In downconversion, can you get coherence between the signals from two sources?

[al onestone]

In the ZWM study of 1991 Mandel et al get coherence between the signal photons of two distinct downconversion sources. This is shown with an interference collection between the two possible sources of signal photons (even though they are entangled with the idler). This is achievable because the two downconversions are pumped by a common source which is beamsplit, and because the idler from one downconversion crystal is transmitted through the other downconversion crystal and superimposed with the second idler (made indistinguishable). Thus the name of his paper "Induced coherence without induced emission".

Is it possible to get coherence between two distinct sources of signals (without the transmission and superimposing of the two idlers) if you simply add the two idlers at a beamsplitter to make them indistinguishable (which negates the which-path information)? You would have to look for interference between the two signals in the same way as you would look for interference between any two separate indistinguishable sources of photons (Hull 1949, Pfleegor & Mandel 1967). You can "postselect" the observations that are in phase or you can look for temporal interference between two detectors. Note this is dis-similar to the DCQE.

A more thorough explaination of this thought experiment is at http://modifiedzwm.webs.com/

[SamBridge]

 

"In the ZWM study of 1991 Mandel et al get coherence between the signal photons of two distinct downconversion sources. This is shown with an interference collection between the two possible sources of signal photons (even though they are entangled with the idler). This is achievable because the two downconversions are pumped by a common source which is beamsplit, and because the idler from one downconversion crystal is transmitted through the other downconversion crystal and superimposed with the second idler (made indistinguishable). Thus the name of his paper "Induced coherence without induced emission".

Is it possible to get coherence between two distinct sources of signals (without the transmission and superimposing of the two idlers) if you simply add the two idlers at a beamsplitter to make them indistinguishable (which negates the which-path information)? You would have to look for interference between the two signals in the same way as you would look for interference between any two separate indistinguishable sources of photons (Hull 1949, Pfleegor & Mandel 1967). You can "postselect" the observations that are in phase or you can look for temporal interference between two detectors. Note this is dis-similar to the DCQE.

A more thorough explaination of this thought experiment is at http://modifiedzwm.webs.com/"

 

 

 

 

I don't quite understand what your saying, but I think the answer is both yes and no depending on the interpretation. When you run a photon through a beam splitter, it's probability distribution gets divided up based on the angle. You can make it so that light goes 50% in one direction and 50% in another, but once a single photon is measured it will go into an Eigenstate and it's wave function will collapse only to a single point, so still only 1 detector would detect one photon at a time, 2 detectors will not measure the same single photon even if the probability is distributed 50/50. It's like flipping a coin. If you and your friend are gathered around a coin you flip in the air, you call heads he calls tails, you only get one or the other.

So, if you split the beam, then rejoin it again, it doesn't matter, a photon will still go into an Eigenstate and only be measured once, and there is no way to predict exactly which one will measure it or to trace the exact trajectory if it was in a probabilistic wave form prior to measurement.

 

Edited January 12, 2013 by SamBridge

[SamBridge]

Ok, so this is my second post on the topic, why do I still see that it has "0" views? Is that some glitch in the system?

[quantropy]

 

In the ZWM study of 1991 Mandel et al ... Is it possible to get coherence between two distinct sources of signals (without the transmission and superimposing of the two idlers) if you simply add the two idlers at a beamsplitter to make them indistinguishable (which negates the which-path information)? ... Note this is dis-similar to the DCQE. A more thorough explaination of this thought experiment is at http://modifiedzwm.webs.com/

 

 

There's a discussion of FTL by downconverting at http://sci.tech-archive.net/Archive/sci.physics.research/2004-10/0473.html, which (I think) explains why downconverting doesn't allow FTL signalling. I'm not sure whether that is what is being asked - are you saying that you know that the FTL idea won't work, and looking at something else?

[al onestone]

Sorry for the delayed response but they moved my thread from physics to speculations, and I didn't even know this section existed or I would have originally posted here.

 

Quantropy, I do know exactly why conventional ideas in FTL don't work, and its for the reason that no-signalling theorem states, operations on Bob's system do not disturb the partial state operator of Alice's system (assuming Alice and Bob have members of a downconverted pair). But these are "conventional" ideas. I looked at the link you gave (from 2004) and it does give a relevant thread where a similar idea is presented, but that idea was just a response from someone who had recently read about the delayed choice quantum eraser. The real original paper concerning downconverted pairs and the possibility of FTL is the ZWM paper,(http://quantmag.ppole.ru/Articles/experiment/Mandel(1991).pdf) I highly suggest it.

In the ZWM paper the authors show how interference between signals (second order interference) is collected even though the signals are entangled with the idlers that are being collected coincidentally. The ZWM is not capable of FTL because the idler from one downconversion is transmitted through the other downconversion so that the phases are locked. This is how they get coherence and thus interference, as well as for the reason that the two crystals are commonly pumped by one source. This is required for coherence because downconversion is spontaneous and highly improbable. If the pump goes to the one crystal and downconverts, you can't assume that if it had gone to the other crystal it would have downconverted there as well, its one in a billion atleast.

What I propose is to separate the crystals all together, and pump them separately. So how do you get coherence between separate emissions of signal photons? The same way you would get coherence between any two seperate sources of photons, the way they did it back in the 60's (see Pfleegor & Mandel 1967,1968), by simply monitoring the relative phase between the two emissions and postselecting the observations that are in phase. For a detailed explaination I suggest taking a closer look at my thought experiment at http://modifiedzwm.webs.com