Delayed Choice Quantum Eraser

[heyo]

Okay... this thing has got me tweaked.<br /><br />Some of you may be familiar with the Delayed Choice Quantum Eraser experiment. If not it is a bit much typing for me to explain here, but it can be found on the web. It can also be found in the book Fabric of the Cosmos, Brian Green's new one.<br /><br />As far as I understand, they have only performed this experiment over very small distances, therefore very small time differences, but it does seem as though what happens in the future affects what has happened in the past.<br /><br />Let's make a few changes to the experiment. First off, instead of a random 50 50 beam splitter for the idler photons, lets change that to a switch that you can set to either send the photon one way, where it will retain "which-path" information, or the other way where which-path information would be erased.<br /><br />According to the experiment, even if the path the idler photon travels is LONGER than the path the signal photon takes to the detector, the photon still apparently "knows" as evident by that particular subest of photons hitting the detector forming either an interference pattern, or the expected particle-type blob of photon strikes.<br /><br />So if the switch we add on the beam splitter is set to reflect to the side that will record which-path information, no interference pattern, and if the switch is set to the "quantum eraser" where which-path information is erased, we get the interference pattern.<br /><br />This is extremely counter intuitive, since the signal photon strikes the target before the idler photon even reaches the beam splitter and is told which way it will go.<br /><br />Now, lets take the path that the idler photon has to travel and stretch it out.. way out to say, 10 light minutes. (how this would be done is debatable, distant mirrors out in space? long fiber optic cable it has to travel down maybe? Anyway the specific method is not important)<br /><br />Now, I turn on my photon generator.. and what do I see? Either a

 

[fangsheath]

I'm a bt unclear about your hypothetical experiment, but perhaps this will answer your question. If you see/observe/measure the state of the signal photon at any point, you will find that at the conclusion of the experiment that everything agrees wth what you observed. It is only because the state of the photon is not measured until it hits the detector that the experiment works the way it does. It does not matter how long the paths are.<br /><br />This applies to the older quantum eraser experiments as well. It does not matter how far the photon travels before the "which-path" information iis "erased." What matters is that no measurement is taken before that which would resolve the question. Placing a polarizer in the path does not constitute a measurement. If you actually measure the state of the photon at that point, you will always get the same result.<br />

 

[heyo]

So you are saying then, that when you measure the which-path information after the signal photon hits the detector, you should still get an interference pattern at the detector?<br /><br />Then how come, in the DCQE, when they look at the subset of photons that correspond to idler photons for which which-path information was recorded, that subset does not fill out an interference pattern?<br /><br />Heyo<br />

 

[fangsheath]

No, that's not what I'm saying. What I'm saying is that a measurement is, by definition, irreversible. What is "erased" in these erasure experiments is not a measurement. In point of fact, in the Kim et al. study, which I assume is the one you're referring to, you do not see the interference pattern until you find the subset of detections that corresponds to "no-which-path" information. This is, I think, exactly what you would get in your hypothetical. You will not see the interference until the other side of the experiment is resolved.

 

[heyo]

It's a known experiment. Check out this link to get familiar with it.<br /><br />http://www.bottomlayer.com/bottom/kim-scully/kim-scully-web.htm<br /><br />Heyo

 

[heyo]

Yes, but the photons which make up the interference pattern would have had to "choose" their identity as a wave or particle before they "knew" wether which path information would be available in the future.<br /><br />The modifications to the experiment I propse, is to #1 extend the delay time to a macroscopic interval, and #2, to to replace the first beam splitter the idler photons hit with a switch that can be switched to send the photon on either the path where which-path information will be retained, or the path where it will not be retained.<br /><br />Then I am not talking about looking at subsets, I am talking about having the switch one way or the other and looking at the entire data set per run, manipulating the switch long after whatever pattern is going to appear on the signal photon detector has already appeared.<br /><br />Heyo

 

[bobvanx]

>>extend the delay time to a macroscopic interval,<br /><br />This is the point of the Quantum Astronomy experiment, written up at SDC a couple weeks ago. Photons from a gravitationally lensed quasar can either be collected as single hits or as an interference pattern.<br /><br />You know Radio is photons, just like infrared and visible light? So, we look at the images of the quasar in visible, and we see multiples. One of those images has a longer path to our eyes, and so we see distinct images. The photons remain discrete. As the wavelength observed lengthens, the photon path length difference is "reduced" and possibly at some point, the wavelength is long enough that the photons interfere with each other again.<br /><br />IOW short wavelengths, troughs and crests stay seperate and we see the images. Long wavelengths, the troughs and crests become closely aligned enough that we get the interference patterns.<br /><br />So, how does the photon, emitted as many as 13 billion years ago, "know" whether it is going to be detected as a wave, or a particle? Does the act of our detecting it, somehow transfer back in time to its start?

 

[fangsheath]

Perhaps I need to be more expository about these detectors and what is measured. The D0 detector in Kim et al. simply records "hits." That is, it registers a "1" if a photon hits it. It moves perpendicular to the incoming photons, back and forth, over a distance of a few millimeters, and simply records the hiits a different points. When the number of hits at each location is tallied, no interference pattern will be evident, even if you make sure that no "which-path" information wll ever be available. This is because there are actually 2 interference patterns, exactly out of phase with each other, so the pattern you see is a smooth curve. You can't tease the results apart and see whether there is interference or not without comparing D0 to the other detectors.<br />

 

[vogon13]

For some reason this is reminding me of Asimov's endochronomic propeties of re-sublimated thiotimoline. <br /><br />And I had a premonition this was all going to happen.<br /><br /> <div class="Discussion_UserSignature"> <p><font color="#ff0000"><strong>TPTB went to Dallas and all I got was Plucked !!</strong></font></p><p><font color="#339966"><strong>So many people, so few recipes !!</strong></font></p><p><font color="#0000ff"><strong>Let's clean up this stinkhole !!</strong></font> </p> </div>

 

[rprogrammer]

It tweaked me too, in two ways.<br />Both of which assuredly get asked to the point of monotony.<br />But just so we're clear:<br /><br /><br />1. What Heyo was saying, let me take a shot at saying it with fewer words.<br /><br />Given the delayed choice quantum eraser experiment by Kim et al,<br />Replacing the beamsplitters with switches to control the preservation or loss of the which-way information.<br /><br />If one slowed the photon carrying the which-way info so that one could check the measured pattern (for particle or wave patterns) before one made the decision to keep or destroy the which-way information, what would the pattern be? And would changing the preservation of the which-way info after the pattern has been recorded have any effect?<br /><br /> Which I believe was answered that once the pattern measuring device has been checked, its results are set in stone and changing the which-way preservation does nothing.<br /><br />This is just a repeat to see if I understood the summary of what was said. Please correct me if I'm wrong.<br /><br /><br /><br />2. However, I have another question which undoubtedly has been asked so much your ears surely bleed from it. But I cannot resist my curiosity:<br /><br /><br />Given that the switch (BSA+BSB) controls the which-way preservation and thus the pattern shown.<br />Does this not constitute as FTL Communication?<br /><br />I.E. The reciever stays at the bulk of the apparatus and at a predetermined time checks the pattern. While the transmitter, before the predetermined time, flips (or not) the switch regarding the which-way information.