Slippery when wet

Nanoscale friction is governed by laws that are different from those that we experience on the macroscopic scale. A new study using atomic force microscopy shows that condensation of liquid bridges between the sliding surfaces plays a key role in atomic-scale friction - these bridges produce a capillary force that acts to increase static friction, but which decreases once the surfaces start moving.

30 April 2002

Charlene Lobo

Capillary liquid bridges form in the contact area between the AFM tip and the sample surface, producing a capillary force between the two surfaces.

We all know how friction works on a macroscopic scale: it increases with applied load, and is independent of both contact area and velocity. But it is friction on the atomic scale that affects technological devices with miniature moving parts, such as microelectromechanical systems and hard-disk drives. Nanoscopic sliding friction in high-vacuum environments has been studied by atomic force microscopy (AFM), and attributed solely to a mechanism known as 'atomic stick-slip', in which the friction force between the AFM tip and the sample surface repeatedly builds up and then quickly slips at each atomic site. The velocity dependence of nanoscopic friction has also been investigated by AFM. In ultrahigh vacuum and controlled environments, nanoscale friction was found to increase logarithmically with velocity, as expected for an atomic stick-slip mechanism. However, at ambient conditions, these studies have produced some conflicting results, and the velocity dependence has been found to vary with the material being studied. Although humidity has been known to play a role in nanoscopic friction, its effect has not been fully understood. Now, in Physical Review Letters, Elisa Riedo and colleagues at the Ecole Polytechnique Fedéralé de Lausanne have developed a model for nanoscopic friction based on the condensation of capillary liquid bridges between rough sliding surfaces, enabling them to quantify the effects of humidity and surface wettability on nanoscopic friction.

Riedo and colleagues studied sliding friction on CrN and diamond-like carbon (DLC) films, both of which are commonly used as hard coatings in applications where small friction forces are required. Whereas DLC films are smooth and hydrophobic, the CrN films are rough, and have a surface wettability (hydrophobic or hydrophilic) that depends on the growth temperature. The authors investigated the velocity dependence of sliding friction for each of these materials during AFM in a controlled-humidity environment by varying the scan frequency of the silicon AFM tip. Simultaneous recording of the normal and lateral deflections of the cantilever enabled determination of the normal and friction forces between the tip and sample.

For both hydrophilic (high-temperature CrN) and hydrophobic (DLC) surfaces, the authors found that friction increased logarithmically with relative humidity. This dependence of the friction force on the relative humidity was found to have its origin in the variation of the adhesion force - the attractive or repulsive force between atoms on the surface and those on the end of the AFM tip. In a humid environment, this adhesion force results not only from the direct adhesion of the Si tip to the sample surface, but also from the capillary force: that is, the force resulting from the condensation of capillary liquid bridges at many different points along the tip-sample contact area. When the tip is at a fixed position over the sample, this capillary force increases as a function of time, because of the continuous formation of liquid bridges between the two surfaces. However, when the tip is being scanned over the sample, the time available for liquid bridges to form, and thus the capillary force, decreases as a function of velocity. Hence the capillary force acts in opposition to atomic stick-slip, causing a logarithmic decrease of friction with increasing velocity. This effect is much greater on hydrophilic surfaces than on hydrophobic surfaces, resulting in a large variation in the velocity dependence of sliding friction with humidity.

Riedo and colleagues have demonstrated that capillary condensation due to ambient humidity must be considered along with atomic stick-slip in order to correctly describe nanoscale friction. Their work has conclusively shown that for hydrophilic surfaces, nanoscopic friction decreases with scan velocity, whereas it increases with scan velocity for hydrophobic surfaces. The explanation of this phenomenon in terms of thermally activated capillary condensation provides a leap forward in the understanding of atomic-scale friction, and should lead to improvements in the design and operation of micromechanical devices.