Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Mixing electromagnetic frequencies

  1. Jul 25, 2003 #1
    If one can mix certain colors of the visible spectrum to produce other colors, is it possible to mix frequencies from other parts of the electromagnetic spectrum to produce visible colors?
  2. jcsd
  3. Jul 25, 2003 #2


    User Avatar
    Science Advisor

    I doubt it. The reason we can mix colors to get other colors is that the eye (cones and rods on the retina) can respond to the different colors and the brain interprets the result. Frequencies outside the visible would have no effect on the eye.
  4. Jul 25, 2003 #3
    Well our retina respond to a range of frequencies of EM radiation. In principle, couldn't we re-construct a frequency in the visual range with frequencies outside of that range using Fourier Synthesis? Of course using high energy radiation would cause cell damage but ignoring this detail, I don't see why this wouldn't work...

  5. Jul 25, 2003 #4


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    How about making green light as a "beat" frequency?

    Two high frequency signals which combine constructively at the frequency of green

    then the green light would be physically present (not just an artifact of the retina cells and how they perceive mixed frequencies)

    edit: I see now that may be what you were suggesting, Alain
  6. Jul 25, 2003 #5
    Two waves in the electromagnetic spectrum with different frequencies do not generate a wave with frequency the sum of they both. Waves with different frequencies do not interact between them.

    The colors depends only on the brain interpretation, as said mathman.
    The fact that a sum of stimuls (different wavelenghts) generates certain colors is only physiological.
  7. Jul 25, 2003 #6
    Try looking into how your eyes work.

    We will interpret a mixture of yellow and blue light as green light, even if there is no green light present. This is due to our perception not because yellow and blue frequencies make green (they don't).
  8. Jul 28, 2003 #7
    It can, but perhaps not in the way you are thinking. This can be done in an optical parametric amplification (OPA) system. Two laser beams (a pump and a seed) generated from an ultrafast laser source (for example, a Ti:Sapphire amplifier) enter the OPA system, and undergo sum frequency generation inside a non-linear BBO crystal. Typically, the pump and seed pulses might have wavelengths around 800 nm, placing them in the near-IR region of the spectrum. By adjusting various paramaters (primarily, the orientation of the crystal) one can tune the output wavelength of the OPA system to cover a large range, say 400 nm (visible) --> 18 microns (far-IR).
  9. Jul 28, 2003 #8
    The three different types of cones in our retina have different absorption characteristics as a function of EM wavelength with peak absorptions in the Red, Green and Blue regions of the optical spectrum. The colors we perceive in an interpretation of the information coming to the brain from these three categories of stimulus. This I agree with.

    The question in contention can be reformulated as follow: Would a Cone Cell built to respond to a Redish frequency actually fire if stimulated by an EM wave composed of many frequencies that actually yields a red frequency at intereaction time with the cone?

    I believe the answer is yes: EM waves form interferece patterns in the double-slit experiment so they DO interact constructively and destructively. As such, I think it's resonnable to assume that waves of different frequencies would locally (at cone location) interact and be "perceived" by the cone as the synthesized magnitude and frequency of all incident radiation. I'm no physicists but I believe you would model this in QM as the wavefuntion superposition of all incident photons. Remember that all energy transfer is quantized, and the quantum phenomena of interaction is interpreted as the notorious "collapse of the wavefunction" which is a superposition of all wavefunctions present in a specific spatial location, bearing in mind the uncertainty principle.
  10. Jul 28, 2003 #9
    Really the peak of sensibility of the eye is at 555 nm in photopic vision( this is when the vision is centered in the phovea, where the cone density is the most high), and at 510 nm with scotopic vision (at night for example).

    I never heard about three types of cones. If you are thinking in the RGB system, this is only a system for color specification. (Not only the one)

    Well, I think the eye can not perceive nothing out of its range of sensibility. (this is the visible light).

    Leaving away some non-linear phenomena you need tree basic things to make two monochromatic waves to interact.
    -They must have the same wavelenght.
    -They must have the same plane of polarization
    -They must be coherent.

    The result will be a wave with the same wavelenght than the original.

  11. Jul 28, 2003 #10

    Actually there are two complementing theory of colour vision: Trichromatic Theory and Opponent-Process Theory. These two theories are complementary and explain processes that operate at different levels of the visual system. In the former, there are three types of receptors, each with a different peak sensitivity: S-Cone: 445nm, M-Cone: 535nm and L-Cone: 570nm


    You are totally right about this. From www.opticsexpress.org/abstract.cfm?URI=OPEX-11-8-944

    April 21, 2003
    "When two electromagnetic fields of different frequencies are physically superposed, the linear superposition equation implies that the fields readjust themselves into a new mean frequency whose common amplitude undulates at half their difference frequency. Neither of these mathematical frequencies are measurable quantities. We present a set of experiments underscoring that optical fields do not interfere with each other or modify themselves into a new frequency even when they are physically superposed. The multi-frequency interference effects are manifest only in materials with broad absorption bands as their constituent diploes attempt to respond collectively and simultaneously to all the optical frequencies of the superposed fields. Interference is causal and real since the dipoles carry out the operation of summation dictated by their quantum mechanical properties."

    Futz was right too :smile:

    Puzzeling though????
  12. Jul 28, 2003 #11
    You may be interested in an effect, I beleive the correct name is the Hadley-Twiss-Brown effect (delightfully British name) which is when two light sources have a large spatial seperation but a small angular separation as seen by the obeserver, the light from both tends to be in phase. The example given was the light from a street lamp and from a star. It's purely quantum mechanical and has to do with Bose statistics.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook