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

FTL-gedanken experiment

  1. Jul 3, 2011 #1
    FTL-gedanken experiment.

    I never really got feedback on my last proposal in 'Communication systems and entanglement'- so I try here with a slightly modified version:

    A source produces entangled photons against Alice: p beam - and against Bob: s beam.
    These are used to transmit from Alice to Bob.

    Setting T (0): Polarized beam splitter PBS (v) followed by two detectors. PBS (v) forcing the photons to choose between being polarized in the direction v degrees or perpendicular = direction v+90. T (0) maintaining an agreed time and should create an interference pattern at Bob's receiver.

    Setting T (1): Polarized beam splitter PBS (v +45), followed by two detectors. PBS (v +45) forces the photons to choose between being polarized in the direction v +45 degrees or in the direction v-45. T (1) maintains the same scheduled time and should not create an interference pattern with Bob.

    Starts with a PBS (v+90). The transmitted beam encounters a device to read any interference pattern - a double slit or (if the double slit is problematic), an interferometer (for example Mach Zehnder with BS = a half silvered mirror). The reflected beam is stopped by a detector.
    The distance between the source and Alice's detectors are less than the distance from the source and into the beginning of the receiver so that the photons will be measured at Alice place before they reach the receivers PBS (v+90).

    T (0): p-photon v degrees so is the corresponding s-photon perpendicular = v+90. All of these are transmitted by PBS (v+90) and would like to form an interference pattern.
    p-photon v+90  s-photon v, ie. reflected by the PBS (v+90) and detected.
    Together, the system works here as a 'half Coincidence counter': Of the 'entangled' only the desired reach the double slit / interferometer and can form an interference pattern.
    Noise will not be stopped. But since this is a gedanken experiment imagined the noise to be minimal.

    T (1): p-photon v+45 degrees, so is the corresponding s-photon perpendicular = v-45. Half of those are transmitted by PBS (v+90) and would like to form an interference pattern.
    p-photon v-45 degrees then the corresponding s-photon perpendicular = v+45. Half of those are transmitted by PBS (v+90) and would like to form an interference pattern.
    Because of reflection should be a half-wave difference between p: v+45 and p: v-45, so the two patterns are shifted half-wave - as - for example, a fringe pattern and an anti-fringe pattern. Together equalize each other.

    Example: Walborn et al: Double-slit quantum eraser.
    With Quarter Wave Plates and polarizer set to theta =v.
    FIG 4 shows a fringe pattern.
  2. jcsd
  3. Jul 3, 2011 #2


    User Avatar
    Science Advisor
    Gold Member

    And I keep telling you that:

    a) Entangled photons do not produce interference patterns, and I provided a reference from Zeilinger on that point. It was pointed out that entangled photons behave as incoherent light, and so the double slit pattern will not appear.

    b) The pattern never changes, even when you detect a photon as a wave; you only see an interference pattern when you perform coincidence counting.

    c) The ordering of measurements on entangled photons is always immaterial to the results.
  4. Jul 3, 2011 #3
    >>a) But how / why get Walborn then interference?

    And this is only a problem for a double slit – not an interferometer.

    >>b) Of course there is a problem with noise - and that only a fraction – for example 1 % - of s-photons reaching the double slit - but what else?

    >>c) Maybe - but I prefer a traditional sequence – and maybe it's because of the coincidence counter.
  5. Jul 3, 2011 #4


    User Avatar
    Science Advisor

    Two-photon interference is something different than single-photon interference. The commonly known interference patterns from the double slit or Michelson interferometer are single photon interferences. They require a certain coherence time/length. Two photon interferences are what is seen in experiments on entanglement and show up due to properties of the entangled two-photon state. Therefore you always need coincidence counting to see them and they show up only in the coincidence counts. Using coincidence counting is not just a matter of filtering out noise in these experiments. Note that single- and two-photon interference are complementary. You cannot have both at the same time.
  6. Jul 3, 2011 #5


    User Avatar
    Science Advisor
    Gold Member

    a) Because they use coincidence counting. And there is no double slit pattern.

    b) Noise has nothing to do with it. In a perfect situation where all photons are perfectly entangled, nothing ever changes.

    c) I appreciate that you prefer a certain ordering. And yet, ordering makes no difference. Ever.
  7. Jul 4, 2011 #6
    additional related DrChinese >>c) The ordering of measurements on entangled photons is always immaterial to the results.<<

    In two photon interference experiments - for example Zeilinger p 290; Walborn; Kim - meetings, s-photons a double slit before it or its p-partner has been measured. Interference / non-interference are therefore part of the photon-twins shared history before they are measured. (Further: if the meeting with a double slit temporarily suspend entanglement(?) - as a PBS does - there is no temporal surprise at all, that the results are independent of the measured order.)

    Because I measure the p-photons before the s-photons reach an interference-device - and interference is not a part of the common history - there are probably talking about single-photon interference in my gedanken experiment.
  8. Jul 4, 2011 #7


    User Avatar
    Science Advisor

    This is not true. It does not matter whether photons in one arm meet a double slit or MZ-interferometer or something similar in one arm before or after the photons in the other arm are detected.

    As I said before, single- and two-photon interference are complementary (see Abouraddy et al., Phys. Rev. A 63, 063803 (2001)). So that means if you are indeed talking about single photon interference, you have already broken entanglement.
  9. Jul 4, 2011 #8
    Do you have an example? - In the 3 cases I have mentioned are both photons measured after double slit - but alternately relative to each other.
  10. Jul 4, 2011 #9


    User Avatar
    Science Advisor
    Gold Member

    Your Walborn reference shows this clearly, so what is the question? That is what delayed choice experiments demonstrate: ordering is irrelevant, exactly as predicted.
  11. Jul 4, 2011 #10


    User Avatar
    Science Advisor
    Gold Member

    Here is the arxiv version:

  12. Jul 5, 2011 #11
    Thanks for the link.

    re Walborn: ‘Experimental setup and procedure: … The double-slit and the quarter-wave plates are placed in path s, 42 cm from the BBO crystal. Detectors Ds and Dp are located 125 cm and 98 cm from the BBO crystal …
    The delayed erasure setup is similar, with two changes: (i) detector Dp amd POL1 were placed at a new distance of 2 meters from BBO crystal …’

    So there is thus no question of a measurement where Dp is closer than the double slit – as in my experiment.

    (One way to understand this Walborn experiment is that the meeting with the double slit locks relationship between s and p with regard to interference and polarization - and therefore it does not matter when you measure them, as it only further happens that through Dp selects a particular subset to consider.)
  13. Jul 5, 2011 #12


    User Avatar
    Science Advisor

    This has been mentioned before, but it bears repeating: for a given experimental setup, the order in which the entangled photons pass the different optical elements, including the double slit (with or without QWP's) and the polarizer in the Dp branch are completely irrelevant to the results of the experiment. You can place them wherever you like along the beampaths, so the photons encounter them in any order ... nothing about the experimental results (i.e. the coincidence statistics and the one-photon measurements) will change one little bit. That is why we say that for such experiments, the results depend on the entire context of the experiment.

    So, for your example, it doesn't matter if the s-photon hits the double slit first, or the p-photon hits the polarizer first, the selecting of particular subsets in the coincidence measurements works out to give precisely the same results.
  14. Jul 6, 2011 #13
    Why I think the order may be important in some types of experiments:

    S-photons encounter a PBS (0) - polarize horizontally / vertically (and the vertically are detected).
    P-photon encounters a PBS (45) - polarization diagonal positive / negative (and both the diagonal positive and negative are detected).
    Both interrupts entanglement and both causes in addition a difference of half a wave between the transmitted and reflected.

    If s first meetings PBS (0): It will transmit Beam with roughly the same wavelength shift as before for all.

    If p first meetings PBS (45): s-beam will be oriented diagonally negative / positive - with a difference of half a wave - and when this beam subsequent meetings PBS (0): half of each type will be transmitted - so this time the resulting beam consists of a fifty-fifty blend with a half wave difference.
  15. Jul 6, 2011 #14
    As far as I understand it compares 'Abouraddy et al' picture all forms (entangled and noise) with the image the entangled form alone - ie noise filtered off through coincidence counter.
    The surprising result is that although the interference of the entangled alone increases so can the complete picture show less interference.
    This must surely assume that the noise constitutes a considerable part. If virtually all were entangled could this difference probably does not occur? OR??
  16. Jul 8, 2011 #15
    I have a problem with Zeilinger p. 290 - Fig. 2 + Fig. 3.

    Is the experiment in Fig. 3 only a gedanken experiment - or is it done of B. Dopfer in 1998?
    I have not been able to find online a more accurate description of it - not even at Zeilinger's website.
    When I want to look at it, it is because the allegation in Fig. 2: ’The beams of particle 1 then pass a double-slit assembly. Because of the perfect correlation of the two particles particle 2 can serve to find out which slit particle 1 passed […]’

    I have hitherto understood that for a photon to interfere with itself should it pass both slits - and not just the one.
    It seems that this double slit must be very close to the source and with long distance between the two slit?

    (PS - I understand Which Path information in the direction of that someone disturbs a path (more or less) and thereby disappears / (decrease) interference.)
  17. Jul 8, 2011 #16


    User Avatar
    Science Advisor
    Gold Member

    The distance to the slits is not really a factor, as you can route the beam wherever you like. The issue is that Alice COULD learn which path info - even if you aren't trying to - and that means Bob must act accordingly. And vice versa. And again, these photons are incoherent but I must admit I don't know all the rules on that.

    Perhaps someone else can help out on this point?
  18. Jul 9, 2011 #17
    Zeilinger Fig 2.: "Particle 1 is either emitted into beams a or a' "

    so again: 'I have hitherto understood that for a photon to interfere with itself should it pass both slits - and not just the one.' ??
  19. Jul 9, 2011 #18


    User Avatar
    Science Advisor

    Yup .. and that's why there is no interference pattern in the one photon measurements for entangled photons.
  20. Jul 10, 2011 #19
    Ok - so to get right around 'Zeilinger':
    I put a lens in front of the transmitter - PBS + detectors - (so the PBS matches the lens focal plane).
    And one problem less.
  21. Jul 11, 2011 #20
    I found Dopfer (Zeilinger) on the German-speaking part of the network:

    A quick look at the 'figures' gives the following:

    p 36 - Fig. 4.5: interference by 'virtually infinitely distant source' = max interference.

    p.37-Fig. 4.6: interference by 'virtually point source' very close to the double slit = no interference.

    P 86 - Table 4.7: shows that even with a large deviation from the 'ideal infinitely distant source' and quite close to the 'punctate' more than 90% of the interference contrast are preserved.

    This last may explain why, for example, Walborn achieves interference in his attempts.

    - And should it be necessary, one can correct it with a lens.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook