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

Delayed choice quantum eraser experiment

  1. May 25, 2009 #1
    I would like to ask some questions regarding "delayed choice quantum eraser experiment". (I think it is possible for those who are familiar with this experiment to skip the text and go ahead into questions 1 and 2.) It is mentioned e.g. on these links:
    Let's talk about picture depicted in http://strangepaths.com/wp-content/uploads/2007/03/qe.png (from the first link). To briefly describe the experiment:
    There are two slits A and B from which (from A or B) two entangled photons are emitted. One goes into detector D0 and the other (its counterpart) goes into one of the detectors D1, D2, D3, or D4. If it comes into the D3 or D4 detector, we can deduce what slit (A or B) it came from - on the other hand when it comes into D1 or D2, we cannot.
    Now comes interesting part: photons detected on D0 don't exhibit any interference pattern when watching them altogether (ie. for their counterparts detected on D1, D2, D3 and D4 altogether). The same is true when watching only photons detected on D0 whose counterparts were detected only on D3 (or only on D4). But when watching photons detected on D0 whose counterparts were detected only on D1 (or only on D2) we see interference pattern on D0.
    But the most interesting thing is that the above is true even if the distance between A(or B) and D0 is smaller than distance between A(or B) and D1(D2, D3 or D4) - and even smaller than distance between A(or B) and BSA(or BSB) - see picture mentioned above. Suppose for the next text that this distance assumption is true.
    These are the facts (or am I wrong?) But it seems it's really strange. My questions are:

    1) Is the same as above (especially presence of interference pattern) true when we will send photons less frequently so that the next one will be send after the previous hits one of the D1,D2,D3,D4 detectors (and of course its counterpart hits D0)?

    2) (Key question of this topic) What if (in case point 1 is true) the beam splitter BS (see picture mentioned above) is present at the beginning of emiting one photon though A or B slit and will be removed just after the photon has reached D0 detector (ie. before its counterpart hits BS - or even before its counterpart hits BSA(BSB) but the latter is not so important). Will we also notice the interference pattern on D0 detector only for those photons whose counterparts have reached only detector D1 (or only D2)? (Before emisson of the next photon, the beam splitter BS will be installed on its original place again and so on...)

    Thank you very much for answering. I hope I have explained my questions clearly.

  2. jcsd
  3. May 26, 2009 #2


    User Avatar
    Science Advisor
    Gold Member

    1. You can slow down the rate of photons, and you will still see the interference. PDC photons are produced relatively sporadically anyway - on an order of hundreds per second depending on the setup.

    2. If you remove the BS as the photon flies by so, then you will know the which-path information completely. No interference will be seen once you correlate. That is true even if the detection event at D0 has already occurred.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook