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

How wide is a photon?

  1. Oct 13, 2006 #1
    When thinking about the double-slit experiment, I realized - even after having a degree in physics - that I never learned about how wide a photon is? Yes I learned about wavelengths, but amplitude was used to measure intensity, and was never expressed in units of length. What have I overlooked/forgotten/never-learned! here?
     
  2. jcsd
  3. Oct 13, 2006 #2

    HallsofIvy

    User Avatar
    Staff Emeritus
    Science Advisor

    If you are imagining a photon as being a little "sphere", that's an error. At the quantum level, such things as shape and size have no meaning.
     
  4. Oct 13, 2006 #3
    Not at all.

    If we can assign a scale length (wavelength) in one direction, shouldn't there be others for the transverse dimensions - otherwise photons would be purely 1 dimensional linear objects?
     
  5. Oct 13, 2006 #4

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    The reason why you never learned about this while in school is because photons were never defined to have a definite size in space. It is only defined as quanta in energy dimensions. While there is a characteristic length associated with a collection of photons, which is the wavelength, a single photon was never defined to have a size in real space.

    Zz.
     
  6. Oct 13, 2006 #5
    If single photons can interfere with themselves and create double-slit interference results, it would seem that they do indeed have a scale length (wavelength) so this can't be right.
     
  7. Oct 13, 2006 #6

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    But associating a "wavelength" to a photon doesn't mean that the "size" of the photon is that wavelength.

    The problem here is that we are used to the concept of "wavelength", which is a classical wave concept. We then try to adapt that to "single photons", which certainly are not classical waves. Now, is the 2-slit interference a unique feature of "waves", even when you shoot one photon at a time? I can show you the Feynman path integral approach that would invoke no waves. Even in Marcella's paper where he derieved the single, double, and multi slit interference using purely QM and not classical wave, you'd be hard pressed to come up with a "wavelength" associated with a classical wave (the QM wavefunction isn't it).

    So no, there still isn't a "size" of a photon. Even if you use the interference phenomenon, you'll never find any books or papers claiming that there's anything associated with the size of a photon. This is why you didn't read about it while in school.

    Zz.
     
  8. Oct 13, 2006 #7
    To argue there is no size to photons and that wavelengths are just abstract properties would seem to totally undermine arguments explaining interference fringes. How could there be interference if there is no size?
     
  9. Oct 13, 2006 #8

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    I never said they were "abstract properties", whatever that is. You did. I said they came from archaic description of light based on the classical wave property. It got carried over into the photon picture.

    You can easily falsify what I just said by pointing out valid physics papers that have made claims about photon sizes.

    BTW, where, even in wave decription, is "size" of the wave is required to describe interference? The slit size has to be comparable to the wavelength for a clear interference pattern, sure. But this has nothing to do with the size of "light". That concept is undefined in wave theory. I could get the same interference pattern for the current in SQUIDs devices. What "size" do we consider there now?

    Zz.
     
  10. Oct 13, 2006 #9
    Indeed. But if the slit size has to be comparable to wavelength, then maybe single-slit and double-slit experiments are telling us something about the transverse dimensions of photons? While this is not the perspective from which we learn about these experiments, nor is it the intent with with they are conducted, I wonder if there's something useful here to be learned if they were analyzed from this approach also?
     
  11. Oct 13, 2006 #10

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    No. I've described several times on here that the "single slit" diffraction is easily explained via the HUP.

    Do a search on here for "marcella". He has a terrific paper on Eur. J. Phys. that derived ALL of the interference pattern without having to resort to any classical wave picture. Out of it, the HUP falls right onto your lap.

    If you want to speculate that such a thing has anything to do with a "size" of a photon, then there's nothing to stop you, except that you have to at least realize that you are making nothing more than speculation that isn't shared by professionals in this field. You just have one heck of an explanation to provide on why the slit size can easily be SMALLER and smaller and smaller without somehow squeezing out the individual photon. How does this reconcile with it having a "size"?

    Zz.
     
  12. Oct 13, 2006 #11
    Perhaps buckyball interference, in the limit of small slits, would provide insight. I was surprised (initially) that interference can be seen even when the wavelength is smaller than the diameter.
     
  13. Oct 13, 2006 #12
    There are lots of references to, but no easily accessible sources, of the Marcella paper. But to a large extent, I'd probably consider it irrelevant unless it provided some terrific new insight into quantum phenomena. QM's curse seems to be that it has great and very precise equations that no one understands. Yet another formalism or path to the same results doesn't address my interests unless it casts some light onto the basics.

    I think while physicists use an appropriate level of rigorous control of words and concepts when dealing with very tricky fundamentals they sometimes hide or retreat into these, and hence have a kind of dogmatism to shut out questions that are quite rational and reasonable that they feel uncomfortable with.

    As for the single slit diffraction case, since the FWHM respond to the slit width, this is consistent with the photon being distorted or affected and hence having a size. The elephant in the room here is that interference and diffraction strongly suggest photons have transverse extent and therefore I don't see it as being unreasonable to ask about it.
     
    Last edited: Oct 13, 2006
  14. Oct 13, 2006 #13

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    But there is a difference here between asking and INSISTING that what you asked makes sense. I can ask you, for example, for the flavor of blue, or the sharpness of charge. Just because I can put together a valid sentence to ask a question, doesn't mean the question has any degree of validity.

    You are asking for the property of something in which that thing were never defined with in the first place. Yet, you still insist that there has to be an answer. So tell me what is the degree of happiness of a neutrino?

    Zz.
     
  15. Oct 13, 2006 #14
    I heard neutrinos were loners.
     
  16. Oct 14, 2006 #15
    So you consider a question concerning fundamental physical properties (the size/scale-lengths) of photons as being comparable to asking about a neutrino's happiness or the flavor of a color. Whatever.
     
  17. Oct 14, 2006 #16

    Office_Shredder

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    a photon is about a quarter of an inch wide

    So what is the spin of your computer?
     
  18. Oct 14, 2006 #17
    .25 inches... Sounds reasonable. Any idea if this is wavelength dependent?

    As for my computer: S1 = 7200rpm for the disk, s2 = 600rpm for the fan - a guess, and Sz is 23mins 56 secs.
     
  19. Oct 14, 2006 #18
    i think the question is not really unreasonable.... I guess what padraighaz means is the "localization" of photon. How can I know where a photon is approximately at with certain probability? and how "wide" would this probability density be? i mean what is the shape of this distribution, is it like a normal curve or what? from what I know, in order to "localize" a light wave, one would need a fourier summations of varies wave functions with different frequencies... so does the probability density of "one photon" exists? or is it non-sense to ask such question since photons can never be consisted of one frequency solely?
     
    Last edited: Oct 14, 2006
  20. Oct 15, 2006 #19

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    If this is true, then it contradicts experimental observations. We DO have "single photon sources". However, if such a thing is simply a "wavepacket" consisting of a fourier sum of a series of wavelength, then it is no longer a monochromatic "photon", and thus, has no single, characteristic wavelength. This means that using a single, unique wavelength to describe a photon is meaningless. Not only that, applying such a thing to something like the photoelectric effect would be wrong!

    In other words, in trying to offer the possibility of a photon being defined to have a 'size', something that it was never defined to have, you have thrown out other verifiable experimental evidence.

    Zz.
     
  21. Oct 15, 2006 #20
    How does the photon having a size contradict experimental evidence?
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?



Similar Discussions: How wide is a photon?
Loading...