Molecular electronics in 2023

This past week I was fortunate to attend this meeting, the most recent in an every-few-years series that brings together a group of researchers interested in electronic transport in molecular systems.  This brings together physicists and chemists, and this was the first one I've attended since this one in 2015.

The evolution of the field over the years has been very interesting.  Generally gone are the discussions of using actual chemically synthesized molecules as electronic devices in eventual ultrascaled computing applications.  Rather, there is a broad recognition that these systems are important testbeds for our understanding of physics that can have broad ramifications for understanding chemical processes (e.g. quantum interference in molecules leading to sharply energy dependent electronic transmission and therefore enhanced thermoelectric effects - more here), light emission (e.g. the role of local vibrations, Franck-Condon effects, and quantum interference in determining the lineshape of light from a single molecule), and the right ways to think about dissipation and the flow of energy at the extreme nanoscale in open, driven quantum systems.  In terms of the underlying physics, the processes at work in molecular devices are the same ones relevant in eventual single-nm CMOS electronic devices.

There were two particular lingering problems/mysteries discussed at the workshop that might be of particular broad interest.

• Current-induced spin selectivity (CISS) remains an intriguing and confusing set of phenomena.  The broad observation, advanced initially by the group of Prof. Ron Naaman, is that in several different experimental implementations, is that chiral molecules seem to couple nontrivially to electron spin - e.g., photoemission through chiral molecules can generate spin-polarized electrons, with the handedness of the chiral molecule and the direction of electron motion picking out a preferred spin orientation.  This has led to a diverse array of experiments (reviewed here) and proposed theoretical explanations (reviewed here).  CISS has been used, e.g., to get LEDs to emit circularly polarized light by spin-polarizing injected carriers.  The situation is very complicated, though, and while some kind of spin-orbit coupling must be at work, getting good agreement from theory calculations has proven challenging.  Recent measurements in chiral solids (not molecules) look comparatively clean to me (see here and here), bringing device design and spin Hall-based detection into play.
• Charge transport over through thick films of biomolecules remains surprising and mysterious.  In single-molecule experiments, when there are no molecular levels resonant with the electrons of the source and drain electrodes, conduction of electrons is through off-resonant tunneling.  As tunneling is exponentially suppressed with distance, this implies that the conductance \(G \sim \exp(-\beta L)\), where \(L\) is the length of the molecule, and \(\beta\) is a parameter that describes how quickly conduction falls off, and is typically on the order of 0.5 inverse Angstroms.   For longer molecules or thick films of molecules, conduction typically takes place through some flavor of thermally-activated hopping and is steeply suppressed as temperature is lowered.  In surprising contrast to this, thick (30-50 nm) films of some biomolecules show almost temperature-independent conduction from room temperature down to cryogenic temperatures.  This is really surprising!  

It's heartening to see how much is now understood about electronic transport and related phenomena down to molecular scales, and how there is still more left to learn.