Photons from separated sources can be entangled - after they were detected

[Lord Crc]

As someone who has only read pop-sci stuff about QM, I'm a bit slow on how to interpret the setup proposed in the first post. If anyone could help me out I'd be very grateful as I find these things very fascinating! My main confusion lies with what Alice and Bob will measure.

Let's say Alice sets her detector to theta degrees and say Bob also sets his to theta degrees. If photons A and D are entangled then they will measure the same 100% of the time (or 0% if the photons are anti-correlated). Now, if A and D are NOT entangled, then, since the pairs AB and CD are independent, they would measure the same 50% of the time (ie independent uniformly random distributions).

Assuming the above is correct, then I fail to see how Charlie's information is needed for Alice and Bob to figure out if the photons are entangled. That is, I understand they can't do so on a single pair basis, but assume that Charlie ensures that the pairs are always entangled. Then how is this different from the situation where Charlie doesn't exist and Alice and Bob performs the measurements on a single pair of entangled photons?

However if it results in the same situation as a single pair measurement, then it would seem that Charlie has no free will, assuming he doesn't change his mind too often :)

Due to this I assume that I missed a turn somewhere, invalidating my subsequent assumptions. Anyone care to shed some light? Thanks in advance.

 

[zonde]

Here is a great new experimental report on the subject. Entanglement from fully independent sources are entangled, in a cleaner and much more accurate fashion that some of the earlier references provided. Here S=2.54, and Bell's Inequality is violated by 27 standard deviations.

http://arxiv.org/abs/1005.1426

Experimental non-local generation of entanglement from independent sources

Authors: Xian-Min Jin, Jügen Röch, Juan Yin, Tao Yang (2010)

"We experimentally demonstrate a non-local generation of entanglement from two independent photonic sources in an ancilla-free process . Two bosons (photons) are entangled in polarization space by steering into a novel interferometer setup, in which they have never meet each other. The entangled photons are delivered to polarization analyzers in different sites, respectively, and a non-local interaction is observed. Entanglement is further verified by the way of the measured violation of a CHSH type Bell's inequality with S-values of 2.54 and 27 standard deviations. Our results will shine a new light into the understanding on how quantum mechanics works, have possible philosophic consequences on the one hand and provide an essential element for quantum information processing on the other hand. Potential applications of our results are briefly discussed. "

I would say that experiment from Zeilinger's group is much better then this one. Here this non-locality argument is quite dubious as all photons go through the same interferometer.

And if you compare them from perspective of practical application it's clear that experiment from Zeilinger's group can be used to develop quantum repeaters where this one can't. I think it's quite reasonable way how to make distinction between true quantum "non-locality" and feigned quantum "non-locality".

EDIT: Thought that in last sentence I should have used "true/feigned non-local generation of entanglement" instead of "true/feigned quantum non-locality".

 

[DrChinese]

As someone who has only read pop-sci stuff about QM, I'm a bit slow on how to interpret the setup proposed in the first post. If anyone could help me out I'd be very grateful as I find these things very fascinating! My main confusion lies with what Alice and Bob will measure.

Let's say Alice sets her detector to theta degrees and say Bob also sets his to theta degrees. If photons A and D are entangled then they will measure the same 100% of the time (or 0% if the photons are anti-correlated). Now, if A and D are NOT entangled, then, since the pairs AB and CD are independent, they would measure the same 50% of the time (ie independent uniformly random distributions).

Assuming the above is correct, then I fail to see how Charlie's information is needed for Alice and Bob to figure out if the photons are entangled. That is, I understand they can't do so on a single pair basis, but assume that Charlie ensures that the pairs are always entangled. Then how is this different from the situation where Charlie doesn't exist and Alice and Bob performs the measurements on a single pair of entangled photons?

However if it results in the same situation as a single pair measurement, then it would seem that Charlie has no free will, assuming he doesn't change his mind too often :)

Due to this I assume that I missed a turn somewhere, invalidating my subsequent assumptions. Anyone care to shed some light? Thanks in advance.

I think you have summarized things pretty well. The issue is that clearly, sometimes A & D ARE in fact entangled. And that depends on what happens with B & C. There is no questioning the effect really. It is just: how do you interpret it? (And no interpretation is even required!)

 

[zonde]

Assuming the above is correct, then I fail to see how Charlie's information is needed for Alice and Bob to figure out if the photons are entangled. That is, I understand they can't do so on a single pair basis, but assume that Charlie ensures that the pairs are always entangled. Then how is this different from the situation where Charlie doesn't exist and Alice and Bob performs the measurements on a single pair of entangled photons?

Two photons from two different pairs interact at Charlie's beam splitter. Now there are two possibilities either both photons appear at the same output of Charlie's beam splitter or they appear at different outputs. Entangled pairs at Alice and Bob are those where corresponding photons at Charlie are detected in different outputs. And Charlie have to communicate this information to Alice and Bob so that they could create appropriate subsample. If Alice and Bob do not check they pairs against Charlie's data they do not see any entanglement.

 

[Lord Crc]

Entangled pairs at Alice and Bob are those where corresponding photons at Charlie are detected in different outputs.

Ah there's my misunderstanding. I was under the impression Charlie could determine which pair to entangle, not that he merely observed which ones were which. That changes things a bit :)

Thanks to both of you. While the experiment seems a bit less weird now, it's never the less very fascinating to contemplate.