Is it possible to have matter that reflects all the light and does not absorb light at all? If thats possible then we can store light inside it right? I think we can trap light inside that matter in shape of box for example and also we can release it by making very small dot in that box.

Im not saying that this is useful or not (I think its useless though because we can already emit light in other ways). Im just curios about this. Im noob in physics so sorry for my noob question. Thanks in advance.

Edit: If it does not make sense for matter to have reflection then what about shapes to reflect all light?

  • 4
    $\begingroup$ Light does not reflect from matter. Light reflects from an interface between two media. You can have such an interface, that 100% of light will undergo complete internal reflection when bouncing at an appropriate angle of incidence. $\endgroup$
    – LLlAMnYP
    Oct 25, 2015 at 12:33
  • 1
    $\begingroup$ See en.wikipedia.org/wiki/Total_internal_reflection . While storing light indefinitely isn't possible, because there are always losses, I believe it's possible to have light bouncing around inside an optic fibre for a measurable fraction of a second. (This happens whenever signals are sent over under-sea fiber-optic cables.) $\endgroup$
    – N. Virgo
    Oct 25, 2015 at 14:03

2 Answers 2


First, "light" is just the name we give to a small set of frequencies in the electromagnetic waves (and it is sometime extended to neighboors such as UV and IR). -> do you mean reflecting 100% at any electromagnetic frequencies, or specifically in the dark-red to dark-violet range ? or would you be happy if is was 100% only at some given laser-like frequency ?

Second, "exactly 100%" is... alot. Indeed, exact zero or exact 100% are notions that exist more in the world of maths than in the natural world. Think about this: suppose a 10x10x10 cm³ box which facing mirrors have reflection 0.999999999 (i.e. absorption 10⁻⁹) - assuming such magic material exist. At the speed of light, in 1 second a ray would bounces 3.10⁹ ( 3 billions ) times. Then this "really almost zero absorption per bounce" will yield a total absorption of... 95% ! ( $1- (1-10^{-9})^{3.10^9}$ ). So you can't expect to keep energy more than a few seconds even it this case.

Now you could have something else than matter. e.g. if you could bend or slow down to zero rays with some physical phenomenon. But well, assuming it exists, it would require a lot of energy.

So there would be a way more efficient solution: capture these photons with photovoltaic cells, then turn them back to photons with some LEDs. ;-) . Or more natural: let 'store' them temporarily as offset in electronic orbitals, then let these excitated orbitals release them back as photons later. This is... phosphorescence ;-) . These both solutions would give you back a lot more photons than a pure optic capture device.

  • $\begingroup$ You've missed a $1-\cdot$ in your formula in parentheses for total absorption. What you have in parentheses instead is total reflection ($=4.97\%$). $\endgroup$
    – Ruslan
    Oct 25, 2015 at 12:20
  • 1
    $\begingroup$ There are mirrors with more than 90% efficiency after 0.1 sec and 1 billion bounces, ie the ones of Haroche and Wineland. $\endgroup$
    – user46925
    Oct 25, 2015 at 15:12
  • 1
    $\begingroup$ wow ! (not far from my "almost perfect" above). Do you have some link ? I'm curious: is it only in the very normal direction ? in the (whole) visible band ? And a point I didn't addressed: I guess vaccum is required, since air absorption might get important ? $\endgroup$ Oct 25, 2015 at 15:31

Source http://www.researchgate.net/post/Can_the_100_reflection_be_achieved

1) The highest reflectivity conventional mirrors made for reflecting a laser beam have a reflectivity of about 99.999%. These mirrors are made for the advanced LIGO experiment which is an attempt to detect gravitational waves using a power recycled Michelson interferometer. The loss of these dielectric mirrors is primarily because of transmission loss. Absorption is less than 1 part per million. 2) Superconductors are not superconductive above a cut off frequency which is typically in the radio frequency range. However, even if the bulk material exhibits superconductive properties at the frequency being reflected, the superconductive mirrors would still exhibit some loss because of surface defects such as surface contamination, roughness and imperfect super conductive properties of the surface. 3) Total internal reflection (TIR) is virtually 100% reflection, but this answer is probably unsatisfactory. TIR requires an incidence angle less than a critical angle which for air/glass interface is about 42 degrees. Fiber optics utilize TIR and fiber optics can transmit laser beams over many kilometers with low loss. The loss that exists is primarily due to absorption and scatter in the bulk material, not reflection loss. Therefore, there are no conventional mirrors which produce 100% reflection.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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