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

How is light experimentally proven to be a particle?

  1. May 1, 2007 #1
    I had an argument with my Physics teacher. I know that light has been clearly proven to be a wave, because of the diffraction experiment with the slits. BUT my teacher kept arguing that it is a particle. But his proof states the Photoelectric effect...

    OK. I searched the internet but to no avail... My previous understanding of the Photoelectric Effect is that, when radiation (i.e. light), it excites the electrons, By transferring energy to the electron, thus the electrons have enough energy to escape. But how does this show that it is a particle?
  2. jcsd
  3. May 2, 2007 #2
    In simple point of fact, it doesn't show any such thing. The strongest statement one can confidently make based on the photoelectric effect is that the transfer of energy between light and the electrons in a metal is quantized. That is light of frequency [tex]\nu[/tex] can only transfer energy to the metal in integer multiples of [tex]h\nu[/tex]. The simplest interpretation of this is that light consists of particles, each of which has energy [tex]h\nu[/tex]. However, there are other interpretations of this same result.

    Just for example, it is possible to recover all of the classic characteristics of the photoelectric effect with a fully quantum mechanical metal and fully classical light. See, for example, Lamb and Scully, "The Photoelectric Effect Without Photons," Polarisation, Matiere et Rayonnement (Presses Univ. De France, Paris 1969), pp. 363-9. (This may be a bit difficult to find, but, as I recall, it's a rather clear treatment.)

    However, there are other phenomena that cannot be explained so easily with a semi-classical theory. In fact, if one takes a photoelectric effect experiment using light from a high efficiency diode, driven by a constant current and also detected with high efficiency and analyzes the noise in the resulting photocurrent, one may find that the noise is smaller than the minimum allowed by the semiclassical theory.

    The idea here is that you get less noise (that is, random variations in the resulting current) if the light source can generate one photon for each electron that passes through it than you could if the light emitted had to be classical waves.

    There are, of course, a wide variety of other experiments which demonstrate the same fact about light.

    And, since it turns out that light actually does have a particle nature, there's no real need to consider other interpretations of the photoelectric effect.
  4. May 2, 2007 #3
    The day-to-day proof in many laboratories is "photon counting".
    When intensity is reduced, only individual photons can be observed.
    Many devices can be used in photon-counting.
    Photomultipliers are the most common.
  5. May 2, 2007 #4


    User Avatar

    first of all, the OP's teacher is wrong in saying that light is only particles. light has some wave-like properties and also has particle-like properties. now we know that matter also has both wave-like and particle-like properties.

    E&M radiation (of which visible light is) is neither exclusively waves nor exclusively particles. EM radiation is both, but manifests itself as waves in some contexts and as particles in other contexts. same for subatomic particles.

    not very reasonable interpretations.

    keeping the frequency, [itex]\nu[/itex], (or color) of the light fixed, but reducing the intensity (watts/m2), to an arbitrarily low level, you will still have electrons emitted with energy [itex]h \nu[/itex] minus some "work function" that is the energy required to get them airborne. you cannot reasonably account for why, given the cross-section area of any particular atom on the surface of the metal, it can store up enough energy, in the short period of time the light is turned onto the metal surface, to launch an electron with energy [itex]h \nu[/itex]. why, in a homogeneous metal surface, would the energy impinged upon many atoms somehow be transfered to a single atom in the neighborhood, so that it would have [itex]h \nu[/itex] amount of energy and launch an electron? only if the energy was "packeted" in these quanta of energy, can such a thing reasonably happen.
  6. May 2, 2007 #5
    So in a very simplistic view, the idea behind the Photoelectric effect is that, light has a rapid, if not instantaneous transfer of energy to the electrons when hitting them - is it like an elastic collision between 2 objects? Thus it is considered to be like a particle??????
  7. May 2, 2007 #6
    If your teacher showed you something and claims that it is a proof then you are gaurenteed that he made an invalid claim. Observations do not and can not prove an hypothesis. E.g. one can't prove that light consists of particles. We do numerous experiments to check this idea out and when we have exhausted our knowledge on how to verify that light is a particle then some scietists incorrectly say that they have proved something.

    For a comprehensive discussion on this see Rohrlich's treatment on the subject at


    Scroll down and then click on "Philosophy of science." I'm sure you'll find Rohrlich's treatment to be very well stated.

    I have a journal article in my possesion which is authored by Dr. Willis Lamb. Dr. Lamb won the Nobel prize in physics for his contributions to the theory of quantum electrodynamics. In this article Lamb walks you through various experiments that have been done and explains why each one does not prove that light consists of partilcles.

    I can e-mail you this article if you'd like?

    Last edited: May 2, 2007
  8. May 2, 2007 #7
    In fact, the rapid photoelectron emission is predicted by models that treat the metal as quantum mechanical but the light as classical waves. In this case, you could think of it arising out of the energy-time uncertainty in the the metal.

    That said, such semiclassical models turn out to be much harder to work with than the photon model and, since the quantization of light can be demonstrated by other means, turns out to be unnecessary. But, the point still stands that the photoelectric effect alone is not enough to unambiguously demonstrate the quantum nature of light.
  9. May 2, 2007 #8
    The point of photoelectric emission and its 'proof' of quantised light is that monochromatic light below a certain energy never liberates electrons. you have to go over a threshold of energy to do anything at all to them. No variation of a beam of light of longer wavelength will liberate electrons; it can only occur witha light of a certain wavelength - which implies that the process is not due to the power of the beam but the mode by which the energy is transferred - discretely.

    Basically, photons are not particles and they are not waves. They are quanta. Quanta are neither but show properties of both.
  10. May 2, 2007 #9


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    The classical and semi-classical picture do not have the ability to account for the which-way experiment and the photon anti-bunching experiment.[1]

    Furthermore, and this has been talked about ad nauseum, beyond just the photoelectric effect, if you go deeper, the intricate details that one get out of (i) resonant photoemission (ii) angle-resolved photoemission and (iii) multiphoton photoemission so far can only be described via the photon picture. No classical or semiclassical anything have been offered to explain all of them as of now.


    1. H. Paul, Rev. Mod. Phys. v.54, p.1061 (1982).
    Last edited: May 2, 2007
  11. May 2, 2007 #10


    User Avatar
    Science Advisor

    Let's see: Compton Scattering; photoproduction of most anything, radiative corrections to scattering of charged particles, ... Photons play a substantial role in particle physics, in which photons are considered to be particles -- it's all in the kinematics. The point is: in particle physics, the best game in town is to assume, that is ASSUME, that photons are particles. the assumption seems to work quite nicely.

    But, forget about photons when dealing with, say, radar. Recall that the quantum field generated by a classical current is composed of coherent states -- which give a Poisson probability distribution, hence the classical current's field can have zero to an infinity of photons. So, it's generally easier to go with standard E&M fields, that is, waves.

    It all depends.
    Reilly Atkinson
    Last edited: May 3, 2007
  12. May 2, 2007 #11
    I'm very interested in.
  13. May 2, 2007 #12
    Me too, I'd love to see this paper. Can you put a link here? or email me as well..


  14. May 2, 2007 #13
    A reference would be good, Pete. o:)
  15. May 3, 2007 #14
    Sorry for the late reply again. Anyway, I talked to the HOD of Physics. He said practically what is being said here. Except, he did state though that light almost definitely; can be said to be a wave (or travels as a wave). Sojourner01 states that light is neither. Can I have a renconsiled statement?
  16. May 3, 2007 #15
    I think a better way to frame it up is for everyone to repeat after me: "We Do Not Know What Light IS!".

    We might think we understand a lot about light but the fact is we know nothing of it. We know about reactions like a martian would know about cars from a distance. They see them moving around, following paths and such. They see how they bounce off each other and so forth. But from Mars they couldn't possible know what they really are, or what their purpose is.

    We understand how light moves, how it goes through objects, how it reflects off to give us colors, how its absorbed to give us heat and energy. However, we have no clue what it actually is, how it physically moves from one spot in space to another, and most importantly, whether its a wave or a particle. In my opinion, it can't be both. But there I go, giving another opinion.

  17. May 3, 2007 #16


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    This is as silly as saying no one understands QM.

    Question: How often do you depend your life and the lives of your loved ones on something that no one understands? For example, no one understands about telekinesis. Practically NOTHING about it has been known or accepted as science. Would you put your life on it? Would you use it to treat a serious disease? Would you use it to save you as you jump off a building?

    We know A LOT about light. When you know something well enough to USE it, it means we have a good understanding of what it is. And if you look closely enough, everything around you that you think you know about, really means that you know the PROPERTY of what it is. You simply confuse that with thinking that you know what it is, when it actuality, all you have is simply an awareness of the various properties and behavior of that object. The SAME goes with light. We know A LOT of its properties, which means that you cannot say we know nothing of it! All those synchrotron centers all over the world is my evidence for falsifying what you just claimed here.

    Please do not hijack this thread into some philosophical discussion about what we mean to "know" about something. The OP asked for specific evidence particle-like behavior. Unless you have some, do not railroad this thread into what it isn't supposed to be.

  18. May 3, 2007 #17
    lol, sorry, wasn't hijacking anything. Was responding to another poster that was still asking the same question because no one had answered the question of the OP yet. Not really sure why you're jumping me... I never claimed to understand anything. I'm only quoting what I'm reading in this forum. You guys all argue amongst yourselves about these concepts. I never said that we knew nothing of light. In fact I said we knew a lot about its properties but not about what it physically is. Therefore for anyone to tell me they "Understand" how light moves is not being totally honest. If we knew how exactly it moved then we'd have no more need for further experiments.

    Again, I'm not trying to get philosophical about anything. Its QM that gets that going because no one can properly explain it and the explanations end up in philosophical debates.

    Just as an example, I read a post last night that someone had posted about spin and thought that I was finally starting to understand it, only to be given another post stating that those tests were not conclusive and had problems. Its a never ending circle. And again, I'm only responding to those claiming understanding as if its there to have. I never claimed anything so not sure why you want to jump on me...

  19. May 3, 2007 #18


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Really! The way you go around and making "proclamations" about various things in physics, you could have fooled me!

    You should also stop mimicking what you read here especially when you have not understood the physics behind it. For someone who is so insistent on urging people to "understand", you are very quick in grabbing something without understanding it.

    Do not equate your level of "understanding" the physics being talked about here with what physicists do understand.

  20. May 3, 2007 #19
    Yes sir, I'm sorry I offended you. I think if you spent time to actually read my posts you'd see I haven't made any claims. I haven't stated anything as fact and I haven't repeated what someone said as if I understood it.

    I think I've spent more time asking questions than anything. Not sure why you feel the need to get personal. You act as though I'm attacking your personal research or something, just because I don't accept some of the responses I've read to explain things on this forum.

  21. May 3, 2007 #20


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Let's see...

Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?

Similar Discussions: How is light experimentally proven to be a particle?