Superconductivity Dependent Friction

Abstract

In the work reported on here, Quartz Crystal Mircobalances are used to study the change in friction between a sliding layer and a substrate at a superconducting transition. Studies of this nature are one of the few able to isolate electronic and phononic contributions to friction. Superconductivity dependent friction was first observed in 1998 for N2 sliding on a Pb surface. This discovery generated much interest both experimentally and a theoretically. However success in reproducing and modeling the results was difficult due to the unique care required for this experiment.

The most aggressive efforts to reproduce the affect where made by Renner, Taborek, and Rutledge. In their experiment Renner et al. tried to reproduce the experiment with improved the temperature stability. Renner et al. designed a system, in which the sample would be cooled over many hours. The improved thermal stability came at the expense of surface quality. The slow cooling of samples by Renner et al. most likely led to increased surface contamination which pinned the sliding layers. Pinning prevented Renner et al. from reproducing the experiment and reports of their unsuccessful attempts led many to question the existence of superconductivity dependent friction.

Marginal success in reproducing the experiment was made in 2001 , while still using the same sample transfer technique of blowing Argon as utilized in the original experiment. In an effort to mitigate some of the expense associated with the large amount of liquid helium consumed during this experiment a temperature control loop was implemented and a number of other traditional low temperature physics techniques. Once again improved thermal stability came at the cost of surface quality, which led to pinning problems in that work as well.

Many theoretical efforts were made to explain superconductivity dependent friction within the established understanding of electronic friction. Modeling the experiment was made difficult by ambiguity in two values that would prove crucial to many of the models. The electron mean free path l of the samples used was not known and could only be estimated from bulk values for Pb. Additionally, since the samples were exposed to atmosphere during transfer an unknown amount of oxide formed on the surface. This made the distance between the adsorbate and the superconducting substrate unknown, which introduced large variable to the numerical estimates made by a number of models.

A few models were able to make verifiable experimental predication based on there work. Bruch predicted in 2000 that the magnitude of the change in frictional force experience by a sliding layer at a superconducting transition would depend on the polarity of the adsorbates in the sliding layer. It is Bruch's predication about the nature of superconductivity dependent friction that has motivated much of the work I present here.

I have studied the change in friction felt at a superconducting transition by layers composed of absorbates with varying polarities (N2, He, and H2O) sliding on a Pb surface transferred to an experimental cell in-situ from the deposition chamber. The work in this dissertation shows that superconductivity dependent friction is present in all systems were the effects of pinning can be mitigated. I compare these results to the predictions made by Bruch and show that our results agree with his theory of superconductivity dependent friction. Additional comments are made on the unique superconducting behavior of the samples in this study, and on apparent coupling between sliding layers and magnetic fields. Finally comments are made as to the state of theoretical treatments to this problem and possible future experiments are suggested.