The behavior of charge transport in molecular scale electronics is governed by the innate chemical structure of the molecular system as well as it’s coupling to the macroscopic electrical contacts. An in depth understanding of this relationship could allow for the design of nano-electronic devices one atom at a time. Thus far, our understanding of the relationship between molecular structure and transport phenomena has been deduced through systematic studies on families of molecules with subtle variations in their chemical structure which regularly results in conflicting explanations for their underlying behavior. Thus, there is a clear and evident need for a more refined means of directly interrogating the structure of a current carry single molecule. In this work, we address this need by developing an experimental platform capable of simultaneously measuring the electrical and Raman response of a single molecule junction. This approach allows for mutually verifiable evidence of single molecule sensitivity in our measurements providing a highly reliable platform for investigating the influence of chemical structure on transport properties. By using a scanning tunneling microscope (STM) based break junction system, combined with a commercially purchased confocal Raman microscope, we are able to create single-molecule junctions between the apex of the STM tip and a conductive substrate supporting a monolayer of a target molecule and measure its conductance. Consequently, this nano-junction has a lightning rod effect when illuminated by a Raman excitation source, creating an intense and highly localized field within the junction allowing for detectable Raman scattering originating from a single molecule, thus allowing to probe the structural dynamics of a current-carrying single molecule.
We then demonstrate the utility of this platform in both the field of molecule electronics as well as chemical analysis. We show that through correlations in the conductance signal and Raman scattering from the molecular junction, it is possible to directly observe structural changes in the backbone of a single molecule junction and how this affects its conductance as well as see signatures providing insight on different binding motifs. We then apply this technique to a subset of amino acids and show that, with the aid of machine learning models, it is possible to reliably identify single amino acids using both the conductance and Raman spectral information but not either independently, demonstrating the need and utility for this platform for future advances in single molecule identification.