Molecular Transport Group

Electronic and conducting properties of molecular nanojunctions


In the Molecular Transport group, we are interested in electronic processes in molecules at surfaces or interfaces. Our main focus is the theory of electron transport across single molecules. However, we are also interested in molecular surface science and, in a broader scope, in the electronic structure of molecules in different environments. Our work is rather interdisciplinary and involves concepts and approaches from physics, chemistry and materials science. We use computational methods based on Density-Functional Theory. Atomistic simulations provide a picture of the main physical mechanisms that govern electronic structure of current flow in these systems. Simulations also rationalize and predict experimental findings.

The group works either alone or in collaboration with experimental partners. In recent years, we have focused on three main areas:

1) Design of single molecule circuits

The electron transport properties of single molecules depend sensitively on the chemical backbone of the molecule and on how it is bonded to the source and drain electrodes via chemical linker groups. We have investigated several examples of how the conducting properties can be tuned through a judicious choice of the molecule:

Antiaromaticity and conductance

Aromaticity is a key concept in chemistry. It is described by Hückel’s rule, which states that a flat, cyclic, conjugated molecule is aromatic if it has 4n+2 electrons in the pi system (where n is an integer), antiaromatic if this number is 4n, and nonaromatic otherwise. Since the 1970s antiaromatic molecules had been thought to possess outstanding conducting properties but they are unstable and difficult to synthesize. Together with experimental partners, we studied the conductance of a genuinely antiaromatic molecule for the first time . We showed that the antiaromatic molecule was much more conducting than its aromatic counterpart due to a more favorable alignment of molecular orbitals at the junction. Our work showed for the first time the potential of antiaromatic species in molecular nanoelectronic studies.

Metal-molecule contacts

We also investigated extensively the role on conductance of metal-molecule contacts and the influence of chemical linker groups. Sulfur linkers are the most commonly used since the birth of molecular electronics, but they can bind to gold electrodes in several possible configurations whose conductance in these geometries differ by 2 orders of magnitude. However, these binding sites cannot be measured easily and there can be fluctuations between them under usual experimental conditions. Together with international partners, we showed that a combination of experimental techniques and simulations could identify the adsorption site of these molecular junctions and thus help overcome a major challenge in molecular-scale nanoelectronics.

We have also paid particular attention to N-heterocyclic carbenes (NHCs), a rather recent class of molecules first synthesized in 1991. These carbenes have a divalent carbon atom with a 6-electron valence shell and bind strongly to Au. In a series of papers, we saw how the LUMO position and conductance depended strongly on the atomic structure of the electrode tips, which has consequences for the Joule heating of the junction (see below). Together with experimental colleagues, we demonstrated electron transport across NHC-bonded single molecules for the first time.

We also studied highly-conducting Au-C metal-molecule bonds. We realized them by soft ion sputtering in fullerene islands, and proposed a bipodal platform based on biphenylene where Au-C led to a stable and electronically transparent contact to the substrate.

2) Electron-vibration interaction in single molecule junctions

Interaction between tunneling electrons and vibrations localized on the molecule leads to features in the second derivative of the current with respect to the voltage d2I/dV2 at energies of junction vibrational modes. In molecular junctions, usually these features are peaks in the Inelastic Electron Tunneling Spectrum (IETS), but not always. We predicted that in Au-C bonded oligophenyls there is a crossover between IETS dips (for the shorter molecule) to IETS peaks (for longer molecules) by simply increasing the number of benzene rings. We have also worked on the origin of the IETS signal. We developed a theoretical methodology to map where inside the junction each IETS peak is generated, spatially mapping each IETS peak. We exemplified it with experimental colleagues using a CO-functionalized tip on FePc. We explained why submolecular resolution in the IETS amplitude (not frequency!) could be achieved when using one vibrational mode of CO but not another.

3) Current-induced heating (and cooling!) of molecular circuits

In a molecular circuit, electronic and vibrational degrees of freedom can also exchange energy. Energy transfer to vibrations locally heats the conducting molecule, which can become unstable and break if these transfer processes are not balanced by dissipation mechanisms, in what are called vibrational instabilities. Energy transfer from vibrational to electronic degrees of freedom locally cool the junction, since electrons release this energy far away. We have calculated these current-induced energy exchange processes from first principles. An accurate description of the electronic structure under bias is essential to correctly calculate these energy exchange rates. The balance between heating and cooling mechanisms can be tuned by modifying the interface electronic structure. We found that, as the voltage is increased, the molecular junction can be heated (as expected) but also cooled!. This can be influenced by adsorbates positioned close to the conducting molecule, even if they are not bonded to it. Finally, we predicted the conditions for vibrational instabilities (junction breakdown) to be observed. We found that it is a much more general phenomenon that previously thought. We generalized previous works to a broader class of molecules under more general conditions.