Thursday 30 July 2015 - Eduard Driessen - STM experiments on electrically connected superconducting TiN nanowires locally driven out of equilibrium

Eduard Driessen presents a new tool to study the non-equilibrium properties of disordered superconductors he entitles „critical current microscopy“: A STM current is used to locally perturb a nanowire where the non-equilibrium state is probed by the critical current.

Eduard motivates his talk with possible applications of disordered superconductors for e.g. single photon detectors. At the same time he points at the so-far only poorly understood microwave response in strongly disordered TiN: The DOS one needs to feed into Mattis-Bardeen equations to fit the dynamical conductivity does not compare to the tunneling DOS. 

Eduard continues with the main objective of his talk: We know (maybe, a bit) about the equilibrium properties of disordered superconductors but basically nothing about the non-equilibrium properties. With the technological state-of-the-art possibilities of structuring at the nanoscale this has become feasible. The systems Eduard tailors and studies are superconducting TiN nanowires (4µm x 200nm, 1KOhm -> close to the SIT) at 50mK through which a current is sent. Using a blunt STM tip positioned above the wire, a small current (0.2 - 2nA) is locally injected. To prevent the tip crashing into the insulating substrate, the nanowire is neighbored by other nanowires separated by a distance small enough to keep the tip above the substrate.  The current through the wire is ramped up until its critical value is reached.

Prior to the non-equilibrium properties, Eduard addresses the properties with the perturbation switched off. On average, only very small coherence peaks are present and a large zero-bias anomaly (>0.5) is observed.  Eduard states that it is impossible to fit the particular shape to any formula known (either a fit captures the suppression of coherence peaks and fails at the zero-bias or vice versa). Also on the local scale, a fully gapped DOS is not resolved. This is in some sense different compared to what was observed previously (studies of B. Sacepe) and in disordered NbN, where the coherence peaks are also suppressed, but a fully gapped DOS is observed at various sites, as Pratap Raychaudhuri points out. The (equilibrium) critical current is 1.2µA.

Upon switching on the STM current injection (1nA), Eduard reports a tremendous drop of the (non-equilibrium) critical current from 1.2µA to 250nA. (“like pouring a glass of water into a river which thereupon stops flowing”). At different spots along the wire the critical current varies by 25% (qualitatively) similar to the observed variation in the (equilibrium) zero-bias map. The correlation between the variation of critical current and local DOS in the sense “low critical current <-> weak spots”, however, is delicate, as also geometric influences need to be considered. In general, weaker SC is more likely to be found close to the edges of the wire, similar as it has been inferred from optical-absorption studies.

It turns out that Ic depends on the quasiparticle energy. Starting at several meV, the critical current increases as the energy goes down, experiences a maximum before it is suppressed again. This is somewhat counterintuitive as in the limit of low quasiparticle energy the impact on the critical current becomes more severe. Eduard drops the idea of the interplay between thermalization and recombination as a possible explanation. Using a higher STM current, the absolute values of Ic are reduced, however the energy dependence persists. He notes that the maximum of the Ic curve happens to roughly coincide with the energy gap of the tunneling spectra. Something to think about.

In summary, Eduard presents a powerful new method to look at the non-equilibrium properties of disordered superconductors - apparently a field full of open questions – and a fascinating first view on its nature.  

Blogged by Uwe Pracht