Everyone who has taken an introductory class to physics has seen a Prism bend light: https://en.wikipedia.org/wiki/Prism

Or, more visually: https://www.google.com/search?q=optical+prism&tbm=isch

That is, white light goes in one end, all of the colors of light come out the other end.

That is, the light is split into its constituent parts. One wave is split into the waves of various different wavelengths comprising it.

Now, I totally understand that.

But if that's the case, then it stands to reason that a Prism, under the right conditions, might output or be made to output other non-visible wavelengths on the electromagnetic spectrum:


Now, increasing in frequency from visible light we have ultraviolet, x-rays, gamma rays, and the like.

I'm not interested in any of those. (It seems unlikely that you're going to get a higher frequency wave from a lower frequency one.)

But, directly below the frequency of visible light, we have infrared, microwaves, and below that, rf (radio frequency).

My question is simply this... has any physicist out there used a light source, a prism, and and a detector of any kind and observed any kind of non-visible electromagnetic wave/wavelength (in addition to the various colors of light) being emitted by the prism?

Infrared would be the first thing to try and detect, and if successful, then maybe the next thing to try to detect would be high-frequency microwaves, then lower frequency ones, and finally, if that could be accomplished, RF.

Maybe it wouldn't work.

But the logic looks something like this:

a) Light is an electromagnetic wave comprised of other (slower) electromagnetic waves.

b) A Prism is able to separate out those slower electromagnetic waves and show the visible ones as light.

But, are there any invisible ones as well?

It seems logical that under certain conditions (very big Prism, strong angle of attack, ?) that you might be able to get non-visible slower electromagnetic waves...

I don't have the tools or materials to confirm or deny this, and I was looking for a discussion about it from the StackExchange community...

Anyway, that's the question.


There are two main cases:

  1. white light that is made up of only visible wavelength photons, in this case there would be no non-visible components in the prizm

  2. the Sun's white light (it is a misconception that the Sun has yellow light, in reality it is white, made up of all the visible wavelengths and non visibles too), in this case, the light coming from the Sun does include non-visible wavelength photons that you talk about, and the prizm would separate those too.

Now you are saying that light is an EM wave that is made up of other waves that are slower. In reality all wavelength photons travel at speed c in vacuum when measured locally. It is at the edge of the other medium, in this case the prizm that the different wavelength photons change angle in different amounts, so the different wavelength photons will travel different paths in the prizm.


First off, its important to remember that a prism cannot create EM radiation of any frequency. It merely separates out radiation of different wavelengths via a method called dispersion.

The amount of dispersion we see is based on the refractive index of a material, which is a function of wavelength. That's why different wavelengths get bent differently. The relationship between refractive index and wavelength is not simple. There are several empirical curve fits, from simple Abbe numbers to more complex Cauchy and Sellmeier equations.

We'd have to get really deep into the materials to discussion whether particular setups work. However, we can get a sense of the complexity of this by looking at the refractive index of air across a wide band of frequencies, borrowing an image from another stack exchange answer:

Refractive index

Clearly, while refractive indexes are simple over small bands, looking at wider ranges of bands is quite a complicated task indeed!


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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