Article pubs.acs.org/Langmuir
Impact of Hydrophilic/Hydrophobic Surface Chemistry on Hydration Forces in the Absence of Confinement Gillian B. Kaggwa,† Prathima C. Nalam,‡ Jason I. Kilpatrick,† Nicholas D. Spencer,‡ and Suzanne P. Jarvis*,† †
Conway Institute, University College Dublin, Belfield, Dublin 4, Ireland Laboratory for Surface Science and Technology, Department of Materials, ETH Zurich, Wolfgang-Pauli-Strasse 10, 8093 Zurich, Switzerland
‡
ABSTRACT: The oscillatory force profile, observed in liquids due to molecular ordering at interfaces, has been extensively investigated by means of atomic force microscopy, but it remains unclear whether molecular ordering is present at the tip apex. Using a displacement-sensitive, low-noise atomic force microscope (AFM) operated in dynamic mode, with a tip of radius < 1 nm, we have investigated the force profile between two approaching surfaces of the same or different hydrophilic and hydrophobic character. By directly comparing different surface chemistry interactions, we have been able to elucidate whether an oscillatory force profile is due to structured water layers adjacent to the surface, the tip, or a combination of the two. We have found that an oscillatory force profile is observed when the surface is hydrophilic in nature, irrespective of whether the tip is hydrophilic or hydrophobic. When the surface is hydrophobic, an oscillatory force profile is not measured, but rather a monotonic repulsive or a short-range attractive force is observed for interactions with a hydrophilic or a hydrophobic tip, respectively. Thus, we attribute the measurement of an oscillatory force profile, in the absence of lateral confinement effects, solely to water layers adjacent to a hydrophilic surface rather than the structuring of water at the tip apex. This is the first direct evidence that solvation forces occur solely as a result of water layers adjacent to the substrate.
■
and results in the force profile being “smeared out” with no clearly detectable oscillations.10−12 In contrast to the requirement for smooth rigid surfaces to observe solvation forces in SFA, atomic force microscopy has been shown to resolve oscillatory force profiles for both rough and flexible interfaces.5,13,14 While SFA measures forces over large areas, AFM measurements are highly localized and detect the liquiddensity profile directly beneath the tip. As such, interactions with the surface can occur on length scales smaller than that of the surface roughness. AFM has been used to measure the oscillatory solvation force for both nonpolar and polar liquids, including octamethylcyclotetrasiloxane (OMCTS)7,15,16 and water5,17 on a range of surfaces: highly ordered pyrolytic graphite (HOPG),15 selfassembled monolayers,7 and model lipid bilayers.14,18 To date, little is known about the effect of tip hydrophobicity on the measurement of solvation forces, although the effects of tip geometry and aspect ratio have previously been investigated.5,19 Li et al. investigated the role of surface chemistry and roughness on the confinement of water molecules between a surface and a nanosized spherical silicon AFM tip.20 The surfaces investigated were mica, glass, and HOPG. The authors observed strong oscillatory forces due to confinement on both mica and glass surfaces and less pronounced oscillations when the underlying
INTRODUCTION The behavior of water molecules in close proximity to surfaces has been the subject of numerous studies, primarily due to water molecules playing significant roles in many processes of practical importance. These include acting as potential energy barriers to biomolecule adsorption and influencing the stability of colloidal particles in aqueous suspensions.1,2 Water molecules within a few nanometers of a surface exhibit behavior distinct from that of the bulk liquid, with the molecules ordering into discrete layers known as solvation or hydration layers.3 These solvation layers may exist as a result of hydrogen bonding between the surface and the water molecules but they may also arise when the liquid is physically confined between two surfaces. This ordering at the solid−liquid interface has been shown to extend for multiple layers into the bulk and results in an oscillatory or solvation force profile, which has been measured extensively with both the surface forces apparatus3,4 (SFA) and atomic force microscope (AFM),5−7 and is consistent with results from molecular dynamics simulations.8,9 In SFA measurements, the oscillatory force profile is detected when two macroscopic and atomically smooth mica surfaces, with typical radii of curvature of the order of centimeters, are brought into contact. The molecules of the intervening fluid are squeezed out in discrete molecular layers as the two surfaces approach due to density fluctuations (stratification) in the confined liquid.4 A small degree of roughness and/or the use of nonrigid surfaces has been shown to be sufficient to disrupt continuous layering at the interface © 2012 American Chemical Society
Received: January 13, 2012 Revised: March 28, 2012 Published: April 2, 2012 6589
dx.doi.org/10.1021/la300155c | Langmuir 2012, 28, 6589−6594
Langmuir
Article
imaging and force experiments. All measurements were carried out in a background 10 mM phosphate buffer solution (PBS, Sigma (Dublin, Ireland)) at pH 7.4, which was made up with high-purity Milli-Q water (resistivity of 18.2 MΩ·cm at 25 °C) and small additions of 1 M NaOH. The background salt was used to screen any excess charge in the system which could mask the presence of solvation forces. Imaging and Force Measurements. Images and force data were obtained using a bespoke, liquid-environment, low-noise AFM.21 All measurements were made with supersharp silicon cantilevers (SSNCH, Nanosensors Switzerland) with a typical spring constant of ∼20 N/m, and a resonance frequency of ∼135 kHz in phosphate buffer solution (PBS). The spring constant and resonance frequency were individually characterized for each measurement. The cantilever was oscillated with an amplitude of ∼0.2 nm. For increased imaging stability, the cantilever was excited at the second harmonic, which had a typical frequency of ∼830 kHz for all imaging experiments. In FMAFM, changes in the resonant frequency and damping of an oscillating cantilever occur in response to changes in the conservative and dissipative tip−sample interactions, respectively. The driving frequency and driving amplitude are controlled by two independent feedback loops, which ensure that the cantilever oscillates with constant amplitude and is driven at the resonant frequency. For the determination of FM feedback loop gains, we use a digital-tuning methodology developed by Kilpatrick et al. in which a robust tuning algorithm is used to calculate initial stable feedback gains for the FM feedback loops for a given cantilever environment, greatly simplifying the operation of the FM-AFM system.25 Under conditions where the cantilever is driven at the true resonant frequency with constant amplitude, the relative contributions of the conservative and dissipative components of the tip−sample interaction can be independently determined. For the data presented in this paper, the frequency shift is converted to a force as a function of an apparent tip−sample separation using the formula proposed by Sader and Jarvis.26 The force measurements were taken at three different locations on the substrate, in order to reduce any influence of local surface features. In addition, each experiment was repeated at least three times. For imaging, an additional feedback loop is employed to control the tip−sample separation by modulating the scanner height in response to changes in frequency shift. All AFM measurements were made at room temperature. Blind Tip Reconstruction. Tip profiles were obtained by employing blind tip reconstruction to atomic resolution (15 × 15 nm) images of mica in PBS using the FM-AFM imaging parameters outlined above. For the tip profiles presented in this study, the same tip was used for force profile measurements followed by mica imaging. The tip profiles obtained are considered to be representative of that used for the force profile data collection. Blind tip reconstruction calculations were performed using the partial method described by Villarrubia27 as implemented in Gwyddion28 using images which were first-order plane fitted and first-order flattened. Due to the susceptibility of blind tip reconstruction to noise,29 images were also median flattened with a 1 nm kernel and then median filtered (7 pixels). Reconstructions were obtained for 20 × 20 pixels (0.59 × 0.59 nm) with a 0.1 pm noise threshold. Tip radii were determined by fitting the X profile of the reconstruction, as this represents the upper limit to the tip radius, and were found to be 139 ± 65 pm and 131 ± 99 pm for the hydrophilic and hydrophobic tips, respectively. Fits were constrained to 9 points for the hydrophilic tip and 10 points for the hydrophobic tip to ensure that the fit was representative of the local tip radius.
surface was HOPG. The tip material was hydrophilic, and no attempts were made to investigate the effects of tip hydrophobicity, as the study concentrated on the effects of surface chemistry on liquid confinement, neglecting the effects of tip chemistry on the observed force profiles. Here we systematically investigate the interaction of different hydrophilic and hydrophobic tip−sample combinations using a home-built, low-noise AFM operated in frequency-modulation (FM) mode.21 The tip and surface chemistries have been controlled through the preparation of clean hydrophilic oxide layers and silanization chemistry.22 The main advantage of the FM-AFM technique is that it utilizes a stiff oscillating cantilever to obtain very high force sensitivity. This enables the measurement of the complete force profile, including strong, short-range attractive interactions with high force gradients, while avoiding the “jump to contact” commonly observed using other modes of operation which rely upon the use of lowstiffness levers to optimize force sensitivity. In addition, the enhanced cantilever displacement sensitivity of the low-noise AFM allows the cantilever to be oscillated at small amplitudes (
