Friction, commensurability, and superlubricity

In the limit of clean surfaces, friction has its origins in the microscopic, chemical interactions at the interface between the two objects in question.  One of the more amazing (to me, anyway) consequences of this is the extremely important role played by commensurability between the surfaces.  Let me explain with an example.  Consider a gold crystal terminated at the (111) surface, and another gold crystal also terminated at the (111) surface.  Now, if those two surfaces are brought into contact, with the right orientation so that they match up as if they were two adjacent layers of atoms inside a larger gold crystal, what will happen?  The answer is, in the absence of adsorbed contaminants, the surfaces will stick.  This is called "cold welding".  In contrast, if you bring together two ultraclean surfaces that are incommensurate, they can slide past each other with essentially no friction.  This is called "superlubricity".  Here are two great examples (pdf of first one; pdf of second one) of this.

In this new paper, Liu et al. are able to do some very cute experiments in this regard, looking at the motion of thin graphite flakes (exfoliated from and) sliding on graphite pedestals.  It's clear from the observations that graphite flakes shifted relative to the underlying graphite substrate can slide essentially frictionlessly over micron scales.  Very neat and elegant, and surprising since there is not any rotation at work here to break commensurability.  This is a very firm reminder that our macroscale physical intuition about materials and their interactions can fail badly at the nanoscale.