PerspectiveMaterials Science

Understanding friction in layered materials

See allHide authors and affiliations

Science  08 May 2015:
Vol. 348, Issue 6235, pp. 632-633
DOI: 10.1126/science.aab0930

As we move to the digital age, we may forget that a lot of correspondence was accomplished by pencil and paper. Nonetheless, when we do still write with “lead” pencils, we are making use of the weak interlayer forces in graphite to strip off tiny flakes of graphite and deposit them on paper in a complex interplay between adhesion and friction. Understanding the nature of such weak interlayer forces and the attendant opportunities afforded by such understanding are becoming more important with the advent of nano- and microelectromechanical systems (NEMS/MEMS); machines whose moving components are at the nanometer and micrometer scales embodied in their names. Because the surface area of such components is much larger than their volume, surface forces (for example, van der Waals, electrostatic, and capillary forces) become the dominant loading to be considered in the design and reliability of NEMS and MEMS devices (1). On page 679 of this issue, Koren et al. (2) have devised an innovative experiment that quantifies our day-to-day experience with lead pencils, but with much higher fidelity than we might expect. They show that two adjacent planes of carbon atoms can be sheared in highly ordered pyrolytic graphite (HOPG) and observe the interplay between adhesion and friction forces in this layered material. This experiment with HOPG can be considered as a model for two planes of graphene or other so-called two-dimensional materials sliding over one another. The findings are exciting because they point to new ideas for data storage at small scales, as well as novel actuation concepts in electromechanical devices.

Taking a measure of friction

Illustrating the method of Koren et al. with a toy model of a layered system under lateral shear and torque.

ILLUSTRATION: P. HUEY/SCIENCE

The challenge of measuring relatively weak forces between components has been addressed by a number of instrumented indentation and lateral loading devices, which all make use of probes that are brought into contact or near contact with the target surface while the normal and lateral (friction) load on and displacement of the probe are measured. With sharp enough probes, it is possible to observe oscillations in the lateral force-displacement response whose maxima and minima, respectively, coincide with the onset of slipping and attachment of the probe to the target surface and were commensurate with the interatomic spacing of the mica surface that was being probed (3). The novelty of the experiment by Koren et al. lies in the use of an atomic force microscope to apply a lateral shear force to disks (mesas) of HOPG. The disks, 50 nm thick and with radii that ranged from 50 to 300 nm, were sheared along the basal plane of the graphite, about 10 nm from their bases. A classic frictional lateral load-displacement response was observed, which was largely reversible except for dynamic dissipative events that occurred during sliding. They can be seen as slight differences in the force-displacement response during sliding in opposite directions. These latter events were related to slight rotational misalignments between the graphite lattices above and below the slip plane. In addition, in a second set of experiments, when a torque was applied to the mesas, stable orientations were observed every 60°, in registration with the sixfold symmetry of the hexagonal carbon atom lattice. A third set of experiments, with a structure exhibiting mechanical bistability, demonstrated the potential to produce cell structures for memory devices that are entirely mechanical in operation. The various equilibrium states of these tiny cells could then be used as switches in such memory devices.

The idea behind the method developed by Koren et al. can be illustrated more easily if we consider, for example, a toy system consisting of ping-pong balls glued together in a hexagonal pattern (close packed) in two planes that are placed on top of one another (see the figure). The in-plane attachment is much stronger than the interaction between the planes, just as in layered, two-dimensional materials. One can imagine sliding or rotating the planes relative to one another and finding a series of equilibrium positions.

Another opportunity afforded by the experiment of Koren et al. is that it may present a close link to multiscale modeling from atomistic to continuum scales in that the interface or slip plane was free of any contamination and the geometry was relatively simple, which is not usually the case for contact pairs. Although atomistic or molecular analyses have provided insights into deformation and dissipation mechanisms and other effects that are difficult to separate out experimentally, one-to-one comparisons are limited by the high shear velocities and nanometer-scale probe tips that are currently accessible to analysis. Some progress in this direction has been provided by quartz microbalance experiments (4) and molecular analyses (5) of krypton monolayers sliding on gold, where the close agreement between simulation and experiment identified mechanisms for phononic contributions to friction. Molecular self-assembled monolayers provide another, slightly more complex avenue for linking experiments and analysis in addition to having technological relevance to MEMS. The issue of time and spatial scaling in atomistic analyses of friction between self-assembled monolayers has been identified as one of the current challenges in single asperity friction (6). One possible way of addressing this issue is through hybrid analyses with schemes that combine molecular analyses with phenomenological potentials (7, 8) that have allowed typical sliding speeds to be accessed.

Koren et al. have come up with a novel experiment that provides a better understanding of friction in layered materials. Although graphite was the subject of this study, the extension to the plethora of two-dimensional materials that are becoming available is clear. The authors have also provided much food for thought in developing potential applications of the observed phenomena to novel memory devices and machines.

References and Notes

  1. Acknowledgments: Supported by the National Science Foundation and the Basic Energy Sciences program of the U.S. Department of Energy.

Navigate This Article