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

Entanglement in delayed choice quantum eraser (DCQE)

  1. Jul 15, 2011 #1
    In the delayed choice quantum eraser (DCQE), such as the walborn paper, link below:

    http://arxiv.org/PS_cache/quant-ph/pdf/0106/0106078v1.pdf" [Broken]

    we try to find out the polarization/path via quarter wave plates (see diagram on page 7 of the paper)

    Now does not entanglement break (i.e. wave function collapse) at that point?

    later when we erase which-way info, does the same entanglement again re-join?

    what is the popular explanation among physicist on this?
    Last edited by a moderator: May 5, 2017
  2. jcsd
  3. Jul 15, 2011 #2


    User Avatar
    Science Advisor

    Just putting quarter wave plates in does not constitute a measurement as it is a reversible process. As long as no measurement has been performed on one of the photons, entanglement should not be broken.

    "The opera ain't over till the fat lady sings" holds for some branches of physics, too.
  4. Jul 15, 2011 #3
    thanks Cthugha, well put.

    when has the measurement, in your opinion, been performed (i.e. the fat lady has finally sang..;)):

    1. when the photon has passed through the QWPs
    2. when one of the (twin pair) photon has struck the screen
    3. when both the pairs have stuck the screen
    4. when both the pairs have stuck the screen and have been compared/matched via co-incidence counter

    I guess it would be 2 above and you mean the same.

    if let' say scientists figured a way where the quarter wave plate would flash or something.....when the photon passed through

    would the measurement then have taken place?
    Last edited: Jul 15, 2011
  5. Jul 15, 2011 #4


    User Avatar

    You may think about the experiment that way:

    We count only coincidences with detector D_p clicks (that is Anton Zeilinger's answer to the question: what is photon? Photon is just a click in our photon detector!) and we ignore all clicks in D_s which occur when D_p stay silent. As photons in both branches (p and s) are entangled, we are sure that we measure only those photons, which, at the path BBO-QWPs are linearly polarized, perpendicularily to POL1 axis. It doesn't matter if we reject some clicks because the coincidence hadn't taken place, or because the light had been absorbed in the polarizator polarizing the beam s linearily.

    Thus our setup is equivalent to pretty classical experiment with light bulb as a source (at BBO) followed on its path by polarizator. You don't need any QM to analyze it - high school wave optics is pretty sufficient to find the results.

    If you really want to think about it in terms of Bohr's measurements, wave function collapse, and Bell's mysteries, you should just take that final measurement is the moment when some coincidence logic fills a click into the histogram - (4) of your list. Probably that is the only tricky point about entanglement: you may think about measurement (in Copenhagen meaning) only after the coincidence between both elements of entangled pair is computed.
  6. Jul 15, 2011 #5


    User Avatar
    Science Advisor
    Gold Member

    No. You can perform all kinds of manipulations on entangled particles which do NOT collapse the wave function. The reason is that the manipulation is applied similarly against all possible paths. Collapse only occurs when some paths are selected over others, which doesn't happen with a wave plate.
  7. Jul 15, 2011 #6
    all kinds of manipulations but none that would change/skew the probability of getting a particular spin?

    interesting. thanks for the info DrChinese

    if we were to look with a telescope: would some paths be selected over others?

    if we were to use a polarizer: would some paths be selected over others?
  8. Jul 15, 2011 #7


    User Avatar
    Science Advisor

    Ok, so I have a question related to this. The way I have always understood this is that, when one of the particles interacts with a device like this (the double slit with QWP's, or a PBS), what happens is that the polarization of that particle becomes entangled with the different spatial paths after the device. You can then choose to detect the particle (thereby obtaining which path information), or you can re-combine the paths (as in an MZ interferometer), to recreate the polarization-entangled state.

    My question related to this is, what the correct way to describe the entanglement particle B (assuming it is not interacting with any sort of apparatus), while particle A is spatially entangled with a PBS ( or whatever)? In other words, what is particle B entangled with during this interval? Obviously it must still be entangled with particle A in some fashion, because we can choose a context where the original polarization entangled state is restored. Or is this simply not a meaningful question to ask, without first defining the context of the rest of the experiment?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook