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

Thought Experiment

  1. Dec 7, 2005 #1
    I have come accross this question in another forum and thought to post it here, hopefully will get some interesting views.
    This is a thought experiment so the setting might not be achivable with todays technology and knowledge, instead try to look at the implications of this set up rather then the possibility of achiving such set up.
    Imagine if we have a enclosed vacuum space, nothing can go in or out. We keep adding photons inside the enclosed space. (assume that whatever material or field we are using to enclose it will not absorb the photons, as well as the material used to generate the photons will not absorbe them). What would happen if we keep adding photons? What would happen if we increase the density of the photons in an enclosed vacuum space? Would they interact with each other? would they form some other form of particle that can carry more energy? What would happen?
    Thanks to all in advance.
  2. jcsd
  3. Dec 7, 2005 #2


    User Avatar
    Science Advisor
    Homework Helper

    If nothing can go in or out then how would you add photons?
  4. Dec 7, 2005 #3
    I think the closest I can come to answering your question is this:

    If you have a space completely enclosed by perfect mirrors (only reflect light, no absorbtion), no the photons won't interact. Assuming we are looking at this purely from the particle portion of the duality, the photons will just sort of bounce around forever.
  5. Dec 7, 2005 #4
    Again do not look at how can we achive to do this experiment. I know that this is not possible to achieve today. I am more interested in the implication of this thought experiment.
    TIDE: Assume the photon source is inside of the enclosed space.
    KingNothing: Would they bounce forever if the density of photons would be very high and keep increasing?
  6. Dec 7, 2005 #5
    Yes. They do not take up any space, hence why they are called "point particles". As such, density does not impose any limits on them.
  7. Dec 7, 2005 #6
    True that density does not impose any limits, but because of large number of photons there might be some interference between each other.
    Since photon is defined as wave and particle and in both concepts interference is a possibility that might as well happen in this case. But if so what kind of interference we might expect? What changes can we expect?
  8. Dec 8, 2005 #7


    User Avatar
    Gold Member

    This is a 'had an awful lot of beers, off the top of my head' response. I would think that eventually the photonic pressure would become large enough to rupture the enclosure.
  9. Dec 8, 2005 #8
    Anyone else? I thought that this might spur some wild responses and speculations. Thanks
  10. Dec 8, 2005 #9
    it would matter, if the photons are particles or waves, it particle you could fill the box , but if there wave you can never fill the box,
  11. Dec 8, 2005 #10


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Think again. I have a standing wave cavity that I fill every day. The RF that I send in there has a filling time, etc. with very little, if any, leaking out.

  12. Dec 8, 2005 #11
    when you say fill time, are you talking about the time it takes to excite the electrons? or are you saying that you are filling it with photons.
  13. Dec 8, 2005 #12


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    You do know what "RF" is, don't you? How do you think particle accelerators work? They accelerate electrons using EM field (i.e. photons) using such cavities. You design such cavities so that it can sustain the mode of the RF frequency that you are using.

  14. Dec 8, 2005 #13


    User Avatar
    Staff Emeritus
    Science Advisor

    One thing that's worthy of note. There is a critical field strength (volts/meter) at which one will start to create particles out of the vacuum, known as the Schwinger critical field.

    So if one set up an RF cavity (or any other method of putting photons in a box) and got the required volts/meter (about 10^18) - (I think) theoretically one would start to create electron-positron pairs.

    Google hasn't found much elementary discussion of the Schwinger critical field, one paper that talks about it a bit is


    Getting to 10^18 volt meter is left as an exercise for the student - actually I've seen papers arguing that we might be able to do this with lasers. IMO a box just complicates the problem with phenomenon such as field-aided emission, where the field (even below the Schwinger limit) would start tear electrons out of the conductors of the box via quantum tunnelling (assuming a resonant cavity to conatin the photons, which requires conducting walls).

    The only reason to use a box is that's what the original poster asked about.

    Issues of breakdown of boxes with high electrical fields brings up the interesting semi-related question of just how much voltage a vacuum capacitor can have without breaking down.

    Eventually I'd expect electrons to be torn from the plates via field-aided emission, and the high energy electrons hitting the plates would knock lose other electrons, causing avalanche breakdown, even in a perfect vacuum.

    I haven't seen anything numerical about this, though (i.e. limits on field strength in a vacuum capacitor). High work-functions for the electrodes would help in minimizing field-aided emission, and of course you need very smooth plates to keep the field strength down (points generate very high fields).
  15. Dec 8, 2005 #14


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    There are problems with this, pervect.

    1. Copper cavities, which are the most common material used, consistently break down at around 35 MV/m. It is why I get employed in my present field of study - people want a better acceleration mechanism rather than using copper LINAC. :) The superconducting cavity of Cornell/DESY made of Nb and Nb alloy are steadily getting to that level. So unless we make major technical progress, I don't think we'll be getting to the gradient you mention anytime soon.

    2. Field emission is quite well-known in such cavities - they are the source of dark currents that we detect all the time.

    3. Breakdown or arcing, on the other hand, is still a highly debated issue. The "consensus" so far is that while it can be initiated via field emission, this isn't the sole mechanism for breakdown. And even when it is initiated by field emission, it isn't what one typically thinks. What happens is that protrusions, having a higher field enhancement factor, will become extremely good field emitters. However, due to such small area of these emitters, there is a very powerful resistive heating going on even when the current is small. This leads to the melting of the material and POOF! There's the breakdown.

  16. Dec 8, 2005 #15
    The description of the photons as wave or particle is of little interest in this experiment. You can use one or the other or both to sustain an argument. The important questions raised; is there a critical point where:
    1. No longer can add photons (able to fill the box).
    2. Photons change properties as we know them and take some other form of energy carrier due to interaction between them.
    3. Explosion!!
    4. Observe previously unknown phenomena!!!
    5. Other speculations!!!!

    PS: Do not concern yourselfs with the how to set up such experiment, or whether it is posible, it is not part of the question raised.

    Thanks again
  17. Dec 8, 2005 #16
    I would think they won't interact the same manner that electrons will because bosons don't obey the pauli-exclusion principle, so as many of them can go in one point as possible.
  18. Dec 8, 2005 #17


    User Avatar
    Staff Emeritus
    Science Advisor

    I thought I'd already mostly answered this?

    1) No. (but see 2b).

    2a) Yes - there is a small photon-photon cross section in QED. The photons do not "change" propeties, we simply observe the effects of the photon-photon interactions. An example of this sort of photon-photon interaction is the creation of e+ e- (electron/positron) pairs out of the vacuum. This is what the point of talking about the "Schwinger critical field" was. I'm sorry that I don't have a better reference for you, if you google for more information on this you will do yourself a service in answering your own question. If you put enough photons in a box, you'll exceed the Schwinger critical field, and you'll start to see electron-positron pair production in the box.

    [note: I know enough about QED to know that this happens, but not enough to do detailed calculations. You might do some reading and ask for more information in the quantum mechanics forum if you are still interested.]

    As Zapper-Z notes, this is totally impractical to do _in a box_. It's not so impractical to do this _without_ the box, simply by creating a high enough intensity laser beam, though it's still quite ambitious.

    Look at the following abstract, for instance


    2b) Even more exotically, eventually your box will be massive enough to form a black hole. (See the related thread in the GR forum). Of course you can still add photons to the resulting black hole.

    3) Yes - but this is due to mundane failure of the box, not to anything exotic. As you add photons, the pressure increases. Eventually, any box will explode with a high enough pressure. (But this is a realistic concern, which you were not concerned with?)

    4) Unknown phenomenon are unknown, how can we talk about them?

    5) I'd rather not speculate - that's not what PF is here for, it's to explain known science, not to speculate.
    Last edited: Dec 8, 2005
  19. Dec 8, 2005 #18


    User Avatar
    Staff Emeritus
    Science Advisor

    Cool, I've never seen a figure on this before. Thanks.

    So, if we were able to improve performance of cavities by only 11 orders of magnitude, we'd have to start to worry about the Schwinger critical limit? :-)
    Last edited: Dec 8, 2005
  20. Dec 8, 2005 #19
    I may not be interputing the question correctly, so i dont know what asumsion you want to include, so just thoughts,
    1. if photons are particle and have physical dimensions at some point any space would fill. but if the photon are waves of energy then you could continue to add enegy until you over power the box.
    2. there is the posiblity that the photon could combine with other particle (neutrino) to create electrons.
    3. explode - any confined energy can explode
    4. you could find the photon are not electro-magnetic energy, but it own type of energy, and because we can only measure photon using the charges of electrons, we call them electro-magnetic.
    Last edited: Dec 8, 2005
  21. Dec 9, 2005 #20


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Exactly! Now if only you can convince Mother Nature of that.


    The group that I work in is hoping to show a gradient of 100 MV/m for a realistic electron bunch of 1 nC. So cross your fingers and toes.

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

Have something to add?

Similar Discussions: Thought Experiment
  1. A Thought Experiment (Replies: 125)

  2. Thought Experiments (Replies: 16)

  3. Thought experiment (Replies: 3)

  4. Thought Experiment (Replies: 9)

  5. Thought Experiment (Replies: 13)