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

Propagation of photon

  1. Dec 23, 2003 #1
    When a single photon is produced by a point source, and detected by an array of detectors placed on the surface of a sphere around the point source, would only one of the detectors pick up the photon, or would only some of them pick it up, or would all of them pick it up, assuming no extraneous particles or fields are present to change the flow of the photon? I hope not to excite any semantic argument over what a point source might be or what an extraneous field or particle might be, but rather to address the question toward help in understanding the nature of photon, and of what a single photon is.

    Thank you,

  2. jcsd
  3. Dec 23, 2003 #2


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Only one would pick it up.

    - Warren
  4. Dec 24, 2003 #3
    Then if only one photon is produced at a time in some sequence, and only one detector is used to detect it, can we easily select conditions so that the detector collects very nearly every photon produced in a sequence by the point source?

    And if we then select such conditions, and hold them all the same except for the distance of the detector from the point source, (or perhaps the angle of apeture of the detector?) and vary that distance or apeture, is it predicted that a more distant or smaller apeture will detect a smaller number of the total sequential source photons, thus showing a lower probability with distance, perhaps as the inverse square law?

  5. Dec 24, 2003 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    I think the answer to this is yes.
    Consider our photon gun {We call them lasers} and a distant photodetector, we can intercept nearly every photon produced.
    the question is, where is the point source. Since a laser produces a very neary parallel beam the actual point source may be some distance behind the laser. The actual distance could be found if you know the divergence of the laser beam.

    The trouble here is that you said point source and I said laser, I would say that "single photon" and "point source" are contradictory phrases. When an atom emits a photon it is in a random direction. This, of course, can be seen from HUP. We can know the location of the souce atom with great precision, this knowledge of the location of the photon means that we have NO knowledege of its direction. The inverse square law applies ONLY to point sources, since they radiate into a sphere, and implies multiple photons.

    Now the Catch-22 to all of this is that if you get far enough away from ANY source of light it will be seen as a point source. Though if a distant observer of a diverged laser beam were to calculate the the power of the source, assuming a point source his results would be off by orders of magnitude because a laser emits large amounts of energy into a very small solid angle.

    It is not clear the that the second part of your experiment has any meaning. Simply because while you say point source you imply something very different.
  6. Dec 24, 2003 #5
    Ok, we have gotten off into a discussion of terms, as I was afraid was inevitable. I guess I chose to use the point source term because I wanted to avoid the conditions in an atom, which, if I understand correctly, cause light to be propagated randomly in any direction. Also the use of the laser, fascinating as it is, since that involves a huge cascade of photons. I guess I was thinking of something more like the photon pair (traveling in opposite directions, I think?) produced by the collision of a positron and an electron.

    Anyway the function of the idea of a point source in my question is to fix the spatial location of the origin of the photon. Perhaps a laser source could be restricted by an apeture so that only one photon at a time is allowed to pass, and that would suffice, altho I am not familiar with laser technology enough to say if that is a reasonable condition or not.

    What I want (as a thought tool, not necessarily as a benchtop apparatus) is a source that produces photons at a fixed location, one at a time. Then, I want a detector array, perhaps hemispherical rather than spherical if we are using a laser with restricted apeture. The geometric origin of the hemisphere is placed at the source, so that the distance from the source to the detector array surface is a radius. Photons leave the apeture one at a time. Photons arrive at the detector one at a time. Then if I have a source event at a given location in spacetime and a detector event associated by the appropriate time and space interval, can I say that a single photon traversing a given path is responsible for both events? Or have I already violated HUP?


  7. Dec 24, 2003 #6
    No, I think you can't. A 'source event' would mean that a photon is detected. A photon can be detected only if it is absorbed. So a 'source event' would involve at least two photons - one absorbed, and one emitted.
  8. Dec 26, 2003 #7
    This seems amiss to me...I probably do not understand. I imagine one could construct a device that would be expected to emit one photon when energised by some quantum of energy, so I might put a current across a resistor for a measured length of time, and expect a photon to be emitted in return. I know the energy is present, I know the time has gone by, so I can assume the production of a photon?

    Then I can say that in the time in question a photon was emitted, and if my detector detects a photon, I can be confident that the photon emitted in the time and the photon absorbed in the appropriate offset time for the offset distance is the same photon. Please offer any corrections.


Share this great discussion with others via Reddit, Google+, Twitter, or Facebook