Sliding Nanofriction in Low Dimensional Model-Systems

Thanks to novel experimental techniques, physicists are now able to characterize the dynamics of interacting surfaces down to molecular length scales. This possibility has brought fresh interest in the field of friction and has opened a new branch of research, nanotribology, whose aim is the study of tribological properties of sliding systems in terms of fundamental atomistic dissipative mechanisms. Far from being a mere academic problem, understanding and controlling friction at the nano-scales can have a great impact on many technological applications, from energy conversion and saving, to transportation and micro-machining. To reach this goal it is essential to develop theoretical tools able to tackle this problem. In absence of a general theory of friction, molecular dynamics (MD) simulations represent, at the moment, one of the most powerful approaches able to explain and predict the behaviour of nanoscale interfaces. Within the context of nanotribology, this thesis deals with some fundamental aspects of friction between dry crystalline surfaces. Anticipating experiments, it reports results obtained by means of realistic MD simulations of artificial model-systems currently under experimental investigation for application in nanotribology. Qualitatively, the sliding properties of crystalline interfaces can be interpreted based on the mutual interaction between the potential energy landscapes generated by the two touching lattices. These may be described as periodic, two dimensional sequences of wells and hills, corresponding to repulsive and attractive regions of the surfaces. Commensurate geometries correspond to atomically locked configurations, where the atoms of each interacting plane adapt into the wells of the potential landscape generated by the other one. When driven out of equilibrium by an applied shear force, this kind of atomic locking, or interdigitation, generally determines mechanical instabilities, which in turn lead to violent depinning events and high dissipation. Atomic locking may still occur, but may also be avoided in incommensurate interfaces, in which case friction shows more smooth and gentle sliding regimes. The ordered atomic arrangement of crystalline surfaces therefore offers a peculiar way to reduce friction, that is by controlling the geometry of the interface, e.g., by rotating relative to each other two originally aligned and commensurate surfaces. The above picture applies to clean, chemically inert, and atomically flat crystal surfaces. To describe in a quantitative way real systems, one has to take into account many other effects whose interplay determine the overall tribological response. They include elasticity of the surfaces, plastic deformations, and the presence of steps, defects and impurities, not counting chemical interactions and other processes involving the direct excitations of electronic degrees of freedom. From the experimental point of view, the fine features of the interface are hardly accessible because buried. Usually only macroscopic average values of some arising physical quantities are measured, which hinders the possibility to keep track of each distinct mechanism at play. From the theoretical point of view, developing a general theory different from brute-force with quantitative predictive power is also difficult, and phenomenological models are usually adopted which apply only to a reduced number of cases. However, there exist a class of real systems where these complications are absent or mitigated, allowing for detailed experimental and theoretical investigations. On one hand, one (1D) and two dimensional (2D) artificial crystals of charged particles trapped inside optical lattices offer the possibility to study the dynamics of ideal crystalline interfaces with all interface parameters under control, including commensurability and substrate interaction strength. On the other hand, surface science and ultra high vacuum techniques supply clean and atomically flat substrates, suitable for the study of the sliding properties of two dimensional monolayers of adsorbate atoms. Both these systems are very well characterized, and allow for accurate realistic MD simulations where geometrical effects, and interface elastic and plastic deformations effects are investigated in great details. In view of future experiments, this thesis reports results of MD investigations of some fundamental tribological aspects in low dimensional incommensurate interfaces formed by: (i) 1D cold-ion chains trapped in optical lattices, (ii) 2D charged colloids monolayers interacting with laser-induced periodic potentials, (iii) 2D islands of rare gas atoms physisorbed on clean metallic substrates. These simulations show that incommensurate linear chains of trapped cold ions display a rich dynamics when forced to slide over a periodic corrugated potential. That suggests that they can be adopted to investigate in detail the external-load dependent transition between the intermittent stick-slip motion and the smooth sliding regime, as well as the precursor dynamics preceding the onset of motion. Both of them are shown to display paradigmatic behaviours observed in sliding contacts at any length scales. Incommensurate two dimensional interfaces realized by colloidal crystals in periodic fields have been simulated to study the pinning transition from the locked (finite static friction) state, to the ``superlubric'' -- zero static friction -- free-sliding state, which is predicted to occur as a function of decreasing substrate potential strength. In a range of parameters compatible with recent experiments, that is a first order (structural) phase transition of the colloid monolayer, showing analogies with the superlubric to pinned Aubry transition" extensively studied in the one dimensional discrete Frenkel-Kontorova model of dry friction. Moreover, realistic simulations show that relative misfit rotations between the colloidal slider and substrate may significantly affect the dissipation under steady sliding even in genuinely incommensurate geometries, by changing the degree to which the soft deformable slider interdigitates within the hard, non deformable optical substrate. This is a result of more general validity, since a similar mechanism must be at play in any incommensurate 2D crystalline interface. Finally, in order to understand the persistent static friction force observed in quartz crystal microbalance experiments on highly clean surfaces, extensive numerical simulations of substrate-incommensurate model rare-gas islands have been performed, which, in absence of any other sources of pinning, describe how the island edges alone may play the ultimate role in determining the overall barrier preventing the onset of global sliding.