[SY-O3] Scale Dependence of Friction: How Elasticity Destroys Superlubricity
Invited
Friction in single-asperity contacts is studied as a function of contact radius a, substrate stiffness G, atomic structure, adhesive strength and adsorbed layers. Friction between bare, rigid surfaces can be obtained by a simple sum over atomic forces. When surfaces are aligned and have the same periodicity, the forces add in phase and the friction force F rises linearly with area A. When the two surfaces are misaligned or disordered, so that there is no common periodicity, forces add out of phase and F~Ax with x less than or equal to 1/2. This is known as structural superubricity and implies that friction vanishes in the limit of large contact sizes. Most surfaces do not share a common period but friction is almost always observed at macroscopic scales. We use an efficient Greens function technique to study contacts with dimensions of micrometers while resolving atomistic interactions at the surface. A new formulation that includes dynamic effects explicitly will be described. For small tips and high loads the contact area follows predictions for contact of rigid surfaces, x=1 for identical aligned surfaces, x=1/2 for random surfaces and x=1/4 for incommensurate crystals. Elasticity becomes important when a exceeds the core width bcore of interfacial dislocations. For a>bcore parts of the contact can advance independently. The friction for identical aligned surfaces drops as a power law and then saturates at the Peierls stress for edge dislocations. The friction on incommensurate and disordered surfaces saturates at nearly the same value. Thus for all geometries x=1 in large contacts. While this means that elasticity destroys superlubricity, the friction between bare surfaces drops exponentially with the ratio of substrate stiffness to local interfacial shear stress. In contrast, when adsorbed layers are included between surfaces, all geometries have x=1 with a large prefactor.