Friday, August 26, 2011

Maoz Ovadia: The insulating state in amorphous strongly disordered superconductors

The speaker, Maoz Ovadia, presented the basic facts about InOx superconductors obtained from transport measurements by the group of Prof. D. Shahar at the Weizmann Institute (Israel).

After an obligatory introduction consisting of: (1) the list of theoretical papers by Matthew Fisher et al about the superconductor-insulator transition treated as a quantum phase transition, and (2) comparison of the two standard sets of data obtained in amorphous and granular superconductors, the speaker demonstrated two sets of r(B)-curves (dependence of the resistance on the magnetic field ) in InOx. One set was for a relatively clean superconducting film, and the other set for a disordered one. Resistance at different temperatures exhibits maximum and, consequently, the curves r(B) intersect with each other at some region of the magnetic fields. In the clean film it occurs around 11Tesla. This set apparently cannot be considered as an illustration of a quantum phase transition. An interesting feature of this plot (which was not discussed, unfortunately) is a noticeable resistance in the fields about 1-2 Tesla or less without a clear transition into a superconducting state. This plot was contrasted with the data in a more disordered conductor. There, the curves intersect accurately enough at the magnetic field B* about 2.7 Tesla, i.e., at noticeably smaller magnetic field comparing to the clean case. With the lowering of temperature, curves at fields smaller than B* go down (superconductor) while at the fields exceeding the critical go up (insulator). The speaker demonstrated a curve at T=0.02K which was almost rectangular with the resistance shooting abruptly up to MegaOhms for B> B*, i.e., for magnetic fields on the insulating side of the transition. Thus, one indeed may speak about a quantum phase transition (we put aside the question about how universal is the value of the resistance at the point of the intersection in different films; it is not). The phenomenon is not unique for InOx films, but was also observed by other groups in different materials. However, it was not so pronounced as in InOx.

Interestingly enough, the insulating behavior is limited to a finite interval of magnetic fields. At large enough magnetic fields, the reentrance has been observed; resistance starts to fall down when the magnetic field increases. The whole phenomenon, shooting up to MegaOhms and retracing back to the resistance about 10^4Ohm, develops within the interval of magnetic fields from 2 to 10 Tesla. To get a clue about the origin of the phenomenon, the speaker demonstrated resistance as a function of temperature at a fixed magnetic field (about 5 Tesla). They found an activation behavior with a gap comparable with the superconducting gap in this material.

As another hint, the speaker used the experiment by James Valles’ group, where they observed a sort of Little-Parks oscillations with a period corresponding to a superconducting flux quantum per unit cell in an insulating Nano-Honey-Comb film. The speaker interpreted his data as a random array of tunneling junctions, and the effect of the magnetic field was interpreted as a flux-related effect. A few slides from the paper of Dubi, Meir and Avishai illustrated this point of view. (One may guess about how to reconcile the picture of random tunneling junctions with the results of the STM measurements in this material. In other words, why people have to develop a theory of the preformed pairs, if everything can be explained in such, rather conventional, way?)
Then the speaker switched gear, entirely: he discussed the nonlinear transport in InOx at low temperatures, about 0.1mK. There is a hysteresis phenomenon as a function of bias voltage. First, the data at a high magnetic field of 11Tesla were presented. There is a striking similarity with the data in YSi by M. Sanquer et al. A comparison with the theory of the overheating by Altshuler et al published in PRL 2009 has been discussed. As a justification of this picture (it assumes that resistance depends solely on the electron temperature despite that the transport may be either phonon- or electron-assisted), a paper by Marnieros et al PRL 2000 has been quoted. If not to pay attention on the significant disagreement in a numerical prefactor, the comparison of the data at B=11Tesla with the theory is rather convincing. For B=5Tesla the situation is less satisfactory and it requests a modification of the rate of the electron-phonon scattering (at least).

Since the talk consisted of two disconnected parts, it was difficult to make a certain conclusion. The speaker just thanked the audience for the attention.

Blogger Alexander Finkel’stein

No comments:

Post a Comment