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

Seeing color

  1. Oct 3, 2009 #1
    If there is a "red" color ball people told me that if i shine white light on it the "red" color ball absorbs all the light and reflects "red" photons hence we perceive it as "red" colored

    Is this true? if this is right i want to know more details... what does reflect mean here? what does absorb means?
     
  2. jcsd
  3. Oct 3, 2009 #2

    Pythagorean

    User Avatar
    Gold Member

    Yes, it's true.

    When atoms absorb photons, the electron actually absorbs it and transitions up in energy state. The electron will then transition back down, kicking out a photon. The photon then hits your eye and your brain interprets the "color" of the photon and you perceive it as red.

    What we call reflection is actually an absorption followed by a transmission (the electron takes up a photon and transitions up, then transitions down, kicking a photon back out).

    So there is no such thing as color in absolute dark. It's not just that you can't see it, it's literally not there without light.
     
  4. Oct 3, 2009 #3

    alxm

    User Avatar
    Science Advisor

    Objects have color whether there's light being shone on them or not. Saying something has a certain color is saying that it absorbs certain kinds of light in the visual spectrum, which in turn depends on it energy levels/bands. Those band-gaps are still there whether or not light is present.
     
  5. Oct 3, 2009 #4
    Thanks for the explaination.
    So a red ball appears red because the electron in the material which was used to paint it emits red photons when they jump down great so far but what about the shine thats on the paint it looks like as if some of the light is bouncing off of the ball?
     
  6. Oct 3, 2009 #5

    Pythagorean

    User Avatar
    Gold Member

    The band gaps are still there, but in my opinion the actual photon packets carry information about the color. This is more of a philosophical discussion, though...

    I'm not really sure. I'm guessing it's the result of a diffraction grating on an amorphous surface, producing a wider band of wavelengths and directing them to a smaller area (as white light is a mixture of many different colors of light) because of the extreme angle of incidence and the larger variety of distances between gratings.

    You see this with pavement too as your viewing angle of incidence becomes greater (in other words, if you look straight down at the pavement, it doesn't reflect, but if you lay down in the road and look at it from an angle, or just look at pavement far away, you begin to see it reflect).
     
  7. Oct 4, 2009 #6

    alxm

    User Avatar
    Science Advisor

    It's not a philosophical discussion. There are two distinct physical scenarios: One is that the property of color is 'induced' by light and doesn't exist if no light is present, the other is that the property exists independent of whether any light is present or not.

    The property we perceive as the color of an object belongs in the second category. If you're talking about nonlinear optical effects (for instance) then it's in the first category.

    It is. When light hits a medium (such as the paint), some of it is reflected, depending on what the refractive index is. (glossy paint) More light will also be reflected back at you from a smooth surface. You can also make something glossy by putting a coat of transparent paint on top, in which case light will be reflected both at the boundary between the air and the transparent coat, and at the boundary between the transparent coat and the paint.
     
  8. Oct 4, 2009 #7
    I was just trying to imagine how does light reflect when it hits the boundary atoms of the shiny paint it just bounces off and doesnt get absorbed by the electron?
     
  9. Oct 4, 2009 #8

    Pythagorean

    User Avatar
    Gold Member

    You've just narrow-mindedly selected one philosophy and decided to call it the truth ('physical') without considering the definition. Let me show you another:

    I (and many other people) define color by frequency and wavelength of the photon. That's why monochromatic light means "one-color light". It's light that has one wavelength (i.e. one color). You can rest assured that there is no red in the dark. There is only black (which in physics, is a lack of color).

    Bestowing the property of color onto the objects themselves comes from when were ignorant that photons even existed (and still reigns today with many people not understanding how photons must bounce off of an object and then enter the eye for the object to be seen).
     
  10. Oct 4, 2009 #9

    Pythagorean

    User Avatar
    Gold Member

    reflections IS an absorption followed by an emission in the atomic view.
     
  11. Oct 4, 2009 #10
    I'm going to question this simple explanation for the color mechanism. I think there are some pretty exotic molecules involved in the paint, ink and dye industries and I'm not sure electron levels in these molecules are responsible for the colors we see. In particular, I suspect that the special optically active frequencies in these molecules are ones which absorb light, and the colors we see are the result of light which is merely scattered at all other frequencies.
     
  12. Oct 5, 2009 #11
    OK, I'm going to change my answer. Now I think that the colors we see are the ones for which there are active optical transitions in the molecules. (Note in my previous answer I speculated that those were the colors which are absorbed). I think there must be a different mechanism for actual absorption. I propose interaction with lattice vibrations. I still have a problem in explaining why glass is transparent.
     
  13. Oct 5, 2009 #12

    Pythagorean

    User Avatar
    Gold Member

    If the molecule's required bandgap energy is not on the order of the energy of the photon, then the electron will not absorb the photon and it will simply pass through the molecule. (I'm not sure if literally doesn't interact or of it just absorbs and reemits with the same frequency). Transparent objects have band gaps that are not on the order of the energy of the photons associated with visual light.

    Addendum:

    I actually suspect it IS absorbed and reemitted in transmission as this would give an atomic basis for refraction, but I'm curious what someone more knowledgeable than I has to say about it.
     
    Last edited: Oct 5, 2009
  14. Oct 5, 2009 #13
    Thanks a lot Pythagorean you cleared up a lot of things.
     
  15. Oct 5, 2009 #14
    It seems that I am not understanding what mechanism is being proposed for absorption; that is, the conversion of optical frequency light to thermal energy in the solid.
     
  16. Oct 5, 2009 #15

    Pythagorean

    User Avatar
    Gold Member

    That is only one outcome. Thermal energy is often understood in terms of the kinetic energy of a particle. The electron is more excited, it has more kinetic energy, and thus more thermal energy. Here's some other outcomes of absorption:

    from http://science.howstuffworks.com/question404.htm

    I'm still curious whether the last one is specifically true, or whether it's an absorption followed by an immediate emission. In the following physicsforum thread, Chi Meson states it as a an absorption followed by an emission:

    https://www.physicsforums.com/showthread.php?t=11209
     
  17. Oct 6, 2009 #16
    No, that's not what you said before. You said when an electron absorbs a photon, it goes to a higher energy state, and then it emits another photon. That's reflection. Then you said some materials have no available transitions in the optical zone, so the photon passes right through. That's transmission. I still don't see a mechanism for absorption: a photon goes in and nothing comes back out.
     
  18. Oct 6, 2009 #17

    Pythagorean

    User Avatar
    Gold Member

    I was answering a different question than you have in your head, particularly the OP's question. I've only recently begun entertaining your particular questions.

    We also haven't discussed the photoelectric effect, or probably a myriad of other effects that I'm ignorant about. You can't expect an all-encompassing universal panacea of an answer every time a question gets asked.

    well... absorption IS the mechanism of converting a photon into kinetic energy. That's what kicks the electron up an energy state, kinetic energy. It gives up that kinetic energy when it goes back down (kicking out a photon)

    Generally, electrons like to be in their lowest available energy state, but if that energy state is not available (due to exclusion principle or constant supply of energy, such as light) it will stay in its excited state (at least on average). If you have a lot of atoms doing this, you detect it as heat and there's probably still emitted photons associated with it (infrared), but the electron keeps getting supplied, so it doesn't stay in it's lowest available state very long before more energy kicks it back up. Also, you have to consider that thermal energy gets supplied TO the electron as well, depending on the band gap.

    I mean, realistically, there's probably very few situations where the electron stays in one state for a long time, so it's hard to imagine a scenario that you describe... where the electron absorbs and holds energy indefinitely.
     
  19. Oct 6, 2009 #18
    From what I can understand of your explanation, for every photon of a given frequency that enters a solid, another one (or the same one) leaves at the same frequency.
     
  20. Oct 7, 2009 #19

    Pythagorean

    User Avatar
    Gold Member

    Not necessarily. Two smaller photons could come back out, for instance. Or the electron could break completely free of it's nucleus given sufficient energy (the photoelectric effect).

    But there's also entropy to consider. In the conversion, some of the photon's energy will be lost as heat.

    There's also phonon excitation to consider. Phonons are the quantized modes of vibration in the lattice structure itself (i.e. a structure of nuclei in the material basically vibrate in a wave-like motion). Phonons can be caused by free electrons whizzing by, but if I recall correctly, they can also be caused by kinetic energy gained from heat or acoustics.
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Seeing color
  1. Quantum Color (Replies: 7)

Loading...