Alexandre Pourret: Nernst signal generated by superconducting fluctuations in low-Tc disordered superconductors

Dr. Pourret presented Nernst Effect measurements as a probe of novel
phenomena in superconductors. He started with a review of a definition of
the Nernst coefficient, which measures the size of a transverse electric
field generated by a longitudinal temperature gradient in a magnetic field.
He pointed out that simple transport theory indicates that the Nernst effect
is proportional to the derivative of the Hall angle with respect to changes
in the Fermi energy. This dependence leads to the so-called Sondheimer
cancellation of the Nernst effect in simple 1 band metals. He went on to
say that non-superconducting materials with a large ratio of mobility to
Fermi energy display the largest Nernst effect.

Next, Dr. Pourret turned to Nernst effect measurements in superconductors.
He characterized the work by Ong's group as a breakthrough in the use of the
Nernst effect to probe novel phenomena in superconductors. That group
observed large Nernst coefficients above the transition temperature in high
Tc materials in the pseudogap regime. The observations have been taken to
indicate the presence of superconducting fluctuations well above Tc. With
those experiments in mind, he and coworkers embarked on Nernst effect
measurements on disordered superconducting films. The high sheet resistance
NbxSi1-x and Indium Oxide films they employed were expected to exhibit large
fluctuation effects.

The experimental results on the NbSi and Indium Oxide films were:

In NbSi, the Nernst coefficient above Tc is much smaller than that below
Tc. The maximum in the Nernst signal exhibited a cusp in a magnetic field
vs. temperature plane, going to zero at Tc. He compared the Nernst
coefficient behavior to a theory labelled USH that gave agreement over a
range of magnetic field. An improved theory fit their data over almost the
entire range.

In Indium Oxide, the Nernst signal decreased with decreasing temperature and
passed smoothly across Tc on cooling. It exhibited no cusp structure.
Thus, there seemed to be no difference between the amplitude fluctuation and
vortex regimes as seen by the Nernst effect. Armitage queried whether the
experiments might show a difference at lower temperatures than those
investigated.

Blogged by Jim Valles