ARTICLE pubs.acs.org/Langmuir
Adhesion Hysteresis from Interdependent Capillary and Electrostatic Forces Nathan W. Moore* Radiation Effects Research Department, Sandia National Laboratories, P.O. Box 5800, MS-1159, Albuquerque, New Mexico 87185, United States ABSTRACT: Adhesion hysteresis commonly occurs at the nanoscale in humid atmospheres, yet mechanisms are not entirely understood. Here, the adhesion forces between silicon (111) oxide surfaces and tungsten oxide probes have been examined using interfacial force microscopy. The results show that the adhesion forces during surface approach and separation differ not only in magnitude but also in mechanism, arising mainly from capillary and electrostatic forces, respectively. Surface contact leads to a transient intersurface potential on dewetting. This mechanism of adhesion hysteresis differs in not relying singly on hysteretic wetting. Furthermore, by biasing the surfaces, nonadditivity is demonstrated between the capillary and electrostatic forces at the onset of condensation. These results hold important implications on the interpretation of force in nanoprobe geometries in humid atmospheres.
’ INTRODUCTION Adhesion due to condensation of water vapor or wicking of an adsorbed water film between hydrophilic surfaces occurs widely in nature, allowing, for example, insects to walk up smooth walls and for some shorebirds to feed via capillary ratcheting, with potential consequences when the surface tension of seawater changes from chemical or oil spills.13 The nano revolution has fueled a myriad of similar uses for capillary forces, e.g., as a gripping mechanism in microassembly and for moving materials in labon-a-chip technologies, as well as in dip-pen nanolithography.4,5 Negative consequences of capillary adhesion include image degradation in atomic force microscopy (AFM), stiction in microelectromechanical systems (MEMS), and clogging of granular flows.3,6 A fundamental understanding of capillary adhesion is therefore of broad interest, particularly at the nanoscale where macroscopic intuition can sometimes fail.7 Capillary condensation occurs when vapor from the atmosphere nucleates into a liquid bridge between opposing surfaces separated by less than a critical distance. In many studies, this critical separation has been predicted on the basis of the macroscopic KelvinLaplace (KL) equation, which relates changes in the pore fluid pressure to the vapor pressure.8 However, the applicability of this equation for gaps smaller than a few nanometers remains debated.3,7 Directly conflicting with the assumption of thermodynamic equilibrium in the KL equation are experiments and simulations showing hysteresis in the force between surfaces bridged by liquid capillaries (e.g., refs 913). That capillaries can lead to hysteretic adhesion is also apparent when applying the theory of Johnson, Kendall, and Roberts (JKR)14 and its variants,6,8,1518 which suggest a linear correlation between the pull-off force and the thermodynamic work of adhesion between frictionless, axisymetric, elastic bodies, even when the solids are bridged by a liquid capillary r 2011 American Chemical Society
in equilibrium with its vapor.1921 However, experiments involving nanoscale capillaries have shown that such correlation does not always occur.10,13 The reasons for these discrepancies have yet to be clarified. While numerous theories describe the formation and spreading of condensed capillaries, in experiments most attention has focused on the critical pull-off force required to separate two asperities bridged by a capillary, e.g., using AFM or the surface forces apparatus (for a review of capillary forces, see ref 3). However, extracting the magnitude of the capillary force from the pull-off force requires assuming that the other intersurface forces, e.g., electrostatic or hydration forces, are either constant or unimportant, which may not always be justified.22,23 As well, the pull-off force may be governed by multiple hysteretic processes that are not easily distinguished. In ultrahigh vacuum (UHV), adhesion hysteresis can arise from charge transfer from triboelectrification, plastic deformation, material transfer, differences in work function, contact angle hysteresis in solidsolid wetting, and induced molecular vibrations (beating) of the solid surfaces.2428 In the atmosphere, adhesion hysteresis can also arise from contact angle hysteresis in liquidsolid wetting, vapor adsorption hysteresis, squeeze-film damping of a gas or condensate, energy dissipation through vibration of an intervening liquid capillary, and aqueous reactions that change the chemical or physical structure of the interface.9,12,2432 Here, it is shown that adhesion hysteresis between a probe tip and a surface can also arise through charge transfer facilitated by a nanoscale capillary between oxide surfaces. Unlike the mechanisms Received: January 4, 2011 Revised: February 17, 2011 Published: March 11, 2011 3678
dx.doi.org/10.1021/la200043a | Langmuir 2011, 27, 3678–3684
Langmuir
ARTICLE
Figure 1. Schematic of the IFM differential-capacitor force-feedback sensor, showing the capacitor gaps on each end of the common plate, with capacitances C1 and C2.
described above, here the adhesion forces are mechanistically distinct during the approach and separation of the two surfaces. While capillary attraction can dominate the force on approach, strong electrostatic forces can be seen following solidsolid contact. The intersurface potential decreases quickly on dewetting of the nanoscale capillary so that the process is highly repeatable. Besides demonstrating another mechanism for how a nanoscale capillary can lead to adhesion hysteresis, the results explain why correlation is not always seen between the pull-off force and the actual work required to separate nanosized elastic asperities.
’ METHODS AND MATERIALS Interfacial force microscopy (IFM) has recently been reviewed.33 In short, the IFM microfabricated force sensor maintains the position of a probe tip by tracking the tip’s displacement in response to a force as a differential change in the capacitance of two capacitors mounted on each side of a teeter-totter to which the tip is attached (Figure 1). The dc voltage on the capacitor pad behind the tip is continually adjusted using a feedback controller to eliminate the tip displacement, while the applied voltage allows a measure of the force on the tip. With this scheme, the IFM sensor is essentially noncompliant about its rotational axis yet can measure attractive and repulsive forces at all tipsample separations. Instrumental compliance along the normal and tangential coordinates is unimportant over the force range in this study.34 This scheme allows the tip to be kept perpendicular to the sample surface, minimizing tip bending and triboelectrification upon contact.26 An interfacial force microscope built at Sandia National Laboratories (Albuquerque, NM) was used with the following improvements. The voltage proportional to the error in the angular position of the sensor common plate (viz., the input to the proportionalintegralderivative (PID) feedback circuit) was calibrated in terms of the effective tip displacement along the normal coordinate by resting the tip on a surface with the sensor feedback disabled to render the tip relatively stationary and then moving the sensor body toward the surface using a calibrated piezo tube scanner, so that the relationship between the common plate displacement and open-loop error (voltage) could be determined in the linear regime. The calibrated error in the feedback response was then recorded with each measurement and the gains of the feedback loop reduced until the jitter in the tip position was
