I am not talking about mirrors, just a plain window made of glass like material. Would it be possible to allow light pass only in one direction but not the other?

  • 5
    $\begingroup$ It is not possible without a magnetic field, by time reversal invariance. $\endgroup$ – Ron Maimon Jan 3 '12 at 5:54
  • 2
    $\begingroup$ does scattering light allowed for not transmitting? certain windows panes scatter light thru one side such that from one side you can see clearly, while from the other side things are blurred.. $\endgroup$ – Vineet Menon Jan 3 '12 at 11:40

No. Or at least, not without special caveats.

Edit After comments by Mark Beadles, David Zaslavsky, and Ron Maimon, I should clarify that you cannot have a plain window that lets all light through in one direction but reflects all light coming the other direction. Thus, you cannot have a window that doesn't absorb radiation at all and also has the prescribed one-direction property.

Imagine the glass separates two rooms which are completely isolated systems, except that they interact by sending electromagnetic radiation through the glass. We'll say that light can only pass from left to right.

Further suppose the electromagnetic radiation in the rooms is a simple black-body spectrum, so each room has a well-defined temperature. Imagine the temperature is higher on the left.

Because light only passes from left to right, heat will be transferred from the hotter room on the left to the colder room on the right with no other effect. This violates the second law of thermodynamics.

  • 5
    $\begingroup$ Heat could still be transferred the other way via conduction, though... unless you put a layer of vacuum between the two rooms. $\endgroup$ – David Z Jan 2 '12 at 22:55
  • $\begingroup$ @David Yes, I was assuming the conduction was zero due to some unspecified mechanism; a vacuum would be good. $\endgroup$ – Mark Eichenlaub Jan 3 '12 at 2:23
  • 1
    $\begingroup$ @MarkEichenlaub In your scenario, the remarkable part is the zero conduction mechanism, not the one-way glass. $\endgroup$ – Mark Beadles Jan 3 '12 at 2:50
  • 5
    $\begingroup$ @MarkBeadles true, good point... now that I think about it, even if you put a layer of vacuum between the panes of glass, you'd still have to deal with thermal radiation from the panes themselves. And the glass would be at thermal equilibrium with whatever matter it contains, so it would act as an effective radiator based on its blackbody radiation alone. So even if you have a material that prevents radiation transmission in one direction, I don't think you can completely prevent heat transfer in that direction. $\endgroup$ – David Z Jan 3 '12 at 4:18
  • $\begingroup$ @Mark I don't understand your comment. As David pointed out, there's no conduction through a vacuum. $\endgroup$ – Mark Eichenlaub Jan 3 '12 at 4:45

You're looking for something called a "light diode" or "optical isolator". The simplest is two polarizing sheets 45 degrees apart, with a 45 degree Faraday rotator in between. Here's why it doesn't violate the second law of thermodynamics.

One of the components--the Faraday rotator--is thick and bulky and usually covers only a small area. So I guess it doesn't qualify as a "glass like material".

  • $\begingroup$ Do they sell these things? Can we buy an optical isolator or do we have to build one ourself? $\endgroup$ – Pacerier Feb 10 '13 at 15:27
  • $\begingroup$ Would be great to see an image of these things! $\endgroup$ – Fattie Nov 10 '15 at 13:02
  • $\begingroup$ Manoj Peiris explains the Faraday Isolator with laser: youtube.com/watch?v=JJp_-oalQK0 $\endgroup$ – neoneye Sep 5 at 21:53

Yes, evidently. A. B. Khanikaev, C. Wu, H. Mousavi, and G. Shvets, "One-Way Slow Light in Nonreciprocal Low-Symmetry Metamaterials," in Quantum Electronics and Laser Science Conference, OSA Technical Digest (CD) (Optical Society of America, 2011), paper QThA6. http://www.opticsinfobase.org/abstract.cfm?URI=QELS-2011-QThA6

But of course this is with a metamaterial, not "a plain window made of glass like material". I don't know whether a metamaterial fits your criteria or not.

  • $\begingroup$ Can we buy this anywhere? Or is the only way to build one ourselves? $\endgroup$ – Pacerier Feb 10 '13 at 18:11

Fundamentally, yes. One needs to break the time-reversal symmetry by either introducing a magnetic effect or modulating the "glass-like" materials dynamically.

Practically, this is a very active research area. One could, for example, inject carriers into silicon dioxide at a very high frequency to achieve the one-way effect.


protected by Qmechanic Oct 28 '13 at 17:15

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.