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

Optics: Soap Film and Other Things

  1. May 27, 2010 #1
    I understand the idea behind the optical interference that produces colors on thin films but have never figured out the reason that the films have to by 'thin'. What is the lack of similarity that I am missing between a film and a somewhat thick sheet of glass or something that makes these effects not visible in the thick thing? Is it merely that the thin films that do this sort of thing are very much close to perfect surfaces so that the effects can be recognized (and the thick things like glass are very imperfect and you can't see the now *very thin* bands)? Or is there something more sinister at play?
    Last edited: May 27, 2010
  2. jcsd
  3. May 27, 2010 #2


    User Avatar
    Gold Member

    Well, yeah. You're on the right track.

    As the thickness increases, the phases of the incident and reflected rays become more and more randomized.
  4. May 27, 2010 #3
    What is the cause of the randomization of the phases?
  5. May 28, 2010 #4

    Andy Resnick

    User Avatar
    Science Advisor
    Education Advisor

    DaveC426913 is correct, but I would have phrased the answer differently- when the thickness of the soap film exceeds the coherence length of the incident light, there is no interference.

    The coherence length is given by the spread of frequencies present, and is a measure of how unequal the arms of a Mach-Zender interferometer can be while still producing interference fringes.
  6. May 28, 2010 #5


    User Avatar
    Gold Member

    Interesting question.

    Let me explain why I think it interesting. In the case of a double slit setup there are numerous interference fringes. There will be one luminous area where the two paths that the light has followed are equal in length, next to that there will be areas where the length difference of the paths is a single wavelength, next to that the areas where the difference is two wavelengths, etc.
    That illustrates that the the difference in pathlength can be a multiple of the wavelength; you always get interference effects.

    Under the right conditions it should be possible to elicit interference effect with a thick layer.
    If I hazard a guess I think interference effects with a thick layer can only be elicited when the light strikes the layer at right angles.

    In the case of a thin film of petrol on water the interference effects are visible from all angles. I wonder: when light enters a fluid its direction is changed. If we are looking at a puddle of water with a petrol film on it then some of the light has reflected directly on the petrol, and some of the light has entered the petrol, it has reflected on the petrol/water boundary, and then it has exited the petrol again.

    My guess is: if the film's thickness is about as large as a single wavelength of the light then the path of entering the petrol and exiting again is hardly displaced sideways compared to the pure reflection. It is as if the light is from the same source, but with different phase because the reflections were different.

    As DaveC426913 points out, the thicker the layer, the more sideways displacement. It's no longer as if from the same source.

    Generally, interference effects will occur only if the setup allows only a small set of possible pathways for the light
    For instance, in a double slit setup the source of the light (the source that illuminates the double slit) must be a point source. You can use sunlight, but the Sun itself is not a point source, so if the double slit is illuminated by the Sun then there are no interference effects. To get interference effects the Sun must illuminate a barrier with a sufficiently small hole in it, and the light entering through that hole then illuminates the double slit.
  7. May 28, 2010 #6
    I just read in my textbook (Serway) that it's because "If the film is more than a few wavelengths thick, the interference fringes are so close together that
    you cannot resolve them." Not sure exactly how that follows...
  8. May 28, 2010 #7
    I buy that (and that was what I was thinking intuitively). I see that the path length difference per angular displacement will be more for a thicker object and that would explain the thin bands.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook