References about Cloaking
Cloaking is a new concept for making the objects invisible. It is not a dream now. recent researches show that based on advances on metamaterials in enginearing the strcuture of the matters it can be possible. You be able to be involved in this new research topic.
Good Luck,
Added after 6 minutes:
Hear are some interesting papers regarding the cloaking.
U can see also the following link:
https://www.edaboard.com/viewtopic.php?p=892974#892974
Hi!
I am have been doing some research on cloaking and published some papers on that (very recently). I am new to this forum and I just saw that there are lots of interesting post around. Would anyone be interested in discussing cloaks?
Anyway, the topic is interesting because of various electromagnetic phenomena, but as for the real thing - whether an 'invisibility' cloak can be made, I think that I can easily convince anyone serious that it's not really possible - what really can be achieved is lowering the scattering cross-section and reducing radiation (for the case of a source inside the cloak) and this, also, cannot be achieved in some spectacular extent (an order of magnitude would be a reasonable limit).
Regards
Hi qoran1901,
I'm very interested to see your publications, is that possible? You are right, in the real world constructing an ideal invisibility cloak is not possible just because of the loss of the matterials, however, in the theory for lossless media it is perfectly possible.
Hi!
We could do that, but we would have to establish a contact on a more personal basis e.g. over e-mail.
Have you simulated any cloaks? The simulations published so far are done in COMSOL and just for small scatterers (R~lambda). It turns out that there is always some scattering but it is very small if you simulate this case in COMSOL, and when doing so you don't really pay attention what are the values of material parameters used. I am not speaking about losses - that is another issue. Even if losses are neglected, it can not really work - just see what the values of permittivity and permeability have to be (especially on the inner surface of the cloak) and ask yourself is that possible. There are several very recent papers, starting with Phys. Rev. Lett. 99, 113903 (2007), Z. Ruan et al., that investigate what happens if the material parameters are slightly perturbed at this inner surface but they don't actually discuss how stringent the conditions on these parameters are for a given perturbation parameters (it's delta in that paper).
If you don't have acces to Phys. Rev. Lett. you can find a preprint of that paper on arxiv.org, it's arXiv:0704.1183v2 and it's free.
Anyway, my point is that if you are systematic in discussing the thing, it turns out that a real invisibility cloak is not possible (in terms of the above perturbation, the physics doesn't work in the limit when delta goes to zero). The scattering width is reduced - that is right, but when you ask yourself how much it can be reduced you soon realise that it cannot be much over than reducing it by a factor like 10 or something.
Finally, I don't know how much you are into the metamaterial technology. It turns out that it is still really hard to make metamaterials for visible frequencies and by this I mean especially designing the magnetic response, i.e. the permeability. You can read about this in a paper from Costas Soukoulis Science 315, p47 (2007) (January 5, 2007).
And you? Have you done any research on cloaks?
Regards
Hi,
I'm very concern about the Phys. Rev Lett. paper you introduced, it is chellenged one of my work. If you let me we have discution several days latter to have enough time to discuss. Regarding the NRI, yes I'm familiar with such metamaterial structures and I have reviewed most of the papers on this area. You may know that recently the CIT univercity could achive negative refraction on really visible range using the plasmons, I have not yet read that paper. How can I have your Email? You may can find a possibility in this website to send me a private message.
Regards,
Hi!
It seems that people are not really interested in cloaking. Why?
Anyway, if cloaking is too boring for you, check out the light wormholes :)
Here is an article from Nature (November 15, 2007).
pass: greenleaf
Regards
P.S. A. Greenleaf is the guy who 'invented' the light wormholes - check out his recent Phys. Rev. Lett. article
Hi,
Wormhole is a very interesting topic the same as cloaking. But actually, the wormhole is not a new concept; it uses the mapping like cloaks. It is mentioned in the paper that cloaks are making use of blowing up a point while the wormholes are based on blowing up a curve. Unfortunately the Phys. Rev. Lett. is full of new symbols which the electrical engineers are not familiar with them, anybody could understand the paper completely?
Regards,
Added after 13 minutes:
Dear Everybody,
An interesting link: in gigapedia.org you can find lots of free ebooks, check it out you may find something about invisibility, metamaterials and so on. If you find something useful let us know.
Regards,
Well, I think there is a bit more to light wormholes than 'blowing up a curve' :) True, the idea is the same, but then, again, the question is what do we mean by 'the idea'. Transformation media was 'invented' decades ago for use in FEM. Actually, it is even older - it follows from the general covariance of Maxwell equations. Greenleaf is just more systematic on the thing.
I agree that papers by professor A. Greenleaf are hardly readable by engineers:)
Anyway, there are no books on invisibility on gigapedia. There are, I think, two books on metamaterials, however. A book by Caloz and a book by Eleftheriades.
Invisibility is too new to be 'covered' in books. However, a lot of interesting references on invisibility can be found in the recently published papers. There is even an article by Milton Kerker on invisible bodies which was written decades ago...
Regards
Hi,
I have some problems, first of all regarding the symbols on the paper, would you please help?
Regards,
Is that a joke? :)
You can find papers of prof. A. Greenleaf on arxiv.org. There is this huge (~60pages) paper on cloaks. There, he uses all kinds of math stuff. It is interesting and I guess you could start from there picking things up. However, I guess it would take some time since I don't suppose you're a mathematician:)
I don't know enough to fully understand the math of the Phys. Rev. Lett. so I can't really help. I could suggest you some books, but then, again, I don't think you really have time for that. I would start from some basic functional analysis, topology and metrics spaces (which we didn't really cover as engineers). Then, I would play around a bit with Minskowski spacetime and just the basics of general relativity (I mean only the math part) and then do the general covariance. It is not as horrible as it sounds. Being a mathematician, Greenleaf merely uses different words, but if you knew basic stuff about the funny things that happen to metrics in general relativity, I don't think there would be problems reading his papers. It would only require 'translating' to physicist language.
Anyway, I think that at this point we have bigger problems than not understanding jsut his symbols:)
Regards
Ok,
Trapped Rainbow is another application of the metamaterials to slow down or stop the light. There is a recent Nature paper (15 Nov. 2007) submitted by Tsakmakidis et al. from university of Surrey, I upload it. Hope you could find it useful. The second paper is the supplementary material of the first one.
Regards,
Any other new ideas of controlling light are welcomed. The main paper has been submitted twice, the supplementary paper:
Hi all,
which solver do you use in COMSOL to study the cloaking of an object?
I'm using the linear system solver (Direct UMFPACK), but the convergence is not very good.
Can you help me?
Thanks in advance.
Hi iaia,
I upload one of the useful examples of COMSOL software, it is about simulation of scattering from a dielectric. You can also find the PML settings for your design.
Regards,
Hi myebook,
I know the example you have posted. My question was a bit more specific.
Which solver gives the better performance in a problem of scattering?
A direct or an iterative solver?
In particular, I'm working on cloaking and with the default direct solver I can't get the expected results, while the iterative solver does not converge.
Can you give me some suggestion?
Thanks
Hi iaia,
I have done a zillion of cloak simulations in COMSOL with various parameters for the cloak (those from the references and some 'self-made') and it always worked. The solver I invariably use is the default one - the Direct (UMFPACK) Solver. I don't see why it doesn't work in your case. If you have some example send me and I will take a look.
I already had some (private) discussion with myebook on his simulations. The most probable reasons for not getting the right results are
(a) either your mesh is too coarse (lambda/10 is approx. the required max mesh size); if you further refine the mesh at the cloak's inner surface it will produce better results...
(b) perhaps you didn't enter good cloak parameters - the usual 'mistake' is that people put mu_r and mu_phi instead of mu_xx, mu_xy, mu_yx and mu_yy; the transformation from the former ones to the latter ones is simple and you can look it up in the Phys. Rev. E paper from Cummer.
I don't think the solver is the issue here.
However, I also have a question: do you (or anyone else) have experience running Comsol 3.2 on E6400 or E6300 Intel Core 2 Duo machines? I have been running Comsol on two such machines (in one case the computer had 1GB RAM and in the other 2GB RAM) and in both cases I get the message that it has run out of memory if I go over approx 100 000 (its 10^5) degrees of freedom. This bothers me because these machines perform only very slightly better than my laptop which has 0.5GB RAM, and much worse than my friend's laptop which has 2GB RAM on which he can do problems with more than 1000 000 (10^6) degrees of freedom. Also, my friend's desktop computer has 1GB and an older Intel processor and is, also, able to do several hundreds if not a million degrees of freedom. That's the reason I think this issue has to do something with Core 2 Duo.
Another thing: I tried running Comsol in Windows XP and Linux (Ubuntu) and the same thing happens.
Further, when I run Comsol on these E6400-E6300 machines, the system monitor shows that the RAM memory is really not exhausted (on this machine with 2GB, the memory usage never goes above approx 700MB), not the mention the virtual memory.
Finally, I have tried to use an iterative solver for >10^5 cloak structures on this machine (E6400, 2GB RAM) and it took Comsol several (approx 3-4) hours to say that the calculation doesn't converge. So, this is the same problem you mentioned with the iterative solver.
Regards
Added after 14 minutes:
I forgot to add:
ieie - you have asked which of the solvers works better for scattering problems.
I am not familiar with solvers' details - I have this simplified picture: you get some system of equations in the FEM from the variational theorem and you 'just' have to solve it. I think there is nothing specific in scattering problems per se. Which solver is better should depend on the kind of structure you are using. The other very important thing is how your mesh looks like so we can say that the structure is actually determined only when you choose a particular mesh.
The 'variational theorem' means that a solution with minimal energy is sought for. Therefore, if you have N>>1 mesh elements and a large percentage of the overall energy is localized only within few mesh elements n, with n<<N, it is clear that the calculation is going to be unstable. The cloaking structure has this problem on it's inner surface, so you can improve both the accuracy and stability of the simulation by increasing the number of mesh elements near the cloak's inner surface.
So:
(a) the solver issue is not scattering-specific, it is cloak-specific
(b) when you solve, put more mesh elements wherever you think the field energy should be higher.
Regards
Hi goran1901,
I have tried to reproduce the results shown in Appl.Phys. Lett. 91 (Shalaev, 2007) for the exact case (eq. 2). I have used the same parameters of the above paper (except for the wavelength: for the lambda in the paper I got terrible results...), but I'm not sure of the accuracy of my results.
I have the same problems you told me in your helpful answer (about the memory and the convergence) with an Intel processor Core2 with 2GB RAM and Windows XP. Could you please check my file with your computer and tell me if you are able to further refine the mesh and\or if you get better results?
Thanks thanks thanks in advance for your precious help.
Hi iaia,
I just saw your post and took a very brief look on the file you sent me. There are couple of things I don't understand:
1. why are you using the hybrid-wave simulation; the point of the paper in APL from Cai and Shalaev is that they use nonmagnetic cloaks for TM waves.
2. if so, why do you have a magnetic cloak (the values of mu),
3. what is the problem: the results apper to be ok? You compared to the results from the paper. Which results? Have you calculated the scattering cross sections?
4. I just want to add regarding the scatt. cross section and about what we discussed: you know that the fields you are to integrate are just the scattering fields, right? If you integrate the overall fields, you will get 'meaningless' results because of the interference.
Anyway, I need some time to go more carefully through both the APL paper (I haven't done nonmagnetic cloaks) and you simulation.
At this point I just one to add that it is not weird that the cloak doesn't work quite well if you don't use the linear mapping (in the APL paper they use the second order mapping). For example, I have used some exponential mappings and the cloak behaved weird, though it should have been the same as for the linear mapping from the theoretical point of view. The difference is due to the cloak's imperfection (finite mesh size and finite values of epsilon and mu...)
I haven't been using the hybrid-mode so far, so I need to look up what is actually simulated there.
Therefore, I hope to go through it during the weekend and then to get back to you. At the moment I am quite busy with some other work. Thanks for the interesting suggestion.
Perhaps we could start a topic regarding the use of COMSOL on Core 2 Duo machines and try to figure out why does it perform so poorly. Perhaps there are people who got it figured out.
I hope this post is not to confusing :)
Regards
Hi goran1901,
thanks for your answer.
In the above mentioned paper there are two different cloak: the exact one(that I have posted here and corresponding to eq. 2) and the non magnetic one (corresponding to eq. 3).
For the first cloak I have used the hybrid-wave simulation because I need to choose both eps and mu anisotropic (see eq.2). In this case, the cloak should be perfect (as the one proposed by Pendry in Phys. Rev. E 74, 2006) and if you want I can send you my comsol file with the Pendry's cloak (it works perfectly).
I have compared my results (the ones in the file I have posted here) with what they should have been: they should have been exact like the ones by Pendry.
I haven't calculate the scattering cross section yet, but I suppose that I have to substract from the total field the incident plane wave.
Thank you again for your help.
Regards
Hi iaia!
Ok, I understand - you are not doing the nonmagnetic cloak and the paper from APL (Cai and Shalaev) you mentioned only because it deals with the second order mapping.
Yes, you need to substract the incident field (plane wave) from the total field and only then to integrate. This substraction is the same as in the example from the Comsol's EM user guide (the one from which I guess you took the parameters for PML) - the scattering on a dielectric scatterer or however it is called.
Both mu and epsilon are anisotropic and both are equal, yes. I guess you know that in the 2D geometry (i.e. when all the quantities are invariant along the z-axis) where the anisotropy can be described by diagonal tensors (like here in the cylindrical coordinates) the TE and TM modes are decoupled. In other words, you can define TE and TM waves. For TE waves only eps_z (because E has only a z component) and mu_xx,mu_xy,mu_yx,mu_yy (because H has components only in the x-y plane) matter. The other components of the eps and mu tensors do not infuence the TE wave. For TM waves it's mu_z and eps_xx,yx,xy,yy. This is exactly what Comsol offers you to set in these simulations. In other words: you don't have to use the hybrid mode. I don't say it's not good - that's some other thing, I'll have to sit down and read a bit and then I will tell you. Right now, I can't do that.
Ok, back to mapping. Do you understand me what I mean by the mapping? You can put it this way: the mapping is the thing that determines the parameters of the cloak. The number of possible realizations of a cloak with given parameters (R1 - the inner radius and R2 - the outer radius) is the same as a number of functions that map some parameter rho to the radial coordinate r. Ok, these mappings have to satisfy some basic assumptions, but my point here is that there is infinitely many mappings. The cylindrical cloak from the PRE paper (Cummer, the first reported simulation) is the simplest case - for a linear mapping. In the APL paper they discuss I think general mappings and particularly the second order mapping. Theoretically, all these mappings should give ideal cloaks, however it's not so simple because any realistic cloak, including the one implemented in a numerical simulation in Comsol is not perfect. I think there are different kinds of imperfections, but I wouldn't go into that now. So, again, we come to the point that weird things can happen if you use nonlinear mappings. I already mentioned you that I noticed that a exponential-like mapping leads to funny behaviour.
I hope that now you understand that there is no reason to think that the cloak discussed in the APL paper (second order mapping) should give the same results as the first order cloak (i.e. mapping) from the PRE paper. My advise is that you establish a firm criteria for the cloak quality (like the scattering cross section) and that you then compare different cloaks. However, I must warn you that it is highly probable that the mere mesh implementation will lead to very different results. You have to figure out some way to be systematic about it. Perhaps you should read the Phys. Rev. papers published recently on the cylindrical cloak:
Phys. Rev. Lett. 99 113903 (Z. Ruan et al.)
Phys. Rev. B 76 121101 (B. Zhang et al.)
There you go :) I could have already written a paper on cloaks :)
Regards
Added after 37 seconds:
can you access these papers? do you want me to send them to you?