No abstract

A current induces forces on atoms inside the conductor that carries it. It is now possible to compute these forces from scratch, and to perform dynamical simulations of the atomic motion under current. One reason for this interest is that current can be a destructive force--it can cause atoms to migrate, resulting in damage and in the eventual failure of the conductor. But one can also ask, can current be made to do useful work on atoms? In particular, can an atomic-scale motor be driven by electrical current, as it can be by other mechanisms? For this to be possible, the current-induced forces on a suitable rotor must be non-conservative, so that net work can be done per revolution. Here we show that current-induced forces in atomic wires are not conservative and that they can be used, in principle, to drive an atomic-scale waterwheel.

Recent experiments suggest that gold single-atom contacts and atomic chains break at applied voltages of 1 to 2 V. In order to understand why current flow affects these defect-free conductors, we have calculated the current-induced forces on atoms in a Au chain between two Au electrodes. These forces are not by themselves sufficient to rupture the chain. However, the current reduces the work to break the chain, which results in a dramatic increase in the probability of thermally activated spontaneous fracture of the chain. This current-induced embrittlement poses a fundamental limit to the current-carrying capacity of atomic wires.

The tight-binding (TB) approach to the modelling of electrical conduction in small structures is introduced. Different equivalent forms of the TB expression for the electrical current in a nanoscale junction are derived. The use of the formalism to calculate the current density and local potential is illustrated by model examples. A first-principles time-dependent TB formalism for calculating current-induced forces and the dynamical response of atoms is presented. An earlier expression for current-induced forces under steady-state conditions is generalized beyond local charge neutrality and beyond orthogonal TB. Future directions in the modelling of power dissipation and local heating in nanoscale conductors are discussed.

Using an exact single-particle scattering formalism we have carried out the first calculation of the conductance of a metallic contact, on the basis of a full dynamic simulation of the evolution of its atomic structure. We find that the contact area evolves discontinuously through a series of mechanical instabilities, which is reflected in a stepwise variation of the conductance also seen in recent experiments. Importantly, we find that the conductance is not simply proportional to the size of the contact and, therefore, the conductance per atom is not constant.

No abstract

We derive and employ a semiclassical Langevin equation obtained from path integrals to describe the ionic dynamics of a molecular junction in the presence of electrical current. The electronic environment serves as an effective nonequilibrium bath. The bath results in random forces describing Joule heating, current-induced forces including the nonconservative wind force, dissipative frictional forces, and an effective Lorentz-type force due to the Berry phase of the nonequilibrium electrons. Using a generic two-level molecular model, we highlight the importance of both current-induced forces and Joule heating for the stability of the system. We compare the impact of the different forces, and the wide-band approximation for the electronic structure on our result. We examine the current-induced instabilities (excitation of runaway "waterwheel" modes) and investigate the signature of these in the Raman signals.

AbStrad. We develop a general fOIlUd~on of the problem Of elastic transpori between two semi-infinite systems, connected by a system of finite size, and derive expressions for the current in and the differential conductance of such a circuit in lhe limit of zero interactions between the canien. These exppressions are exact in the applied voltage, the coupling of the components of the circuit, and lhe tempen" of lhe drcuiL We then apply OUT mulls in a tight-binding approximation to three specific cases: the one-atom contact, the finite. disordered one-dimensional chain, and the generalized stacldng fault.