Light and Magnets... and Gravity

Most recent answer: 10/22/2007

How far can a magnetic field bend light? Could it be bent enough so that it goes around a three dimensional object and comes out (after going 180 degrees, I guess) the other side, being bent by the magnet, therefore making it seem as though the object had the light bass through it completely, and it does not appear to the human eye? Here is the best example I can give: -->=light, O=Object, *=mag. field, / or \=light bent. ___ |***| If this can be done, please tell --->/*O*\---> me how, and what would be needed to perform such a thing. Thank you. (PS - This would need to be done in a round sense, as in all around the 3-d object, not in just one line, so please keep that in consideration. Also, if you know any other way to get the desired effect please inform me.)
- Jon (age 15)
The King’s Academy, Marion, IN, U.S.A.
Hi Jon --

Nice try. Unfortunately, the path light takes is not affected by the presence of a magnetic field. Light itself is composed of an oscillating electric and magnetic field, and one very important property of electric and magnetic fields is what we call "linearity." That is, if you have two sources of electric and/or magnetic fields, you can predict what the combined field is just by adding the two source fields together. The two fields don’t change each other at all. So if you add the field of a light ray to any other field we can imagine, the light ray will continue as before and the extra field will just stay the same, adding to it in places where the extra field is strong, but having no effect beyond the reach of the extra field. So there is no way that a magnetic field can bend light.

Although magnetic fields might not do the trick for you, there is quite a bit more about light that can be taken advantage of. For instance, if the object is very small (small compared to the wavelength of the light), the light will simply diffract around the object and be invisible or nearly so anyway. Please see our .

Also, even though a magnetic field won’t do anything for the light, a gravitational field, sufficiently strong, will in fact bend light. This was observed early in the 20th century confirming Einstein’s General Theory of Relativity in which light from the planet Mercury was bent by a very tiny amount by the enormous gravitational field of the sun. Unfortunately they needed a total solar eclipse to block out the bright sun’s rays so they could see the feeble light from Mercury on the other side of the sun passing close by where the effect was big enough to be measured.

One of the more spectacular demonstrations of bending light by a gravitational field is "gravitational lensing" of light from very faraway bright objects whose light passes close by another distant object with plenty of mass. Here is a description, along with some *very* nice pictures:


The problem with making an object invisible in this way is that the light cannot be bent any way desired -- what you get is just lensing, and you get multiple images of the object in back of the object doing the light-bending. An object like this will attract quite a lot of attention! These galaxies sure attracted our attention. And you need a galaxy-sized gravitational field to do the trick.

To make a gravitational "invisibility cloak" requires quite a strong gravitational field. Maybe the best way to do it is to have such a strong field that the light just gets sucked into the object and cannot come out. Then you have a black hole. Not invisible because you can tell that your light is gone, but interesting nonetheless. Here’s a tutorial on black holes:

Now the disclaimers back on your original question: If your magnetic field is strong enough over a large enough distance, you can have enough energy stored in it to do gravitational lensing, and then refer to the above answer on gravitational lensing. This however is a very difficult way of getting a strong gravitational field. It is much easier just to collect a galaxy’s worth of matter than to collect the equivalent energy in a magnetic field (neither is particularly easy, I admit!)

The second disclaimer is that there is a small expected deviation from linearity of electric and magnetic fields due to quantum mechanics and the ability of electrons to pop out and go away on microscopic time scales. This only becomes noticeable for very very high frequency light colliding with other very very high-frequency light (it wouldn’t be noticeable and may even have exactly zero effect for a static magnetic field and visible light -- I haven’t done any calculations). There are plans to make such a light-light collider, but it requires a many-mile electron accelerator to get the energy of the light high enough.

If you are really interested in getting light to go around a solid object so that it looks like it went through, the low-tech solution might be the easiest. Magicians use mirrors for this purpose all the time!


(published on 10/22/2007)

Follow-Up #1: shenanigans?

Your saying, light cannot be bent by a magnetic field, and I call shananigans..In fact, what your saying is unproven and not scientific. Assuming the Einstien's equation of general relativity is valid.. E = mc^2 , energy is equal to mass * speed of light ^2 If light, is simply a wave expressed as a ''photon'' and does not distort matter around it because it has no mass..It would seem to indicate that gravity would have no affect on light, mass cannot affect something which is massless, and thus light would not bend due to Gravitation. This is not the case, sooo it is safe to assume that light has mass.. also mathmatically, If a photon does not have mass,then E = mc^2 would seem to imply that m = 0 and E = 0, speed = 0 which is certainly not the case. Now because we assume that a photon has mass, now a photon must then be affected by gravity, and electromagnetism, even if it is a very insignificantly small amount of change. IF you take it that mass, is a distortion in the speed of a traveling wave, because E=mc^c, the waves which have a slower velocity, are also more massive, and more easily disturbed by gravity.. this means that I can put an electric current through a wire and transmit information (like radio), but that the farther it travels the more distortion it has until it is completely incomprehensible to instrumentation because the wave dissipated due to gravity and the interactions between the transmitted waves and the slightly ionic atmosphere. IF we assume the previous 2 paragraphs are correct, then we can assume light is minimally affected by magnetic/gravitational forces, because it has an extremely high frequency, and extremely low mass. Peace
- Dr.Hugo (age 20)
Doc- Your basic point- that since magnetic fields (including that of light) have energy, they give rise to gravity, which can affect other light, is correct. In fact, Tom made exactly that point in the original post: "Now the disclaimers back on your original question: If your magnetic field is strong enough over a large enough distance, you can have enough energy stored in it to do gravitational lensing, and then refer to the above answer on gravitational lensing."

I'm not quite sure what you mean by your last points. The gravitational self-interaction of light would not cause information loss, so far as we know. In practical situations the effect on the light propagation is negligible even on cosmic scales.

Mike W.

(published on 03/27/2010)

Follow-Up #2: Gravity vs. electromagnetic light bending

Is there actual evidence that it is gravity affecting light? I mean, the sun, black holes, galaxies all give of energy in the form of light/radiation, so there must be an EM field present. So how can they tell the difference between gravities influence and that of an EM field?
- Daniel (age 18)
Perth, WA
Sure, this is a question on which there's a ton of evidence. To begin with, we have a beautifully accurate theory of electromagnetism, tested countless times here on earth. So we know that EM fields don't bend the paths of light, except to the negligible extent that their energy is itself a source of gravity, or that there's quantum-mechanical photon-photon scattering. Here on earth, you can shine two laser beams through each other or shine a beam through a big magnetic field, etc. to check that we're not missing something about electromagnetism. You may want to know how accurate our EM theory (quantum electrodynamics) actually is. In one case (the electron gyromagnetic ratio) it predicts a number to an accuracy of one part in one hundred billion.

Then we have a theory of gravity, General Relativity. It also make a wide array of very accurate predictions, including for the bending of light near stars. Its many predictions include the speeding-up of higher-up clocks, an effect which must be accurately considered in order to keep GPS systems from going far off. That gravitational clock speeding-up, incidentally, accounts for half of the bending of light.

For a nice book on the confirmations of GR, see Clifford Wills "Was Einstein Right?".

Your question reminds us of a theme which scientists have done a poor job of communicating. There are many questions in science where we're grappling with a variety of semi-quantitative ideas to understand the behavior of complex systems. There are other areas in which beautiful thoroughly confirmed precise general theories give us the answers. And there are a range of cases in between. This is one of the cases where we're not floundering, we know precisely what's going on.

Mike W.

(published on 03/01/2011)

Follow-up on this answer