Superconducting Thin Films

Current superconducting cavities are generally made of solid Niobium. Due to several decades of improving manufacturing and surface treatment processes, Niobium cavities today reliably reach accelerating fields above 30 MV/m, at high quality factors of above 1010.  The ongoing improvement of Niobium cavities cannot go on indefinitely though, as the theoretical limits of the material are close to being reached.

A different possibility to reduce cost as well as increase the quality factor and/or accelerating fields is to use thin film coated cavities. The principle of adding a very thin (~100nm) layer of superconducting material onto a copper cavity works, as electromagnetic fields only penetrate the superconductor on an order of its penetration depth, for Niobium this is around 40 nm. Thin film coated cavities have several advantages over cavities made from bulk Niobium:

  • Better mechanical properties of Copper allow for easier manufacturing
  • The higher thermal conductivity of copper compared to Niobium protects the cavity from thermal quenches
  • Cost reduction due to less Niobium needed
  • A lower susceptibility to magnetic flux trapping
  • The potential of raising the operating temperature from 2K to 4,2K

Apart from Niobium thin films, other substances such as Niobium-tin, Magnesium diboride, and Niobium nitride are promising candidates. These have critical temperatures significantly higher than Niobium and are thus believed to have a lower surface resistance. Their comparatively low HC1
(I.e. the external magnetic field at which normal conducting vortices penetrate the material and cause high dissipative losses) can be enhanced by applying a thin film in the order of the penetration depth.

Superconducting thin films are thus very interesting for SRF cavities. A problem is that depositing an even, homogenous layer on a curved and closed geometry is no easy task. Even tiny defects or normal conducting spots can completely destroy a cavities performance. This is easy to imagine, as a local normal conducting defect will have a surface resistance six orders of magnitude higher than the surrounding material and create great amounts of heat due to ohmic losses, quickly quenching the entire cavity. In the case of the compound superconductors, general thin film deposition techniques have not yet been established.
To determine whether a certain technique (say magnetron sputtering or energetic condensation) produces a good result with certain parameters (temperature, pressure,...) one has to perform a RF test after each iteration. In these cases manufacturing an entire cavity for each RF-characterization would be a tremendous effort. Smaller, plane samples would have a faster turn-over rate and thus speed up the research process.
 
An ideal setup for the RF characterization of superconducting samples would precisely measure the surface resistance of samples at temperature, frequency and external magnetic fields as used in SRF environments.

Quadrupole Resonator


A method of measuring the surface resistance of samples uses a Quadrupole Resonator, pioneered at CERN in the late 1990’s. It consists of a pillbox-like niobium cavity which acts as a screening cylinder. Four rods are supported from the top-plate of the cavity and are short circuited pairwise by two loops just above the bottom plate. As can be seen in schematic drawing below the screening cylinder has an opening below the two loops from which the calorimetry chamber is mounted, thermally isolated from the screening cavity. The RF power dissipation in the sample is given by:




and can be measured calorimetrically, which is far more accurate than that by RF power measurements. With knowledge of the magnetic fields present on the sample surface, one can then calculate the surface resistance.

The amount of power dissipated on the sample is measured by a setup involving a compensation heater attached to the bottom of the sample, powered such that the sample temperature remains constant when RF is turned on. The difference of the power levels of the heater is exactly the RF power dissipation on the sample.

 The coaxial gap between the calorimetry chamber and the screening cylinder is small enough that the first quadrupole (and dipole) modes are below the cutoff frequency. This means that the electromagnetic fields decay exponentially in this gap and are sufficiently small at the lower flange that they do not interfere with the dissipation measurement. The quadrupole resonator gets its name from the fact that it is a screened four wire transmission line, excited in a quadrupole mode.

Due to the increased interest in superconducting thin films and methods of RF-characterizing superconducting samples, a new measurement facility based around a quadrupole resonator is being constructed at HZB.   With the new apparatus, currently in manufacturing stage, we will be able to measure surface resistance of samples with a nano-ohm resolution, at field levels not attainable by comparable setups. To reach this goal the design of the Quadrupole Resonator was optimized to reduce peak surface fields, a summary of the optimization was presented at the SRF  2013 conference. First measurement results are expected in spring 2014.