What do we know of superconductivity in thin layers?

Question

motivated by another question, i wonder if there are special properties of superconductivity when restricted on 2D or very thin layers related to the effective permittivity in function of the frequency $\epsilon ( \omega )$ near the first-order transitions between the superconducting and normal phases , any references about the main state of the art would be appreciated

Edit: I've noticed that most research papers on superconductivity report on the current transport properties, but they usually don't talk much about the permittivity. Is this because its harder to measure?

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user lurscher

Answer 1

The obvious case for superconductivity superflow in 2D is, well, superconductivity superflow in a 2DEG (2D electron gas). One of the many ways this manifests is in the quantum hall effect. There, one observes a step-like behavior for the resistivity of a 2D sample w.r.t externally controllable parameters such as the magnetic field or carrier concentration. The resistivity is quantized in integer or fractional units as shown in the plot below (courtesy of D.R. Leadley, Warwick University 1997).

When the magnetic field is such that the system lies on the "plateaus", in between the quantized resistance values, the 2DEG in the bulk of the sample is in a superconductivity state exhibiting dissipationless transport. This is also depicted by the green line in the plot which measures the longitudinal resistance $\rho_{xx}$ of the sample. We see that this quantity vanishes precisely when the system is on a plateau indicating the presence of superflow in the longitudinal direction.

Edit: As pointed out by @wsc and @4tnmele in the comments below, it is not quite accurate to describe the plateaus as being in a superconducting state. However on the plateaus the hall strip does exhibit dissipationless flow in the longitudinal direction - even though this is primarily due to edge currents. The bulk remains insulating. So while it might be correct to describe this state as exhibiting "superflow" it is incorrect to call it a "superconductor". I have modified the language in my answer to reflect this change. I am not deleting my answer because I feel that it is still relevant in the context of the OP's question.

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user user346

Comments

+1, i've noticed that most papers about superconductivity don't care too much about measuring the surface permittivity and just focus in the current transport characteristics

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user lurscher

---

-1. This is incredibly wrong -- there is no superflow in QHE! Superconductivity is much more than vanishing $\rho_{xx}$ anyway, and it is also trivially seen that the dc conductivity $\sigma_{xx} = 0$ in quantum Hall systems as well (this is because resistivity is the inverse of the conductivity tensor).

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user wsc

---

Dear @wsc, quoting from Jain's "Composite Fermions", pg. 19 "What is remarkable about the quantization of the Hall resistance is that ... concurrent with the quantized plateaus is a “superflow,” i.e., a dissipationless current flow in the limit of zero temperature." I call it superflow, Dr. Jain calls it superflow. You can call it whatever you like.

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user user346

---

@wsc you might be right in that the state of the electron fluid on the plateaus is not exactly like that of a superconductor in that it does not exhibit cooper pairing. However, the fact is, the best description for the dissipationless flow observed at the plateaus is as "superflow". Likewise the mechanism for superconductivity in high T_c is different from low T_c, but we still call it superconductivity.

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user user346

---

@Deepak Vaid Isn't superconductivity defined as perfect conductor + perfect diamagnet? Not sure FQHE fulfils this...? Although there are a few concepts from superconductivity that arises in FQHE, I am pretty sure that experts don't consider FQH systems as 2d superconductors.

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user Heidar

---

@4tnemele given that you're an expert (IMHO) I'll take your word for it.

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user user346

---

@Deepak thank you for your kind words, but I definitely don't deserve them. I am by no means an expert in this. :)

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user Heidar

---

Regardless of this little labeling & classification debate, this is still relevant to the main concern; as long as the superflow state behaves as a mirror in a range of frequencies and can be sandwiched in layers in the micron range

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user lurscher

---

@lurscher I'm not sure what you mean by "behaves as a mirror", but people have done work on bilayer (and more perhaps) quantum hall systems. So yes, QHE surfaces can absolutely be sandwiched in layers. In fact, I remember attending a colloquim which was about sandwiching two such substrates so that quasiparticles in one pair with quasiparticles in the other one and you get a condensate in this manner! I can't remember the speaker's name but when I think of cool cond-mat experiments Keith Schwab's name comes to mind.

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user user346

---

"behaves as a mirror" in the context of having complex permittivity in some EM range - as metals or mirrors do

This post imported from StackExchange Physics at 2014-04-01 16:44 (UCT), posted by SE-user lurscher