 |
|
> Nanotech forum > Discuss referee reports and reviewer comments > Research proposal on nanowires: Discussing critical reviewer`s comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES
|
|
|
Research proposal on nanowires: Discussing critical reviewer`s comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
Alexey
Join Date:
Mon Apr 2 04:58:18 2007
Posts:78
Reputation:11
 |
Research proposal on nanowires: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
| 3. The proposal on investigation of magnetic impurities on superconductivity depends on the results of the PI's paper (http://www.nanogallery.info/nanogallery/?ipg=129), which I do not find convincing, in spite of its being published in PRL. This effect seems to be rediscovered every few years, most recently by Moses Chan's group in addition to the PI's group, who dub it the anti-proximity effect, but work by Van Haesendonck's group in the early 90's (in PRL) and other unpublished work even earlier has shown similar effects. For example, Van Haesendonck's group showed a Little-Parks effect on a superconducting ring where the Tc increased with magnetic field before it decreased. It can be understood simply as an experimental artifact, due to the fact that measuring a superconducting wire in the resistive state will lead to heating. If the superconducting wire is connected to superconducting leads with a higher transition temperature, the heat generated has nowhere to go, since the leads block heat transport, and so the effective temperature of the wire is elevated. Applying a magnetic field kills superconductivity in the leads first as they have larger area normal to the field. The thermal block is therefore removed, the nanowire cools initially and shows more superconducting properties, critical current, Tc, etc, and it appears that superconductivity is enhanced with increasing field. The problem is even more acute in nanowires, and the trend that the authors note in their PRL (ref 50), that narrower wires show a stronger effect. However, this experimental problem is not even mentioned in the paper. Consequently, I think that basing a proposed experiment on a false premise is not a good idea. |
| Created: Fri Apr 27 23:34:09 2007
|
|
Alexey
Join Date:
Mon Apr 2 04:58:18 2007
Posts:78
Reputation:11
|
Possible objections to this critical comment |
Possible objections to this critical comment:
1. Heating is not important since the measurement of the critical current is made immediately BEFORE the wire switches into a resistive state. Thus the measurement is done in the superconducting state, when there is now noticeable heating. Thus heating is not an issue in these experiments.
2. The main idea of this critical remark is that the heating should be reduced as magnetic field is applied. The explanation of the observed magnetically-enhanced superconductivity is that magnetic field makes creates quasiparticles in the leads and thus makes them able to carry heat away from the wire. Thus its superconducting characteristics become stronger. This phenomenon should depend on the field orientation with respect to the film but it should not depend on the orientation of the field with respect to the wire. Thus the control experiment, presented in Fig.2c of the above-mentioned PRL paper (which shows that the enhancement depends on the field orientation with respect to the wire) proves that the explanation advanced in the critical remark is not correct. The theory proposed in the paper, on the other hand, explains naturally the observed orientation dependence. |
| Created: Fri Apr 27 23:59:27 2007
|
|
Physya The Cat
Join Date:
Wed Apr 18 04:02:53 2007
Posts:19
Reputation:11
|
RE: Research proposal on nanowires: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
| I guess the major and main issue is to show that magnetic impurities do exist in the system. (After long discussions with Andrey, I am not convinced at all that this is the case). What evidence do you have for magnetic impurities? |
| Created: Sat Apr 28 07:08:39 2007
|
|
Alexey
Join Date:
Mon Apr 2 04:58:18 2007
Posts:78
Reputation:11
|
RE: Research proposal on nanowires: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
In fact the paper says explicitly that we do not think there are magnetic impurities at all. The explanation was that it is localized states which are occupied with single electrons provide free spins. These spins or local magnetic moments can influence superconductivity we think.
|
| Created: Mon Apr 30 01:42:43 2007
|
|
Kostia
Join Date:
Fri May 25 18:59:39 2007
Posts:7
Reputation:12
|
Comment on the reviewer's comment: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
"Van Haesendonck's group in the early 90's" probaly means PRL 69, 1268 (1992). This effect (resitive transition anomaly: bump at the top of R(T)) has been observed in wires and loops. The position and the magnitude of the bump can vary from sample to sample, from cooling to cooling and also by measuring current and magnetic field. Though I myself tried initially to propose a "sophisticated model" [PRB 53, 12304 (1996)], later I became convienced that this is a trivial physics related to nucleation of N-S interfaces and the corresponding re-arrangement of the current density across the wire [PRB 59, 6487 (1999)]. Negligence to experiment (e.g. finite DC component if lock-in is used to measure AC resistance) can also contribute to "interesting physics" :)
What is related to the Referee's comment on overheating, it might happen indeed. But a normal experimentalist should work (and report results!) in the linear regime, when the Joule heating is negligible. A rule of thumb test is that variation of the measuring curent by a factor of 10 should not shift the position of the onset of superconductivity (both T-sweep up and down). In this limit the mentioned overheating is absent.
Now about local magnetic moments (or corresponding local states). Hypothetically it is OK. But one should prove their existence. I also see negative magnetoresitance (nMR) in Al nanowires. There one should forget about localized magnetic moments: hosting matrix is known to be immune up to very high concentrations. But the funny thing: why it is size dependent? In my case (and Alexey's also?)this nMR pops up only in sub-15 nm wires and is absent in > 20 nm. Of course, ratio of surface (=hypothetical surface states?) to volume increases, but why so that sharp? |
| Created: Fri May 25 19:27:07 2007
|
|
Alexey
Join Date:
Mon Apr 2 04:58:18 2007
Posts:78
Reputation:11
|
RE: Research proposal on nanowires: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
Hi Kostia,
Thanks for very interesting comments. I will try to discuss some of them within time.
-Heating effects
What you say is correct of course and we always do this check by reducing the bias current and making sure that the R(T) curve is not changed (except that it becomes more noise as the bias current is lowered). But the important thing is that in this paper we report the critical current versus magnetic field measurements. Thus the resistance is zero and there is no heating. In more detail we measure as follows: we sweep the bias current and measure resistance. The resistance remains zero (no heating). At some current we observe a sharp jump of the voltage and resistance. This point is recorded as critical current. Thus all the way up to the current when the jump happens there is no heating. Some heating might occur only after the jump is detected. Thus the measurement is not affected by heating at all.
-Van Haesendonck's results and your discussion of them is very interesting (although they are not related to the magnetic enhancement of the critical current which is discussed here). We might discuss them at our journal club. At this time I wonder weather you can help us to understand our other result observed recently: we see that a phase slip center can be induced in a wire when the wire is exposed to microwave radiation. One puzzle we have is that the dV/dI value for this PSC is much less (perhaps 5 or 10 times less) than the normal resistance of the wire. How this could be? The wire is only about 100 nm long. I would think that quasiparticle imbalance should persist on a longer scale. Since you paper is related to these issues, perhaps you can suggest how to understand this small resistance of this microwave-induced PSC.
|
| Created: Wed May 30 22:38:22 2007
|
|
Kostia
Join Date:
Fri May 25 18:59:39 2007
Posts:7
Reputation:12
|
RE: Research proposal on nanowires: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
I do not see anything suspicious about your low value of differential resistivity of PS. Charactristic relaxation length LQ=sqrt(DtauQ), where (1/tauQ)~(Delta/T)(1/tauE). In your case diffusivity D=Vfermi el is rather small (very small mean free path el). Second, in type-II superconductors tauE can be 2 orders of magnitude smaller than in pure Al or Zn. Hence, you may get LQ < 10 nm. This is at T-->Tc. At low T LQ obviously furhter drops.
About rf radiation and PS look at Tidecks book. Such effect has been observed: something similar to Shapiro steps. |
| Created: Wed May 30 23:14:42 2007
|
|
Alexey
Join Date:
Mon Apr 2 04:58:18 2007
Posts:78
Reputation:11
|
RE: Research proposal on nanowires: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
Thanks!!!
I though that LQ should become larger at low temperatures. Did you really meant to say that LQ should drop at low T? One can think that inelastic processes are less probably at low T.
What is tauE, can you give some hint on how to estimate it for MoGe? |
| Created: Sat Jun 2 21:22:49 2007
|
|
Alexey
Join Date:
Mon Apr 2 04:58:18 2007
Posts:78
Reputation:11
|
RE: Research proposal on nanowires: Discussing critical reviewer's comments-LOCAL MAGNETIC MOMENTS IN NANOWIRES |
Recently I found a news item about two nonmagnetic materials forming localized electronic states at the interface. These localized states have magnetic moments:
http://www.nanogallery.info/nanonews/?id=7984&slid=news&type=nano
So one can speculate that MoGe forms similar states near the surface, which is oxidized, so there is an interface between the metallic alloy and its oxide.
|
| Created: Sat Jun 9 21:34:35 2007
|
|
|
|
|
 |
