## Wednesday, January 27, 2021

### Zero bias peaks - an example of the challenge of experimental (condensed matter) physics

The puzzle-solving aspect of experimental physics is one reason why it can fun, but also why it can be very challenging.  In condensed matter, for example, we have limited experimental tools and can only measure certain quantities (e.g., voltages, currents, frequencies)  in the lab, and we can only tune certain experimental conditions (e.g., temperature, applied magnetic field, voltages on electrodes).  Getting from there to an unambiguous determination of underlying physics can be very difficult.

For example, when measuring electronic conduction in nanostructures, often we care about the differential conductance, $dI/dV$, as a function of the bias voltage $V$ applied across the system between a source and a drain electrode.  In an ideal resistor, $dI/dV$ is just a constant as a function of the bias.  "Zero bias" $(V = 0)$ is a special situation, when the electronic chemical potential (the Fermi level, at $T = 0$) of the source and drain electrodes are the aligned.  In a surprisingly large number of systems, there is some feature in $dI/dV$ that occurs at $V= 0$.  The zero-bias conductance $(dI/dV)(V=0)$ can be suppressed, or it can be enhanced, relative to the high bias limit.  These features are often called "zero bias anomalies", and there are many physical mechanisms that can produce them.

For example, In conduction through a quantum dot containing an odd number of electrons, at sufficiently low temperatures there can be a zero-bias peak in the conductance due to the Kondo Effect, where magnetic processes lead to forward-scattering of electrons through the dot when the Fermi levels are aligned.  This Kondo resonance peak in $dI/dV$ has a maximum possible height of $2e^2/h$, and it splits into two peaks in a particular way as a magnetic field is applied.  In superconducting systems, Andreev processes can lead to zero bias peaks that have very different underlying physics, and different systematic dependences on magnetic field and voltage.

Zero bias anomalies have taken on a new significance in recent years because they are one signature that is predicted for solid-state implementations of Majorana fermions involving superconductors connected to semiconductor nanowires.   These exotic quasiparticles have topological properties that make them appealing as a possible platform for quantum computingObservations of zero bias anomalies in these structures have attracted enormous attention for this reason.

The tricky bit is, it has become increasingly clear that it is extremely difficult to distinguish conclusively between "Majorana zero modes" and cousins of the Andreev features that I mentioned above.  As I mentioned in my last post, there is a whole session at the upcoming APS meeting about this, recent papers, and now a retraction of a major claim in light of new interpretation.  It's a fascinating challenge that shows just how tricky these experiments and their analysis can be!  This stuff is just hard.

(Posting will likely continue to be slow - this is the maximally busy time of the year as department chair....)