Entanglement and interference effects

[ddd123]

When in a pure state, a particle exhibits interference effects in experiments. Wavefunction collapse alters this pure state: if two particles A and B at the opposite sides of a galaxy are entangled, does it mean that interference experiments done using A vary depending upon whether B has been collapsed on some eigenstate?

Thanks.

 

[bhobba]

When in a pure state, a particle exhibits interference effects in experiments. Wavefunction collapse alters this pure state: if two particles A and B at the opposite sides of a galaxy are entangled, does it mean that interference experiments done using A vary depending upon whether B has been collapsed on some eigenstate?

If that's what you want to do then QFT is the correct paradigm to analyse an interference setup the size of the Galaxy. What is done to the field at one point can't instantaneously effect the field at another and it won't be the same as one the size of a lab where that limitation is irrelevant.

Thanks
Bill

 

[ddd123]

I'm kind of confused, a wavefunction collapse would make the EPR effect work even when the setup is the size of a Galaxy.

 

[Catflap]

Yes, but since you have no way of knowing what the result was at B, it makes no observable difference to any measurement you make at A.

If I give you that the result of a spin experiment was 'up' can you tell me if that was a random value from a just-collapsed wavefunction or an inevitable result from a steady state?

 

[ddd123]

Yes, I'm strictly referring to interference experiments. What I mean is something similar, but much simpler, to the quantum eraser setup: one particle goes through slit experiments and the other's wavefunction is collapsed.

Say you have entangled particles A and B. You measure some observable of B and then perform interference experiments on A: the result should be different than that in which you leave B alone.

You don't predict information about what is being observed, but you know whether B has been observed or not, in itself.

 

[atyy]

http://arxiv.org/abs/quant-ph/9905039
Experimental realization of Popper's Experiment: Violation of the Uncertainty Principle?
Yoon-Ho Kim, Yanhua Shih
Found.Phys. 29 (1999) 1849

 

[Nugatory]

I'm kind of confused, a wavefunction collapse would make the EPR effect work even when the setup is the size of a Galaxy.

Indeed it would, but that just tells us that collapse is not a helpful way of way of thinking about galaxy-scale experiments.

Collapse is not a fundamental part of the mathematical formalism of QM; it's one way of thinking about what that formalism is telling us but it's not the only way. So you can stop trying to explain things in terms of collapse if it's not working for you.

 

[bhobba]

Yes, but since you have no way of knowing what the result was at B, it makes no observable difference to any measurement you make at A.

If I give you that the result of a spin experiment was 'up' can you tell me if that was a random value from a just-collapsed wavefunction or an inevitable result from a steady state?

Exactly. They are correlated - that does not mean what is done at A sends information to B, which relativity does not allow.

But that is not what you are talking about - you are talking about some galaxy wide screen and doing this massive double slit experiment on it. At least that's what I think was being asked - it wasn't really clear.

Thanks
Bill

 

[Nugatory]

Say you have entangled particles A and B. You measure some observable of B and then perform interference experiments on A: the result should be different than that in which you leave B alone.

You don't predict information about what is being observed, but you know whether B has been observed or not, in itself.

You have to be precise about what you're going to end up knowing. Are you stating a conclusion about what you learn when you run the experiment on a single AB pair, as your first paragraph suggests? Or are you stating a conclusion about the statistical properties of a large number of experimental runs on a large number of AB pairs?

 

[DevilsAvocado]

I'm kind of confused, a wavefunction collapse would make the EPR effect work even when the setup is the size of a Galaxy.

The word collapse is a little bit 'troublesome' since this is still 'under discussion', but yes, if non-locality turns out to be the solution to EPR-Bell experiments – then there is nothing in the theory stating that; "Okay guys, now you've run this prank for 500 km, time to stop this silly game!"

There is absolutely nothing (known today) setting a limit for entanglement state influences over long distances (except natural disturbances).

Also, the word instantaneous is a little bit 'tentative', because we just don't know the speed of entanglement. Today we know that that the lower bound of "spooky action at a distance" is four orders of magnitude of the speed of light.

We should also note that the possibility to preserve locality is still open, but then we would have to abandon realism (which is even 'spookier' than non-locality).

if two particles A and B at the opposite sides of a galaxy are entangled, does it mean that interference experiments done using A vary depending upon whether B has been collapsed on some eigenstate?

No. Entanglement can never be used to send FTL information (which would be the case if you could see interference Morse'ing on/off at the other end of the universe).

Reason? Entanglement and measurement is theoretically the same thing, thus; Entanglement = No interference.

But... what if we destroy the entanglement on one side of the universe?? Wouldn't that mean the interference would pop up at the other end!?

No, this would be a kind of Delayed choice quantum eraser. We would have to send measurement data by classical channels to the other end, to filter out the actual interference, and before we do this we will only see random noise.

Ron Garret explains everything nicely in this video, and why he will never get the Nobel Prize for his EPRG* Paradox.

https://www.youtube.com/watch?v=dEaecUuEqfc
http://www.youtube.com/embed/dEaecUuEqfc

*Einstein-Podolsky-Rosen-Garret

 

[Catflap]

The question is very unclear. A spin singlet state is a definite state. In that sense, it's already 'collapsed'.
In fact any quantum system has to be in some state or another at all times and is therefore always in a collapsed condition.

The whole point of the EPR conjecture was to try to 'trick' a quantum system into giving two different eigenvalues for an experiment simultaneously. Essentially being in more than one state at a single moment.

Unless you specify exactly what experiment you envision it's impossible to answer further.

 

[ddd123]

Sorry, I'll explain better. The logic of my post was more or less this: I know that it's impossible to send FTL information, and the observables can't be communicated in this way, but in my, obviously wrong, understanding the fact that something has been observed on the other side might show up. So I was asking what I'm missing.

But I guess it was exactly the delayed choice quantum eraser that stops me from seeing this particular information, as DevilsAvocado said.

 

[DevilsAvocado]

Yup!

 

[DrChinese]

When in a pure state, a particle exhibits interference effects in experiments. Wavefunction collapse alters this pure state: if two particles A and B at the opposite sides of a galaxy are entangled, does it mean that interference experiments done using A vary depending upon whether B has been collapsed on some eigenstate?

No. Entangled photons generally do not interfere in a double slit apparatus. See Zeilinger's "Experiment and the foundations of quantum physics"

http://www.hep.yorku.ca/menary/courses/phys2040/misc/foundations.pdf

See figure 2, page 290.

 

[Catflap]

In reply to the OPs last post :-

Well, it might - it probably does - that depend on the nature of what you are measuring.

But the point is - you have no way to know that whatever 'shows up' has any significance or is in any way related to the electron having once been part of an entangled pair. Nor can you learn anything about the nature of the entanglement that might have existed. Or even if it was still entangled before you just performed some measurement on it (that has most definitely untangled it)

 

[ddd123]

In reply to the OPs last post :-

Well, it might - it probably does - that depend on the nature of what you are measuring.

But the point is - you have no way to know that whatever 'shows up' has any significance or is in any way related to the electron having once been part of an entangled pair. Nor can you learn anything about the nature of the entanglement that might have existed. Or even if it was still entangled before you just performed some measurement on it (that has most definitely untangled it)

It could be used as a binary code, 0 = unobserved, 1 = observed. The idea is that different pure states interfere differently, now how exactly would one check that I don't know.

 

[DevilsAvocado]

But the point is - you have no way to know that whatever 'shows up' has any significance or is in any way related to the electron having once been part of an entangled pair.

This is correct; the (local) EPR-Bell outcome is always 100% random, no matter what. Correlations are always established afterwards, by classical channels.

However, by doing a double-slit experiment in one end, and selectively perform a EPR-Bell measurement in the other, you should be able to turn interference on/off like a big Morse signal, and that of course could be used to transmit FTL information (theoretically, to the other end of the universe).

But entanglement and (double-slit) interference don't party together = random noise in all cases.

 

[Catflap]

This is correct; the (local) EPR-Bell outcome is always 100% random, no matter what. Correlations are always established afterwards, by classical channels.

However, by doing a double-slit experiment in one end, and selectively perform a EPR-Bell measurement in the other, you should be able to turn interference on/off like a big Morse signal, and that of course could be used to transmit FTL information (theoretically, to the other end of the universe).

But entanglement and (double-slit) interference don't party together = random noise in all cases.

Now wait a minute - you only get one shot at performing a measurement on a member of an entangled pair. What you get may depend on it's entangled state but you can't tell that. All you get is a result to the experiment you performed. You can't tell the difference between a tangled and untangled particle.
So there's nothing you can turn on and off.

You can perform any sort of experimental measurement you want on one set of entangled particles and whatever experiment you like on the other, you will get a result but only by comparing the results of both experiments will you be able to see any correlation.
You need both sets of results to decode your Morse message.

 

[EskWIRED]

No. Entangled photons generally do not interfere in a double slit apparatus. See Zeilinger's "Experiment and the foundations of quantum physics"

http://www.hep.yorku.ca/menary/courses/phys2040/misc/foundations.pdf

See figure 2, page 290.

Thanks for the link.

Can you also point to a similar, but more recent article? I found it helpful to have lots of experiments discussed in the same context. For an armchair physicist, this type of collection clarifies and condenses months and years of piecemeal readings.

 

[DevilsAvocado]

Now wait a minute - you only get one shot at performing a measurement on a member of an entangled pair. What you get may depend on it's entangled state but you can't tell that. All you get is a result to the experiment you performed. You can't tell the difference between a tangled and untangled particle.
So there's nothing you can turn on and off.

Maybe some misunderstanding, I explained how this was supposed to work, not that it does...

The basic idea in the "Einstein-Podolsky-Rosen-Garret Paradox" is actually pretty smart. If you have an entangled pair that according to theory does not produce interference (on their own), you should be able to destroy the entanglement (i.e. by a quantum eraser) and then the particle left should be 'free' to generate any normal double-slit interference you wish, right?

Now, if this was possible (which it is not) we could use the particle left as a 'digital semaphore', on/off for binary messages, in a double-slit experiment:

on = interference = entanglement erased
off = no interference = entanglement persistent

There's hardly any difficulty to detect the interference in one end, right? And decide if to run a quantum eraser or not in the other, right?

But this EPRG Paradox does not work*.

Check out the video in post #10 for complete coverage (EPRG starts @25:18).


*Note also that this EPRG thought experiment requires the entangled pairs to be generated deterministic and effective, otherwise we would have to use timing and (classical) coincidence counting to establish the entanglement. This is afaik not the case with current technology...

 

[ddd123]

Yes it's exactly what I was asking, I felt stupid for not understanding this point through my study of EPR, but I'm getting used to the fact that in physics nothing comes easy.