Important impact of the experimental platform on the efficient control of electronic and vibrational properties of molecular junctions

Abstract

A recent work reported simultaneous surface-enhanced Raman scattering (SERS) and transport current-voltage I-V measurements in electromigrated molecular junctions based on C60. The vibrational frequencies measured there were found to be slightly but significantly shifted by the applied bias. This effect was attributed to a partial reduction of the C60 molecule. The fact that a complete redox process could not be achieved in that experiment raises the question addressed in this work on whether the experimental platform utilised (electromigrated junctions) represents the most favourable setup to achieve a(n almost) complete bias-driven reduction. The answer presented here is that (i) no matter how high the voltage is, reduction cannot exceed 50% in an experimental setup where the molecule is symmetrically coupled to the (source and drain) electrodes, (ii) reduction can be almost complete in cases of highly asymmetric molecule-electrode couplings, but in this case (iii) biases corresponding to current plateaus are required, which common junctions cannot withstand. We next discuss that, much more than in two-terminal setups, the molecular charge, orbital energies, and vibrational properties can be efficiently controlled in cases of (redox) molecules embedded in scanning tunnelling microscope (STM) junctions in electrochemical (EC) environment. The electrochemical setup turns out to provide a unique chance to control the molecular properties. The key role in the unprecedented efficiency of this control is played by the electrolyte gating, which enables a continuous change in the molecular charge by up to an entire electron (complete redox process). To exemplify, we consider EC-STM experimental transport data for the redox unit (viologen), whose essential constituent is 4,4´-bipyridine (44BPY), a molecule of special interest for the charge transport at the nanoscale.