Model may expose how friction lets loose.

Friction is at play wherever surfaces meet, and it always begins with atoms. Atomic theories of the phenomenon usually focus on interatomic bonds, shared vibrations, and other surface-to-surface interactions as friction's ultimate source.

Now, two Texas-based physicists have modeled surface slippage--friction's retreat--as bands of atoms in the top surface momentarily leaping up from the underlying surface. Millions of such ripples propagate simultaneously along the interface when, for instance, a book slides on a table, they say.

Scientists have long known that friction on the scale of books and tables obeys simple laws. As French physicists Guillaume Amontons and Charles-Augustin de Coulomb established in the 17th and 18th centuries, the sideways force needed to overcome the friction between surfaces is proportional to the forces, such as weight, pressing the surfaces together. Surprisingly, the size of the friction-defeating force is independent of the area of the surfaces in contact.

For years, physicists have tried to explain this large-scale behavior in terms of atomic-scale events. They've had some success by portraying surfaces as jagged on an atomic scale. That way, very little material actually touches. However, scientists still struggle to explain why protrusions from two surfaces would stick together at all.

There's incentive to find out. A better understanding of friction could improve scientists' grasp of countless phenomena, such as engine performance and tool wear. Moreover, friction is particularly vexing for developers of micromachines (SN: 7/22/00, p. 56).

In the new mathematical model, Eric Gerde and Michael P. Marder, both of the University of Texas at Austin, build upon the physics of how cracks form and propagate through solids. Think of a bump in a rug, says Marder. As people know from everyday experience, pushing such bumps along can move a big rug over a floor.

Something similar may be happening at the atomic scale between sliding surfaces. Marder says that the combination of downward and sideways forces on an object sliding along an underlying surface can translate into upward forces that open "cracks" at the interface, akin to bumps in a rug. These cracks amount to a series of arches, each a few atomic diameters across. As these waves of separation advance along the interface, the overlying surface comes back down behind each wave and reconnects with the surface below.

A plus for this hypothesis, presented in the Sept. 20 NATURE, is that it predicts the simple relationship between compressive forces, like weight, and frictional forces. Yet it doesn't require the surfaces to be rough on an atomic scale, as previous models do. Other scientists have theorized about such cracking before and have even seen hints of it in the lab and at earthquake faults. Previous attempts to mathematically represent surface-separating ripples, however, have led to nonsensical implications, including solid surfaces passing through each other like ghosts.

David A. Kessler of Bar-Ilan University in Ramat-Gan, Israel, calls Gerde and Marder's work a "mathematical tour de force" in a commentary in the same issue of NATURE. "Whether it also helps to solve the problem of friction ... remains to be seen," he adds.

Marder concedes that his team's current model may not explain everything about surface sliding, and he looks to experiments planned by Jay Fineberg of the Hebrew University in Jerusalem to fill in gaps or show where the model falls short. Fineberg and his colleagues plan to use high-speed cameras and acoustic sensors to seek signs of the friction-releasing ripples on the surfaces of transparent pieces of Plexiglas as they move over glass surfaces.