Nanoscale Water Condensation on Click-Functionalized Self

ARTICLE pubs.acs.org/Langmuir

Nanoscale Water Condensation on Click-Functionalized Self-Assembled Monolayers Michael James,*,†,‡ Simone Ciampi,‡ Tamim A. Darwish,† Tracey L. Hanley,† Sven O. Sylvester,‡ and J. Justin Gooding‡ † ‡

Australian Nuclear Science and Technology Organisation (ANSTO), Locked Bag 2001, Kirrawee DC NSW 2232, Australia School of Chemistry, The University of New South Wales, Sydney NSW 2052, Australia

bS Supporting Information ABSTRACT: We have examined the nanoscale adsorption of molecular water under ambient conditions onto a series of wellcharacterized functionalized surfaces produced by Cu(I)-catalyzed alkyne
azide cycloaddition (CuAAC or “click”) reactions on alkyne-terminated self-assembled monolayers on silicon. Water contact angle (CA) measurements reveal a range of macroscopic hydrophilicity that does not correlate with the tendency of these surfaces to adsorb water at the molecular level. X-ray reflectometry has been used to follow the kinetics of water adsorption on these “click”-functionalized surfaces, and also shows that dense continuous molecular water layers are formed over 30 h. For example, a highly hydrophilic surface, functionalized by an oligo(ethylene glycol) moiety (with a CA = 34
) showed 2.9 Å of adsorbed water after 30 h, while the almost hydrophobic underlying alkyne-terminated monolayer (CA = 84
) showed 5.6 Å of adsorbed water over the same period. While this study highlights the capacity of X-ray reflectometry to study the structure of adsorbed water on these surfaces, it should also serve as a warning for those intending to characterize self-assembled monolayers and functionalized surfaces to avoid contamination by even trace amounts of water vapor. Moreover, contact angle measurements alone cannot be relied upon to predict the likely degree of moisture uptake on such surfaces.

1. INTRODUCTION The ability of water vapor to condense onto a surface depends on a range of factors including temperature, relative humidity, surface charge, roughness, and the hydrophilicity of the exposed surface. Water contact angle measurements are frequently used to give an indication of the quality of surfaces formed by molecular self-assembly,1
3 as well as quantifying the degree of functionalization that takes place upon reaction of target molecules with SAMs.4
6 Recent molecular dynamics7
11 and experimental studies11
15 have revealed that nanoscale water layers are present on organic self-assembled monolayers (SAMs), even those that are considered almost hydrophobic by classical measures such as water contact angle goniometry. We have shown using X-ray reflectometry (XRR) that the degree of water adsorption on hydrophilic SAMs does not necessarily follow the behavior expected from water contact angle measurements.15 A number of authors have studied the chemical, structural, and physical properties of mixed monolayers, particularly those formed by a combination of hydrophilic and hydrophobic moieties.7,11,16,17 A number of factors contribute to the observed water contact angle in mixed self-assembled monolayers including polarity of the distal groups, homogeneity of the SAMs, and surface roughness effects. While contact angle measurements are routinely used to quantify the extent of coverage of the different molecules r 2011 American Chemical Society

in these SAMs, these studies have shown that the variation in contact angle is not linear with composition. Although contact angle measurements can to some extent quantify the composition of mixed monolayers, they say little in relation to the interactions between the surface and atmospheric water. In common with mixed monolayers formed in one step, contact angle measurements are also used to explore the far more commonly used functional surfaces built in a multistep modular fashion.18 Numerous functionalization strategies have been investigated and ultimately give rise to surfaces with characteristics being a closer reflection to the equivalent SAM. Despite contact angle measurements being used to characterize functional surfaces formed in multiple steps, their capacity to reveal detailed knowledge of the coverage of the functionalized surface is limited. The uppermost surface and the underlying SAM both contribute to the wettability of the functionalized surface. X-ray reflectometry is a key tool for the study of the structure and composition of SAMs and functionalized surfaces bound to silicon. Excellent scattering length density (SLD) contrast exists between air (SLD = 0), silicon (SLD = 2.01  10
5 Å
2), and the Received: June 23, 2011 Revised: July 22, 2011 Published: July 25, 2011 10753

dx.doi.org/10.1021/la202359c | Langmuir 2011, 27, 10753–10762

Langmuir organic molecules of the SAM or functionalized surface (SLD = ∼1  10
5 Å
2). In addition, XRR has the capacity to follow changes in the thickness and composition of functionalized layers with subangstrom precision and the ability to conduct measurements without having to resort to ultrahigh vacuum conditions.4,19
21 Accurate determination of structural parameters of selfassembled monolayers and functionalized surfaces using X-ray reflectometry does however rely critically on the absence of adsorbed water. The presence of even 1
2 Å of adsorbed water (having a comparable SLD to the organic components in the film) can lead to large errors in refined structural parameters, potentially rendering the results completely unphysical. Allowing the alkyne-terminated SAM (Surf-1), for example, to be exposed to ambient conditions for 10 h would lead to a measured film thickness of more than 12 Å,15 which is larger than the fully extended Si
CHdCH(CH2)5CtCH chain length, even assuming that the chains are oriented normal to the SAM surface. In this report, we investigate the nanoscale adsorption of molecular water under ambient conditions onto a range of functionalized surfaces produced by modifying an alkyne-terminated SAM on oxide-free Si(100) by Cu(I)-catalyzed alkyne
azide cycloaddition (CuAAC or “click”)22
24 reactions. Such reactions have in recent years been shown to be an excellent tool for surface immobilization,4,25
30 as they are tolerant to a wide range of functional groups and solvents, benefit from high selectivity, and do not require activation or protection/deprotection steps. These functionalized surfaces, selected due to their relevance in biology, electrochemistry, and photochemistry, were characterized using X-ray reflectometry, X-ray photoelectron spectroscopy, and water contact angle goniometry. We show that conventional measures of hydrophilicity such as water contact angle measurements may be a useful indicator of chemical functionalization; however, they are not a valuable gauge of the interaction of water with a surface at the molecular level.

2. EXPERIMENTAL SECTION 2.1. Materials. All chemicals, unless noted otherwise, were of analytical grade and used as received. Chemicals used in surface modification procedures were of high purity (>99.9%). 1,8-Nonadiyne (1, Alfa Aesar, 97%) was redistilled from sodium borohydride (SigmaAldrich, 99+%) under reduced pressure. Solvents for synthesis of reagents, surface cleaning, and surface modification procedures were redistilled prior to use. Milli-Q water (>18 MΩ cm) was used for surface cleaning and surface modification. Prime grade silicon wafers, 100-oriented (Æ100æ ( 0.05
), p-type (boron), 500 ( 25 μm thick, 13) by addition of sodium hydroxide solution. The product was extracted with dichloromethane (7  25 mL), and the combined organic extracts were dried over Na2SO4, filtered, and evaporated in vacuo. Triphenylphosphine oxide was crystallized from the crude mixture by storing at 0
C for 2 days. The remaining oil was redissolved in the minimal amount of ice-cold water and filtered, and the filtrate evaporated in vacuo to give 1-amino-oligo(ethylene glycol) (OEG) 4 as pale yellow oil (0.36 g, 50%). 1 H NMR (300 MHz, CDCl3) δ: 2.84 (t, 2H, J = 5.0 Hz), 3.38 (t, 2H, J = 5.0 Hz), 3.48 (t, 2H, J = 5.0 Hz), 3.63
3.69 (m, 12H). 13C NMR (75.5 MHz, CDCl3) δ: 41.35, 50.56, 69.91, 70.16, 70.51, 70.53, 70.58, 73.19. 2.2.4. Azidomethylferrocene (5). Azidomethylferrocene (5) was synthesized from hydroxymethylferrocene in the manner previously reported by Ciampi et al.27 Hydroxymethylferrocene (0.6 g, 2.8 mmol) and sodium azide (0.4 g, 6.2 mmol) were dissolved in glacial acetic acid (10 mL) under argon and stirred at 50
C for 3 h. The reaction mixture was diluted with dichloromethane (50 mL), washed with a saturated NaHCO3 solution (3  50 mL) and water (1  50 mL), and dried over Na2SO4. After filtering, the product was dried in vacuo to yield an orange oil. Column chromatography (ethyl acetate/hexane, 1:2) was used to purify the crude product to give azide 5 as a yellow/orange solid (450 mg, 67%). 1H NMR (300 MHz, CDCl3) δ: 4.24 (t, 2H, J = 1.7 Hz), 4.20 (t, 2H, J = 1.7 Hz), 4.17 (bs, 5H), 4.12 (bs, 2H). 13C NMR (75.5 MHz, CDCl3) δ: 82.25, 68.83, 68.0, 51.03. IR (KBr, cm
1): 3101, 2957, 2936, 2099, 2072. 2.2.5. 6-Azido-8-methoxy-spiro(2H-1-benzopyran-2,20 -(2H)-indole). (6-Azido-8-methoxy-BIPS) (6) was synthesized in two steps from the commercially available 8-methoxy-6-nitro-BIPS according to the method reported by Fukushima et al.,31 with modifications. Reduction of the nitro group was achieved by combining 8-methoxy-6-nitro-BIPS (500 mg, 1.41 mmol) with Sn (402 mg, 3.387 mmol) in 2 mL of ethanol in a 50 mL round-bottom flask fitted with a water cooled condenser. The flask was immersed in an ice/salt mixture, and 6 mL of 6 M hydrochloric acid solution was added dropwise over a 20 min period. The mixture was stirred for 15 min at room temperature, before refluxing at 90
C for a further 30 min. After cooling to room temperature, 1 N sodium hydroxide solution was added until pH 11 was reached. Diethyl ether (20 mL) was added and the organic phase extracted, dried over Na2SO4, filtered, and evaporated in vacuo to give 6-amino-8-methoxy-BIPS (380 mg, 1.18 mmol (84%)). Reaction of the aminoBIPS derivative to give the corresponding azide-tagged derivative: a solution of 6-amino-8-methoxy-BIPS (300 mg, 0.93 mmol) in 2 M hydrochloric acid (1.5 mL)/ acetone (1 mL) was cooled in an ice/sodium chloride mixture before 10754

dx.doi.org/10.1021/la202359c |Langmuir 2011, 27, 10753–10762

Langmuir

ARTICLE

Scheme 1. Molecular Self-Assembly on Silicon by ω-Funtionalized 1-Alkyne Reagentsa

a Preparation of the ethynyl-terminated film (Surf-1) from the diyne 1 and its direct CuAAC reactions with the azides 2
6 to form films presenting
CH3 (Surf-2),
NH2 (Surf-3), OEG (Surf-4), ferrocenyl units (Surf-5), and spiropyran-derivatives (Surf-6).

sodium nitrite (78 mg, 1.13 mmol) in water (0.4 mL) was added over 30 min. After 30 min at
5
C, urea (5.6 mg, 0.09 mmol) was added to this solution, before this was added while cold to a solution of sodium azide (123 mg, 1.89 mmol) and sodium acetate (230 mg, 2.80 mmol) in 1:1 water/acetone mixture (5 mL) below 0
C. Stirring continued for 4 h with the temperature held below 0
C, before 10 mL of water was added and the aqueous phase was extracted with diethyl ether (20 mL). The ether layer was washed twice with 1 N sodium hydroxide solution (10 mL each time) and water (10 mL), and then the organic phase was dried over Na2SO4 and evaporated in vacuo to give 6-azido-8-methoxyBIPS (6) as a red brown solid (230 mg; 70%). 1H NMR (400 MHz, CDCl3) δ: 7.15 (t, 1H), 7.04 (d, 1H), 6.81 (t, 1H), 6.75 (d, 1H, J = 10.2 Hz), 6.50 (d, 1H), 6.40 (s, 2H), 5.73 (d, 1H, J = 10.2 Hz), 3.66 (s, 3H), 2.74 (s, 3H), 1.30 (s, 3H), 1.16 (s, 3H).

2.3. Assembly of Monolayers of ω-Funtionalized 1-Alkynes (Surf-1). Assembly of the acetylene-terminated monolayer (Si
CHd

CH(CH2)5CtCH, Surf-1) by covalent attachment of terminal alkyne 1 onto Si(100) wafers, as depicted in Scheme 1, followed previously reported procedures.4,27,30,32 In brief, wafers (10  30 mm) were rinsed with ethanol and dichloromethane and then cleaned for 30 min in hot Piranha solution (100
C, 1 vol 30% by mass aqueous hydrogen peroxide to 3 vol sulfuric acid), before being transferred to an aqueous fluoride solution (2.5% hydrofluoric acid, 11/2 min). The wafers were then transferred, taking care to exclude air from the reaction vessel, to a degassed sample (ca. 1 mL) of diyne 1. The reaction mixture was kept under high purity argon (H2O < 10 ppb, O2 < 1 ppb) and heated at 165
C for 3 h. After cooling to room temperature, the modified silicon wafers were then removed from the reaction flask and rinsed with copious amounts of dichloromethane before being analyzed, or used in subsequent CuAAC functionalization reactions.

2.4. Functionalization of Alkyne Monolayers via CuAAC Reactions (Surf-2
Surf-6). The reactions used to functionalize Surf-1 are summarized in Scheme 1 and described in more detail below. To vessels containing Surf-1 samples were added solutions of the azides 2
5 (10 mM, 2-propanol/water, 2:1), copper(II) sulfate pentahydrate (1 mol % relative to the azide), and sodium ascorbate (10 mol % relative to the azide). Reactions were carried out in the dark at room temperature for a 24 h period with 3-azidopropylamine 3 to give Surf-3, with azidetagged OEG 4 to yield Surf-4, and with azidomethylferrocene 5 to give Surf-5. 1-Azidobutane 2 was reacted for 24 h with Surf-1 in the dark at 35
C to give Surf-2. The “clicked” surface-bound [1,2,3]-triazoles samples were then rinsed consecutively with copious amounts of water and ethanol and then rested for 2 min in a 0.5 M hydrochloric acid solution. Samples were then rinsed with copious amounts water, ethanol, and dichloromethane, dried under vacuum (ca. 100 mbar), and stored under argon gas before being analyzed. Surf-6 was produced by adding 6-azido-8-methoxy-BIPS 6 (7.2 mM, 2-propanol/water/tetrahydrofuran, 2:1:1), copper(II) sulfate pentahydrate (1 mol % relative to the azide), and sodium ascorbate (40 mol % relative to the azide) to a vial containing Surf-1. Reactions were carried out at room temperature in the dark and stopped after 24 h. The surface-bound [1,2,3]-triazole(spiropyran) (Surf-6) was rinsed with copious amounts of water, 2-propanol, and chloroform, dried under vacuum, and stored in dry nitrogen gas. 2.5. Characterization of Functionalized Surfaces. 2.5.1. Water Contact Angle Goniometry Measurements. Water contact angle measurements were made using a Ram
e-Hart 200-F1 goniometer. Samples were prepared in triplicate with at least three separate spots being measured for each sample. Reproducibility of these measurements was (3
. 2.5.2. X-ray Photoelectron Spectroscopy (XPS). XPS measurements were performed on an ESCALAB 220iXL instrument. Monochromatic 10755

dx.doi.org/10.1021/la202359c |Langmuir 2011, 27, 10753–10762

Langmuir Al KR X-rays (1486.6 eV) incident at 58
to the analyzer lens were used to excite electrons from the sample. Emitted photoelectrons were collected on a hemispherical analyzer with multichannel detector at a takeoff angle of 90
from the plane of the sample surface. The analyzing chamber operated below 10
8 Torr and the spot size was approximately 0.5 mm2. The resolution of the spectrometer was ∼0.6 eV. All energies are reported as binding energies in eV and referenced to the C 1s signal (corrected to 285.0 eV). Survey scans were carried out selecting 100 ms dwell time and analyzer pass energy of 100 eV. High-resolution scans were run with 0.1 eV step size and dwell time of 100 ms, and the analyzer pass energy set to 20 eV. The high resolution data were analyzed first by background subtraction using the Shirley routine and a subsequent nonlinear fitting by mixed Gaussian
Lorentzian functions as described previously.4 When detected, the monolayer coverage of oxidized silicon was calculated directly from the oxidized/bulk Si 2p peak area ratio according to the method described by Webb and co-workers for very thin oxide overlayers.33 For Surf-2
Surf-6 samples, the ratios of the integrated areas for the C 1s and N 1s emissions, each normalized for their elemental sensitivity, afforded an estimate for the CUAAC reaction yield.4,27,34 2.5.3. X-ray Reflectometry (XRR). XRR profiles of self-assembled surfaces on silicon were measured under ambient conditions (30
C; 35% relative humidity) on a Panalytical Ltd. X’Pert Pro Reflectometer using Cu KR X-ray radiation (λ = 1.54056 Å). The X-ray beam was collimated using a G€obel mirror with a 0.1 mm slit and a postsample parallel collimator. Reflectivity data were collected over the angular range 0.05
e θ e 5.00
, with a step size of 0.010
and counting times of 10 s per step. Once exposed to air, alignment of the sample on the X-ray reflectometer took approximately 10 min prior to measurement. Structural parameters of the prepared surfaces were refined using the MOTOFIT reflectivity analysis software35 with reflectivity data as a function of the momentum transfer vector normal to the surface Q = 4π(sinθ)/λ. The Levenberg
Marquardt method was selected to minimize χ2 values in the fitting routines. Singlelayer or two-layer models were used to fit the observed data.

3. RESULTS AND DISCUSSION The production of functionalized surfaces involves the formation of an alkyne-terminated self-assembled monolayer (Surf-1) on hydrogen-terminated silicon (Si
H) using a symmetrical R,ω-diyne (1,8-nonadiyne 1, Scheme 1). The hydrosilylation reaction of 1 with the Si
H surface gives an exceedingly stable alkenyl linkage (Si
CdC
)36,37 and results in a well-defined and stable monolayer on the silicon surface.4,15,27,38 XPS spectra acquired on the acetylene-terminated SAM (Surf-1) prepared from the diyne 1 are shown in Figure S1 (Supporting Information), and all the refined spectral features (i.e., C 1s and Si 2p emissions) agree with previous reports.39 The high-resolution XPS C 1s signal (Figure S1b, Supporting Information) was fitted with three component peaks; the major component at 285.0 eV (87%, 1.4 eV fwhm) is assigned to the methylene carbons of the adsorbate.40 The lower binding energy component at 283.8 eV (5%, 0.9 eV fwhm) and the higher binding energy component at 286.4 eV (8%, 1.4 eV fwhm) are assigned respectively to siliconbound olefinic carbons (silylated olefin, Si
CdC)41,42 and to oxygen-bound carbon atoms (C
O).42 The structure of the adventitious C
O bond has not been fully elucidated and is currently debated.40,42,43 On Si
H surfaces, monolayer assembly with 1-alkenes results in alkyl monolayers with a Si
C
C linkage, while 1-alkynes, as in this report, yield alkenyl monolayers with a Si
CdC linkage.36,42,44
47 In contrast to the case of Si
C
C, the Si
CdC linkage is known to limit oxidation of the underlying silicon and therefore enhance the monolayer stability.37 Most importantly, the high-resolution Si 2p scan

ARTICLE

reveals relevant information about the monolayer quality and its ability to prevent appreciable oxidation of the underlying Si substrate. In the Si 2p region (Figure S1c, Supporting Information) obtained for surface Surf-1, the silicon oxide content was below the spectrometer detection limit (ca. 0.06 SiOx monolayers) in the 102
104 eV region. The distal end of this monolayer is an ethynyl function that is amenable to further modification with azide species via the CuAAC (or “click”) reactions.32
34 The syntheses of these different functionalized molecular coatings (Surf-2
Surf-6) are summarized in Scheme 1. 3.1. Characterization of Freshly Functionalized Surfaces. 3.1.1. X-ray Photoelectron Spectroscopy. The composition and quality of the functionalized surfaces (Surf-2 to Surf-6) were characterized using XPS and XRR. The XPS data presented below in Figure 1 are for Surf-3 (amino-terminated SAM), while the equivalent results for functionalized Surf-2, and Surf-4 to Surf-6 are given in the Supporting Information48 (Figures S2
S5, respectively). XPS survey spectra (Figure 1a for Surf-3) show the presence of N 1s peaks at ca. 401 eV, indicating immobilization of 1-azidopropylamine 3 onto the alkyne SAM and successful triazole formation. High-resolution spectra (Figure 1b
d) confirm the presence of high quality monolayer surfaces with the appropriate molecular functionality. The amount of Si detected from all of these surfaces (ca. 100 eV) was also consistent with the film thickness derived from X-ray reflectometry, and essentially no detectable photoemission (i.e., equal or below the detection limit of