Quantum Entanglement

[Mordeth]

Quote

Quoted from Eise:

 

So if I change the charge of one plate, somebody can measure an instantaneous change at the other plate. Wouldn't that be a violation of special relativity?

@Menan

You show that you do not understand entanglement. 

Let's go one step at a time. First a classical example. I have a bag of balls, they are all red or green. Without looking I pick two balls, and I put them in separate boxes. I keep one, and send the other far away. Then I open my box, and see it is a green ball. What can I conclude about the colour of the ball in the remote box? Right, nothing. And why? Because there was nothing special with my picks. It could have been two reds, two greens, or one red and green.

Now I pick, looking of course, one red and one green, and put them in two separate boxes. So what I did here is 'entangle' the balls. Now I shuffle the two boxes, so that I do not know which one is which. If I open one, and see that it is red, I immediately know that the ball in the other box is green. And of course, this is independent on the distance. If I send the second box lightyears away, and only then open my box, I still know immediately what some alien sees when he opens his box. I know it because the observations are correlated. And the correlation already happened at the moment of my picks. That is the moment of entanglement. It is not when the boxes are opened.

Now in quantum physics, there are processes where two particles pop out, which have e.g in one aspect always opposite values. Say the direction of spin. So if I measure the spin e.g. in a vertical direction, say it is 'up', then I immediately know that the other one will measure spin 'down', when also measured in the vertical direction. But as with the balls, the 'moment of entanglement' is when these particles popped into existence. But in quantum physics a few things are different: first, it is impossible to say which particle has which spin without measuring (it is as if I created the green and red balls, including their boxes, without knowing which ball is in which box). But as the two particles are entangled, if I measure both, the measurements will always be correlated. And there is nothing special with correlation: if I send one particle far away, and then measure my particle in the vertical direction, and the alien measures his particle in the same direction, I will always know what he measures: the opposite of my measurement. 

The 'spooky' aspect comes in when we do not know from each other in which direction we measure the spin. It can be vertical, horizontal, 30^o, 45^o, 55.3977^o. What we find is that the correlation is stronger than one would expect if we would assume that the particles already had a definite spin from the beginning. But it still is correlation, not causation. As with the red and green balls, there is no direct causal relationship between my and the alien's observation. The causal relationship goes back to the moment of 'entanglement'. Everything afterwards is just correlation, and therefore cannot be used to transfer information. And because there is no causal relationship between my measurement of the spin of my particle, and the alien's measurement, I cannot use entanglement for sending information. 

And all this is very well understood by all quantum physicists, and is no secret at all.

Hello,

I have a question related to Eise's very useful explanation above, which was provided in another post.  My question is related to what appears to be the crux of quantum entanglement and the subsequent loss of entanglement via measurement.  My question is related to the statement by Eise above that says: 

"The 'spooky' aspect comes in when we do not know from each other in which direction we measure the spin. It can be vertical, horizontal, 30^o, 45^o, 55.3977^o. What we find is that the correlation is stronger than one would expect if we would assume that the particles already had a definite spin from the beginning. "

Is it possible to explain these statements further?  Preferably with details.   Everything else is very easy to follow, except this exact part, which seems to be very critical in understanding the topic.  If it is possible, please contain any explanations to this one particular detail - "...the correlation is stronger than one would expect..." (unless necessary to explain the answer).    I am not arguing or debating any point, but rather attempting to understand.  

Thank you. 

[Strange]

7 minutes ago, Mordeth said:

Is it possible to explain these statements further?  Preferably with details.   Everything else is very easy to follow, except this exact part, which seems to be very critical in understanding the topic.  If it is possible, please contain any explanations to this one particular detail - "...the correlation is stronger than one would expect..." (unless necessary to explain the answer).    I am not arguing or debating any point, but rather attempting to understand.  

This comes down to Bell's Theorem, which showed that the probabilities of certain correlations could not be explained by a classical theory.

If you measure the polarisation of a photon at any angle, it will either be "up" or "down" with respect to that angle, and an entangled photon will have the opposite polarisation. Then the question is, what if you could measure the polarisation at two different angles? What are the probabilities of the first one being "up" or "down" and the second one being "up" or "down" (for all the combinations). Bell showed that for a quantum system you will get different results than for a classical system (e.g. the red and balls in Eise's example, or the pairs of socks often used in examples).

What makes it tricky is that measuring the polarisation at any angle "destroys" the polarisation information at the other angle. This is where entanglement comes in: we generate a pair of entangled photons and measure one angle on the first photon and a different angle on the second photon. We do this lots of times and check the probability of these matching.

What this comes down to is that in the classical examples of the coloured balls, the actual colour in each box is fixed ("real") as soon as we put them in the boxes. In the quantum case, the values are not determined until we actually measure them (in other words: there are no hidden variables).

This is a much better (and probably more accurate) explanation; it is quite detailed but only uses simple math: http://drchinese.com/David/Bell_Theorem_Easy_Math.htm

 

[studiot]

1 hour ago, Mordeth said:

"The 'spooky' aspect comes in when we do not know from each other in which direction we measure the spin. It can be vertical, horizontal, 30^o, 45^o, 55.3977^o. What we find is that the correlation is stronger than one would expect if we would assume that the particles already had a definite spin from the beginning. "

Whilst Eise might spot this and give a better answer (His posts are usually very well thought out) here is a quick answer.

 

Eise's text was stated to be classical spin.

Quantum 'spin' is not the same.

However it has the same property of chirality in that there is no absolute measure of handedness or chirality so If A measures the spin of a quantum particle and a (very) remote B is told (at the speed of light) the spin of my particle was up, he is no wiser unless he also has a means of determining which way the up arrow was pointing.

[Mordeth]

Thank you very much for the explanations Strange and studiot.  Much appreciated!  And thank you for the link Strange.  The lightbulb went off after reading the explanations here and then following that link!   I had initially read the entire wikipedia article on the subject and simply couldn't understand the fundamental nature of the issue.   My initial senses and internal bias appealed to Einstein's view via EPR, but mostly because I didn't even understand what the fundamental "issue" or "debate" was to begin with.  It seemed a non-issue, especially when considering some of the analogies provided.       But I see now what the nature of the problem is (using some basic math to express it was needed) and I very much appreciate your assistance in this understanding.  I wrote it all down myself so that I could see the numbers myself.  

From following more of the links, it is extremely interesting (and surprising) to me that literally every single experiment shows a Bell inequality violation, including those that have apparently closed most of the "loopholes".   My wager was most certainly in favor of "local hidden variables", as it made sense to me.  But I see now that my intuition was wrong.  

[Eise]

18 hours ago, studiot said:

Eise's text was stated to be classical spin.

Quantum 'spin' is not the same.

Ehhmmm, no. It was meant as QM-Spin. Maybe the misunderstanding lies here: you think that I said that the spin measured can be 30^o, 45^o, 55.3977^o, , but I said 'in which direction we measure the spin'. Instead 'up' of down', maybe one should use 1/2 and -1/2. So I have my measuring device under different angles, and measure the correlation between the other, also under different angles.

As examples:

Measurement Angle Device 1  Measurement Angle Device 2 Correlation

         0                                                            0                                      100%

         0                                                           90                                         0%

      90                                                           90                                     100%

       0                                                            45                                        55%

(The last percentage is my invention).

Does that clear it up?

 

 

[studiot]

31 minutes ago, Eise said:

Ehhmmm, no. It was meant as QM-Spin. Maybe the misunderstanding lies here: you think that I said that the spin measured can be 30^o, 45^o, 55.3977^o, , but I said 'in which direction we measure the spin'. Instead 'up' of down', maybe one should use 1/2 and -1/2. So I have my measuring device under different angles, and measure the correlation between the other, also under different angles.

As examples:

Measurement Angle Device 1  Measurement Angle Device 2 Correlation

         0                                                            0                                      100%

         0                                                           90                                         0%

      90                                                           90                                     100%

       0                                                            45                                        55%

(The last percentage is my invention).

Does that clear it up?

 

 

Sorry If I misunderstood, somewhere you mentioned classical physics and I just jumped to the conclusion that the use of classical angles implied classical spin.

Quantum spin does not have a reference axis. Up and down or +1/2 and -1/2 are just references.

But the left hand v right hand of chirality still applies. (and is reflected in the + and - signs)

[swansont]

20 minutes ago, studiot said:

Quantum spin does not have a reference axis. Up and down or +1/2 and -1/2 are just references.

Yes it does, but it's what we pick. We define the quantization axis, and it is with respect to that axis that we measure the spin projection. Eise's description reflects that no matter what quantization axis we choose, we still see these correlations, which is not what you would expect if the spin were actually set, and simply unknown. If you do a measurement and I do one, but we do not communicate about what axis was chosen, then we will not see the expected correlations (except if we choose the same/similar setup by accident)

[Itoero]

7 hours ago, swansont said:

Yes it does, but it's what we pick. We define the quantization axis, and it is with respect to that axis that we measure the spin projection. Eise's description reflects that no matter what quantization axis we choose, we still see these correlations, which is not what you would expect if the spin were actually set, and simply unknown. If you do a measurement and I do one, but we do not communicate about what axis was chosen, then we will not see the expected correlations (except if we choose the same/similar setup by accident)

I probably misinterpret this, but don't you basically mean that quantum spin is a physical spin analogous to the spin of a planet?

[Strange]

35 minutes ago, Itoero said:

I probably misinterpret this, but don't you basically mean that quantum spin is a physical spin analogous to the spin of a planet?

There are a couple of differences from classical spin. 

The first is that a classical object can only have spin around one axis. But we can (by using entangled particles) measure quantum spin in two different axes.

The second is that the value of the spin (up or down, + or -) is not just unknown before we measure it, but is actually not determined until we measure it.

And, particles are modelled as zero-size particles so physical spin doesn't really make sense,

[swansont]

2 hours ago, Itoero said:

I probably misinterpret this, but don't you basically mean that quantum spin is a physical spin analogous to the spin of a planet?

No, I definitely do not mean that.

2 hours ago, Strange said:

There are a couple of differences from classical spin. 

The first is that a classical object can only have spin around one axis. But we can (by using entangled particles) measure quantum spin in two different axes.

You don't even need entanglement. no matter how you orient your coordinate system, there will only be spin up or spin down, relative to that system.