Pino d'Amico, Klaus Richter, Dmitry Ryndyk

Figure 1
A carbon nanotube-C60-carbon nanotube molecular junction.

The field of molecular electronics emerged from the idea to use objects on the molecular-scale as the active components of electronic devices, which would operate identically or analogous to transistors, diodes, conductors and other key elements of today's microcircuits. The success of this proposal paves the way to smaller-denser-faster, and possibly cheaper computer chips. Conventional semiconductor devices are based on the `top-down' miniaturization making them smaller and smaller. Such fabrication techniques have inherent limitations and imperfections at nanometer scales. In contrast, molecular electronics is based on the `bottom-up' manufacture philosophy. The underlying principle is to use directly synthesised molecules or supramolecular structures as circuit elements and, therefore, has the advantage of increased reproducibility, exploitation the diverse electronic properties of molecular complexes, and the capability of synthetic (bio)chemistry. Clearly, progress in this direction requires interdisciplinary activity and poses an active interplay between fundamental and applied research.

The original proposal of a molecular rectifier dates back to 1974 but significant progress on transport across a single molecule has been demonstrated experimentally only in recent years. This owes to the advances in self-assembly techniques, end-group modifications, scanning probe, and break-junction techniques, which allow atomic-scale control and positioning of single molecules and their assemblies. In a typical experimental setup a molecule bridges two metallic electrodes. Proposals and studies of molecular wires range from `simple' molecules to DNA strands. As an example, rectification and negative differential resistance mediated by single molecules have been achieved. These functions are of fundamental importance to electronic devices, and gave new momentum to this field of high promises. Some optimistic views project that in a decade, much of our silicon technology will have been replaced by what today seems an unfulfilled dream.

On the theoretical side, although there is no complete understanding at the present, several efforts have pointed to a number of factors that determine the conductance in such systems. The most important include the molecular electronic structure related to the resonant spectrum and the distribution of the wavefunction along the molecular bridge, the location of the electrodes' equilibrium Fermi energy with respect to the molecular electronic spectrum, charging effects, and the electron-phonon coupling in wires of low conduction. Nevertheless, what crucially matters is the interface between the electrodes and the molecular complexes, which inter-relates the above classified effects.

The theoretical description of electron transport at the molecular scale requires two conceptual steps: a)departing from the traditional study of semiconductor devices using the effective mass equation and the Boltzmann kinetic theory and b)understanding the role of the microscopic electronic structure of the electrodes-molecule hybrid system. The first step has been largely tackled in the past two decades by the emergence of mesoscopic solid-state physics, which succesfully described a variety of nontrivial quantum effects arising in lithographically patterned submicron semiconductor devices. The second is what makes the molecular electronic devices more different and richer than their mesoscopic counterparts. To study in detail this point, several approaches have been developed, which range from tight-binding or semi-empirical to first-principle. As stressed above, it is important to realise that albeit a molecular device is typically divided into three parts, the donor and acceptor electrodes and the molecular compound serving as a bridge, an account of the electronic structure of the system as a whole is required. Such a description coupled to a theory of transport for both zero and finite bias at any temperature, presents another fundamental theoretical challenge.

Our research is focused on several issues concerning the fingerprints of electrodes with size comparable to the molecular bridge, to the conductance of the junction. In contrast to bulky electrodes which are described by a smooth local density of states, bearing also a continuum of possible conducting channels, in low-dimensional transport the geometry of the contact crucially determines the measured conductance. In particular, we study interfacial effects for pure carbon devices consisting of carbon nanotube electrodes attached to a fullerene (C60) molecule (see figure 2). Carbon tubules have been the focus of much experimental and theoretical activity and are the most prominent wiring elements of molecular circuits. On the other hand, C60 is also one of the most intensively studied molecules. To investigate these hybrids, we use both simple tight-binding models and density-funtional-based theories in collaboration with Giovanni Cuniberti (MPI for the Physics of Complex Systems) and the group of Prof. Ruediger Schmidt at TU-Dresden. Our purpose is to: a) advance our qualitative understanding, especially for mesoscopic electrodes, b) reveal the properties of 'all-carbon' molecular electronics together with discussing designs and possibilities, and c) develop new methods for quantitative description of transport at the molecular scale.