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

Refraction of light

  1. Jun 19, 2003 #1
    Light gets refracted when travelling from one medium to another, but why? Is it because the photons interact with the electrons?
  2. jcsd
  3. Jun 19, 2003 #2
    The question of "why" is IMO always problematic. Anyway, I think the answer to your question is: yes.
  4. Jun 20, 2003 #3


    User Avatar

    Light refracts when going from one medium to the next b/c light travels at different speeds in each of the media. It's the same thing as a lawn mower rolling across a grassy field from a concrete sidewalk at an angle. Once the 1st wheel goes over the grass it slows down while the other three wheels are still travelling at the same speed. This causes the lawn mower to veer in the direction of the slow wheel. If you look at the overall path of the lawn mower you'll see that the mower refracts the same way light does. The key is difference in speeds otherwise no refraction occurs.
  5. Jun 20, 2003 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    A photon is adsorbed by an atom, this means an electron is elevated to an excited state. After a brief time the electron decays to a lower energy level, not necessarily where it started. It is this delay between adsorbtion and emission that reduces the aparent speed of light in a medium. The energy (wavelength) of light emitted determins the color or the material. A very good source for a understandable explanintion of the physics behind this is Richard Fynmans QED
  6. Jun 21, 2003 #5
    Integral, thanks for the explanation, but this makes me ask another question. When light is refracted, I was told that frequency stays the same, it is only wavelength that changes, therefore the speed changes. But I am really mixed up with this. Frequency is simply the number of waves per second, measured in hertz, and wavelength is the distance between two peaks of the wave. So why can't you say the wavelength stays the same and it is the frequency that is changing. Anyway if you change the wavelength, for example the wavelength is longer than the wave would be red-shifted, but since the wavelength is longer, the number of waves per second would also decrease, wouldn't it?
  7. Jun 21, 2003 #6
    If frequencies are different in different media, vibration of light wave becomes discontinuous at the boundary between different media because rate of vibration (frequency) is different. Light wave of one ray should be continuous, should'nt it. Even if wavelengths are different in different media, light wave can be continuous at the boundary between different media. Nature dislikes discontinuity.
  8. Jun 21, 2003 #7
    Interesting explanation, Integral. Does that mean that the angle of refraction is actually a statistical quantity (and therefore the index of refraction)? Since the re-emission of an absorbed photon is a probabilistic event, so the time taken and thus angle of refraction should be a probabilistic event as well.
  9. Jun 21, 2003 #8


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Again I will refer you to QED. find a copy of this little book written for the non scientist and read it. The wavelenghth of light determines the color of the light so any shift in wavelenght is due to the mechanism I described above.

    The speed of light between atoms is c, the apparent slowing is due to the time spend adsorbing and reemitting in a medium.

    At any sufrace some light is reflected (ie the emitten photon leaves the material) some is transmitted on into the material to be reemitted. All photons are emitted in a totally random direction. Think about it, different observers in different locations will seen the reflection of an object in different places. Photons are scattered in all directions, the law of "angle of reflection = angle of incidence" is due to the fact that the image appears in a location determined by the minimun optical path length. That is determined by the location of the observer. This is a matter of energy and time.
  10. Jun 21, 2003 #9
    Not exactly. A transparent medium doesn't have any allowed energy transitions by which the photon could be absorbed. Instead, When a photon interacts with an atom in the transparent medium it momentarily forms a combined atom-photon energy state, called a dressed state or a virtual energy state. This virtual state is not an "allowed" energy state of the atom, so the atom almost immediately re-emits the photon. However virtual interactions do take some time to happen, and this is what is responsible for the apparent slowing of light as it passes through the medium.
  11. Jun 21, 2003 #10
    Oh man! So THAT'S why some media are transparent to light! (by the photon theory)

    That'd explain so clearly why heavy elements will stop high energy photons better than light elements (that, along with plain ol' density..)

    Anywayz, the frequency stays the same because the time between one wave crest hitting the line and another wave crest hitting is the same time as the wave crest that was transmitted takes until the other wave crest gets transmitted

    Think of a little picture of a wave going along:


    then when it hits the other medium, it looks like this on the other side: (sort of.. actually it should be scrunched down and not a triangle wave, but there's a limit to what I can do with ascii art)


    anyway, it's going along and hits the medium, like this:


    (the crests have to match up to crests)

    now, the time it takes for the wave to move along one wavelength on the left is the same as it takes to move one wavelength on the right


    I dunno if that's lucid enough.. I got it once I saw it in my Physics tutorial section, but I couldn't explain it to my group-members, so they had to fumble around for half an hour before they understood too
  12. Jun 22, 2003 #11
    Virtual energy state. Uh... I can't understand exactly. I think in a transparent medium absorption and reemitting do not occor. These are related to inelastic scattering between a photon and an electron in an atom. But perhaps, relevant phenomena is elastic scattering. At long enough wavelength, photon( as well as wave) elastically scatters with electrons of atoms, causing polarization. This polarization causes dipole radiation of the atom. According to the property of the dipole radiation, wavelength of light in the medium is determined. Superposing dipole radiations from many atoms causes interference, making a travelling light in the medium. At short enough wavelength, electrons are scattered out by the elastic scattering with photons, which is photoelectric effect by Einstein. Does anyone have another opinion?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook