent. This higher order can be µm. Larger islands emerge for the 2 min adsorption
attributed to the island growth mechanism which is more sample, and the fractal-like shapes of those islands suggest
favored by higher water content in the solvent. Water in that the growth is due to the accumulation of clusters of
the bulk deposition solution may hydrolyze monomers,
which then may undergo condensation with other mono-
(60) Overney, R. M.; Meyer, E.; Frommer, J.; Brodbeck, D.; Luethi,
mers to form solution aggregates. In addition to leading R.; Howald, L.; Guentherodt, H. J.; Fujihira, M.; Takano, H.; Gotoh, Y.
to a faster reaction on the surface, the alkyl chains within Nature (London) 1992, 359, 133.
the aggregates may become “preordered”. The close (61) Bar, G.; Rubin, S.; Parikh, A. N.; Swanson, B. I.; Zawodzinski,
T. A., Jr.; Whangbo, M.-H. Langmuir 1997, 13, 373.
packing of alkyl chains within the aggregate will then (62) Frisbie, C. D.; Rozsnyai, L. F.; Noy, A.; Wrighton, M. S.; Lieber,
C. M. Science 1994, 265, 2071.
(58) Himmelhaus, M.; Eisert, F.; Buck, M.; Grunze, M. J. Phys. Chem. (63) Piner, R. D.; Mirkin, C. A. Langmuir 1997, 13, 6864.
B 2000, 104, 576. (64) Lee, B.; Clark, N. A. Langmuir 1998, 14, 5495.
(59) Löbau, J.; Rumphorst, A.; Galla, K.; Seeger, S.; Wolfrum, K. (65) Zhou, Y.; Fan, H.; Fong, T.; Lopez, G. P. Langmuir 1998, 14,
Thin Solid Films 1996, 289, 272. 660.
4334 Langmuir, Vol. 17, No. 14, 2001 Liu et al.
Figure 4. The 5 × 5 µm LFM images of the partial OTS monolayers formed with low water content solvent with adsorption times
of 1 (A), 2 (B), 5 (C), and 30 min (D).
OTS molecules. However, the shapes of islands for the 5 occupied by island structures was less than 25% of the
min adsorption sample are less fractal-like than that of total surface area, yet the squalane contact angle mea-
the 2 min sample presumably due to monomer filling in surements yield a surface coverage of 52%. Thus, a
the existing island structures and/or two-dimensional significant amount of surface OTS molecules were not
movement of existing surface species. More interestingly, contained in the island structures detected by LFM. This
the contrast between islands and “floor” becomes less than disparity indicates a portion of the OTS monolayers still
that in the 1 and 2 min sample. This decreasing contrast possess disordered alkyl chains, from a uniform growth
indicates that, in addition to island growth, uniform mechanism, and the loosely packed nature of these
growth of OTS molecules occurs during the adsorption disordered films increases the frictional forces. Further
process, which causes the “floor” on the images in Figure evidence can be found by correlating SFG results with
4 to change from hydrophilic to hydrophobic, resulting in LFM images. Within the OTS islands, the alkyl chain
reduced frictional forces. After longer adsorption time, would be expected to have a high degree of order. These
the domains of islands are much less distinctive (Figure regions would represent areas of high density and
4D). The diminishment of the image contrast in Figure 4 therefore should have a high degree of order among the
indicates that the partial OTS monolayers formed by alkyl chains. However, the overall conformation of OTS
uniform growth are initially disordered and have a lower chains for 1-15 min adsorption samples is still highly
density. Therefore, a larger tip-sample contact area is disordered (Figure 1A). Therefore, the large degree of
expected on these surfaces, which causes the frictional disorder that is observed suggests that a mixed growth
force to increase. As the OTS coverage increases, the tip- mechanism must exist for both low and high water
sample contact area of the uniform growth film decreases, samples. Evidently, the partial OTS monolayers formed
causing a decrease in contrast between the OTS islands via a uniform growth mechanism have a high degree of
and the “floor”. Eventually, this contrast should be gauche defects and are random in nature. At longer times
minimal when the film density increases and the overall though, the alkyl chains become more ordered as small
monolayer becomes closely packed. quantities of monomers begin to fill in the remaining sites.
The existence of a uniform growth mechanism in the The kinetic effect of water in the solvent on the OTS
OTS film growth process is also supported by comparing adsorption is clearly seen from the LFM images of partial
the island areas of the LFM images with surface coverage OTS monolayers formed via high water content solvent
calculated by squalane contact angles. For example, on deposition (Figure 5). The high water content solvent
the 5 min sample image shown in Figure 4C, the area initially deposits fractal-like islands whose sizes are much
Alkyl Chain Conformation Changes Langmuir, Vol. 17, No. 14, 2001 4335
Figure 5. The 5 × 5 µm LFM images of the partial OTS monolayers formed with high water content solvent with adsorption times
of 1 (A), 2 (B), 5 (C), and 10 min (D).
larger than those produced by low water content solvent conformation. Finally, a slight increase in surface coverage
for the same period of deposition time. This observation only slightly affects SF intensities. The initial stages of
is consistent with the adsorption of larger OTS aggregates OTS deposition on fused silica surfaces can be probed by
formed in the high water content solution.28 The island LFM, which revealed island formation for both low and
size does not significantly change at 2 min adsorption high water content solvent deposition. The contrast in
probably because these larger OTS aggregates were unable the LFM images is indicative of the uniform growth of
to fill in the existing surface structures. However, from OTS molecules during the deposition with both solvents.
the contact angle measurements (Figure 2), a significant From SFG, contact angle measurements, and LFM images,
increase in surface coverage from 1 to 2 min adsorption it can be concluded that the adsorption of OTS on fused
for the high water content solvent deposition is observed. silica surfaces has two competing growth mechanisms,
Therefore, the OTS molecules must reach the unoccupied island growth and uniform growth. Island growth is
surface area via a uniform growth mechanism, which significant in the initial deposition of OTS molecules and
causes the contrast between the islands and the “floor” to related to the water content of the system. More water in
decrease slightly. With further deposition, a decrease in the solvent facilitates the OTS deposition via island
island size and contrast is observed, consistent with the growth, and the resulting films are more ordered than
filling in of the remaining surface area with OTS molecules that prepared with low water content solution. The
and the concomitant increase in the alkyl chain order. uniform growth mechanism is also responsible for filling
in the remaining spaces among the islands and forming
Conclusion monolayers with well-ordered alkyl chain conformation.
This work demonstrates that the alkyl chain conforma-
Acknowledgment. This work was supported by the
tion of OTS monolayers on fused silica substrates evolves
National Science Foundation (CHE-9709386). Y.L. ac-
in three stages. Initially, the surface coverages increase
knowledges the White Fellowship from Department of
dramatically while the film is still disordered. Then the
Chemistry, Lehigh University.
monolayers experience a rapid change from disordered to
well-ordered with the alkyl chains adopting an all-trans LA010123C
