I Few Questions about: TU delft "A loophole-free Bell test"

[grzegorzsz830402]

Electrons in diamonds are not entangled.

Is above statement correct?

Spin of emitted fotons if detected separately would not show any correlation?

Is above statement correct?

Is there assumption that spin of emitted fotons cannot change when reflecting or passing through beam spliter? Or under any other conditions?

Does Bell test requires for photon to be thought of as massless particle?

 

[StevieTNZ]

Electrons in diamonds are not entangled.

Is above statement correct?

All electrons in the universe are technically entangled.

 

[vanhees71]

I guess you talk about this experiment:

https://www.nature.com/articles/srep30289 (open access!)

Then of course the used electrons are entangled. How else should you perform a Bell test, which is about entangled states by definition? As explained in the first paper in detail, the entanglement is ensured by entanglement swapping using photons.

 

[grzegorzsz830402]

My fault, to be more precise. Are they entangled before (before entanglement swapping)?

Is there assumption that spin of emitted fotons cannot change when reflecting or passing through beam spliter? Or under any other conditions? (In context of hidden variable analysis and statistical predictions)

Does Bell test requires for photon to be thought of as massless particle?
(If current model assumptions are incorrect then based on that model statistical predictions for hidden variable could be also flawed).

 

[vanhees71]

As far as I understood the paper, the electrons are not entangled before but get entangled by entanglement swapping, or more precisely, with help of the photons you select a set of entangled photons. Beam splitters and other optical instruments can simply be thought of in terms of how they act on electromagnetic waves. Photons are just particular states, socalled Fock states, of electromagnetic wave fields, and the polarization etc. can all be calculated using the well-known Fresnel formulae for classical electromagnetic waves.

Bell tests can in principle be done with any system, provided you are able to prepare it in entangled states. Here electrons as well as photons are used to prepare entangled states. That very often photons are used to do Bell tests is simply, because nowadays they are easy to prepare as entangled photon pairs (parametric downconversion) and the electromagnetic interaction is pretty weak, so that it's not too difficult to prevent decoherence through interactions with "the environment".

 

[StevieTNZ]

My fault, to be more precise. Are they entangled before (before entanglement swapping)?

They are, technically. Whether they exhibt a Bell inequality violation is another question. The entanglement swapping would likely achieve that.

 

[grzegorzsz830402]

So, of course, experiment would produce the same results, yet predictions for "hidden variable" could be different, if some current assumptions would turn out to be false, therefore interpretation of results could change aswell.

There is assumption for quantum theory as well as for "hidden variable", that (emitted photon spin = detected photon spin).

So, within constraints of that assumption, quantum theory explanation of experiment results is that, whatever happens at beam spliter, affects electrons trapped in diamonds (they become entangled through entanglement swapping) and that affects entangled with them fotons (their spins) and that is the reason why we get correlation of detected spins higher than chance.

Within constraints of that assumption for "hidden variable" detection correlation higher than chance can be explained only by experiment setup somehow affecting randomness of emitted fotons (their spins).

I do not think, this possibility have been completely excluded, especially if the same pair of electrons is used for each run.
As, whatever is happening, at beam spliter and detectors, can potentially afect next run possibilities.
Although I see it as a viable possibility,
personally, I do not think, that is what is happening.

If assumption (emitted spin = detected spin) is flawed.

Then, emitted photons (their spins) can be altered at beam splitter.

And, I do think that is exactly what is happening.

Comparison of entanglement to entanglement swapping allows to observe some similarities and differences.

Hidden Variable scenario (entanglement):
Photon with a defined spin is divided, and its trajectory is redirected.
Spin of new pair of entangled Photons is independent from the source spin. Yet correlated with each other.

Hidden Variable scenario (entanglement swapping):

Here, we have redirection of trajectory possibility, and maintenance of trajectory possibility, for both photons, those conditions promote opposite spin detection.
(I can provide more details, why would that be the case).
And, for detection at both detectors, both photons have to either reflect, or pass through.

I couldn't find any information, that would exclude the possibility.

I have experiment idea, that could test, and prove or disprove that possibility.
(I can provide more details, if interested.)

If you have information, that exclude this possibility, I would like to examine them closely.

 

[StevieTNZ]

So, of course, experiment would produce the same results, yet predictions for "hidden variable" could be different, if some current assumptions would turn out to be false, therefore interpretation of results could change aswell.

There is assumption for quantum theory as well as for "hidden variable", that (emitted photon spin = detected photon spin).

So, within constraints of that assumption, quantum theory explanation of experiment results is that, whatever happens at beam spliter, affects electrons trapped in diamonds (they become entangled through entanglement swapping) and that affects entangled with them fotons (their spins) and that is the reason why we get correlation of detected spins higher than chance.

Within constraints of that assumption for "hidden variable" detection correlation higher than chance can be explained only by experiment setup somehow affecting randomness of emitted fotons (their spins).

I do not think, this possibility have been completely excluded, especially if the same pair of electrons is used for each run.
As, whatever is happening, at beam spliter and detectors, can potentially afect next run possibilities.
Although I see it as a viable possibility,
personally, I do not think, that is what is happening.

If assumption (emitted spin = detected spin) is flawed.

Then, emitted photons (their spins) can be altered at beam splitter.

And, I do think that is exactly what is happening.

Comparison of entanglement to entanglement swapping allows to observe some similarities and differences.

Hidden Variable scenario (entanglement):
Photon with a defined spin is divided, and its trajectory is redirected.
Spin of new pair of entangled Photons is independent from the source spin. Yet correlated with each other.

Hidden Variable scenario (entanglement swapping):

Here, we have redirection of trajectory possibility, and maintenance of trajectory possibility, for both photons, those conditions promote opposite spin detection.
(I can provide more details, why would that be the case).
And, for detection at both detectors, both photons have to either reflect, or pass through.

I couldn't find any information, that would exclude the possibility.

I have experiment idea, that could test, and prove or disprove that possibility.
(I can provide more details, if interested.)

If you have information, that exclude this possibility, I would like to examine them closely.

I think this is bordering on personal speculation - something, if you are doing so, not permitted on PF.

 

[DrChinese]

There is assumption for quantum theory as well as for "hidden variable", that (emitted photon spin = detected photon spin).

So, within constraints of that assumption, quantum theory explanation of experiment results is that, whatever happens at beam spliter, affects electrons trapped in diamonds (they become entangled through entanglement swapping) and that affects entangled with them fotons (their spins) and that is the reason why we get correlation of detected spins higher than chance.

Within constraints of that assumption for "hidden variable" detection correlation higher than chance can be explained only by experiment setup somehow affecting randomness of emitted fotons (their spins).

I do not think, this possibility have been completely excluded, especially if the same pair of electrons is used for each run.

As, whatever is happening, at beam spliter and detectors, can potentially afect next run possibilities.

Although I see it as a viable possibility,
personally, I do not think, that is what is happening.

If assumption (emitted spin = detected spin) is flawed. Then, emitted photons (their spins) can be altered at beam splitter. And, I do think that is exactly what is happening.

Comparison of entanglement to entanglement swapping allows to observe some similarities and differences.

Hidden Variable scenario (entanglement):
Photon with a defined spin is divided, and its trajectory is redirected.
Spin of new pair of entangled Photons is independent from the source spin. Yet correlated with each other.

Hidden Variable scenario (entanglement swapping):

Here, we have redirection of trajectory possibility, and maintenance of trajectory possibility, for both photons, those conditions promote opposite spin detection.
(I can provide more details, why would that be the case).
And, for detection at both detectors, both photons have to either reflect, or pass through.

I know English is probably not your first language. So I am trying to understand where you are making statements, and where you are asking questions. Also, you refer to assumptions. But there are many differences between assumptions involving Quantum Theory and assumptions regarding Hidden Variable theories (which are generally ruled out by experiments such as the Henson et al paper, cited in post #3 by @vanhees71). Also available below (this is by the same team, but is the earlier version of the experiment):

https://arxiv.org/abs/1508.05949
Experimental loophole-free violation of a Bell inequality using entangled electron spins separated by 1.3 km
B. Hensen,1, 2 H. Bernien,1, 2, ∗ A.E. Dr´eau,1, 2 A. Reiserer,1, 2 N. Kalb,1, 2 M.S. Blok,1, 2 J. Ruitenberg,1, 2 R.F.L. Vermeulen,1, 2 R.N. Schouten,1, 2 C. Abell´an,3 W. Amaya,3 V. Pruneri,3 M.W. Mitchell,3, 4 M. Markham,5 D.J. Twitchen,5 D. Elkouss,1 S. Wehner,1 T.H. Taminiau,1, 2 and R. Hanson1, 2, †

The entangled photons emitted from the PDC crystal are in a superposition of polarization states. It is not possible to assert this is the same state as what was observed later, regardless of whether an entanglement swap is performed or not. I think you understand this, but it was not entirely clear. You are correct that the input photons have a well-defined polarization entering the PDC crystal, and that polarization must be a specific one in order to produce a good quality superposition. Good quality superpositions are needed for the entanglement to exist initially.

You mention the idea that there is some kind of "memory" at the beam splitter that affects the next run probabilities ("possibilities"). There are many reasons your idea cannot be correct. 1) The ONLY factor that figures into the quantum prediction is the relative polarization settings of the 2 observers. That setting can be selected after the beam splitter is encountered. That wouldn't be true if there were additional hidden variables (as other effects must net out to zero impact). 2) The outcomes are random to the limit of our testing. 3) In an entanglement swap, the photons that are identified as being in a suitable Bell state do not enter the swapping apparatus consecutively (which would be necessary to cause some kind of bias). There might literally be >1,000,000 photons entering the beam splitter in between. And it might be minutes/hours between suitable pairs that arrive within the narrow coincidence time window. That is because the creation of entangled PDC pairs occurs at random times, and you need 2 pairs being created within the window. Most input photons do not "split", they go straight through the crystal. Perhaps 1 in 10,000,000 split and exit in an entangled state. 4) There are published references you can read to satisfy yourself that there is no "quantum memory" from run to run. See below for a start.

https://arxiv.org/abs/quant-ph/0205016

Just to be clear: a beam splitter (BS) does not alter/measure photon polarization. Instead, that is done with a polarizing beam splitter (PBS). Both a BS and a PBS are used when an entanglement swap is performed (location C in the reference). On the other hand, a BS is not part of the setup where a Bell/CHSH Inequality is being tested and calculated (locations A and B).

The design of the experiment we are looking at considers many different quantum and classical elements. This is an advanced experiment, and takes a bit of study (including background) to follow it. Another experiment using entanglement swapping is referenced below, and it may be a little easier to follow (although it is advanced too).

https://arxiv.org/abs/0809.3991
High-fidelity entanglement swapping with fully independent sources
Note: one of the authors (Zeilinger) won the 2022 Nobel for his work, including papers such as this.

 

[grzegorzsz830402]

I agree.
"there is no "quantum memory" from run to run."

Simplifying:
For successful runs there are 2 pairs of entangled fotons. And 4 detectors. 2 at location C, 1 at location A and 1 at location B. Detectors at location C show opposite spins correlation and detectors at location A and B also show opposite spins correlation.

For detection at location C there is also detection at location A and B. And that also have to be addressed.

Question:
Is there correlation between detector at location A with detector at location C. And beetween detector at location B with detector C.
(Example A=up, B=down, C1=up,C2=down, A=down B=up, C1=down, C2=up)

 

[DrChinese]

I agree.
"there is no "quantum memory" from run to run."

Simplifying:
For successful runs there are 2 pairs of entangled fotons. And 4 detectors. 2 at location C, 1 at location A and 1 at location B. Detectors at location C show opposite spins correlation and detectors at location A and B also show opposite spins correlation.

For detection at location C there is also detection at location A and B. And that also have to be addressed.

Question:
Is there correlation between detector at location A with detector at location C. And beetween detector at location B with detector C.
(Example A=up, B=down, C1=up,C2=down, A=down B=up, C1=down, C2=up)

The C detectors (where the Bell State Measurement/BSM is performed) are not polarization correlated with the A or B polarization detectors WHEN a successful Bell state is registered. In other cases it could be.

Keep in mind that for the successful BSM to be registered, you cannot possibly know which photon is paired with the A detection (or the B detection). So there is no way to correlate the original 1 & 2 photons, for example. Again, this rule applies when the indistinguishability requirement is met (a requirement for swapping to occur).

 

[grzegorzsz830402]

I want to thank for all replies.
They helped me understand what bit of information I was missing, excluded some possibilities and narrow down my focus.

There is one more thing that I am interested about, yet I could not find it. There was 245 runs that met requirements. Out of those 245 runs 80% show opposite spins correlation at location C and, from what I am guessing, also at location A and B. So my understanding is that 20% showed the same spin detection at location C, and I am interested about what detectors at location A and B have shown.
My guess would be also same spin yet opposite to the one at location C. It would surprise me if it would be totally random.

So, If any one could guide me to raw data, from 4 detectors for those 245 runs, I would be very grateful.

 

[grzegorzsz830402]

Hey Steve.
It kind of was personal speculation, based on incomplete understanding.
Yet, isn't everything is, that threads on edge of knowledge (understanding)?
It is a funny place, where we are all silly. Cheer up, and if you see me here stumbling and tripping, don't let it annoy you, let it entertain you.

 

[DrChinese]

1. There was 245 runs that met requirements. Out of those 245 runs 80% show opposite spins correlation at location C and, from what I am guessing, also at location A and B. So my understanding is that 20% showed the same spin detection at location C, and I am interested about what detectors at location A and B have shown.

2. My guess would be also same spin yet opposite to the one at location C. It would surprise me if it would be totally random.

1. The results for the 245 runs were detections at A and B (not C as you say). They are correlated and marked for the 4 different terms used in the CHSH calculation of S=2.42. (Local realism demands S<=2.) The generic form of the CHSH inequality is presented as equation (1). The important thing to understand is that the 4 terms correspond to 4 different angle pair settings for A and B. That's 2 (for A) x 2 (for B) = 4 permutations, selected randomly mid-flight. You can see what the settings were by looking at the clock hands in Figure 4.a. These pairings are chosen to demonstrate maximal violation relative to the CHSH S value. The quantum expectation value (theoretical prediction) for correlation is: cos^2(theta)-sin^2(theta) where theta=|A-B|.

2. These pair settings have no connection whatsoever to the polarizer angle setting at C, so there is no correlation to look for. For this experiment in fact, polarizers are used at C rather than polarizing beam splitters. That means all of the qualifying events represent the same Bell state, which is psi-. Note that the A & B correlation is a function of theta (A-B) , and the polarization seen at C is not a factor in that function.

 

[grzegorzsz830402]

I do appreciate your patience, and have not intention to abuse it.

There is a reason for detection at location A and B aswell as at location C.

Is the reason is to determine
are measured fotons entangled or not?

245 runs were selected because detection at A and B met the criteria. (coincidence time window)

Does measurement of fotons at location A and B tell as are measured fotons entangled or not?

Is it only to the value that is measured?

Or, would it be safe to say that if they are entangled they would have opposite spins if measured?
And, if they are not entangled they would have same spin if measured?

Does measurement of fotons at location C tell as are measured fotons entangled or not?

From what I understand yes if opposite spins are detected and no if the same spins are detected.

 

[DrChinese]

There is a reason for detection at location A and B aswell as at location C. Is the reason is to determine
are measured fotons entangled or not?

245 runs were selected because detection at A and B met the criteria. (coincidence time window)

Does measurement of fotons at location A and B tell as are measured fotons entangled or not? Is it only to the value that is measured?

Or, would it be safe to say that if they are entangled they would have opposite spins if measured?
And, if they are not entangled they would have same spin if measured?

Does measurement of fotons at location C tell as are measured fotons entangled or not?

From what I understand yes if opposite spins are detected and no if the same spins are detected.

Keep in mind that this experiment is perhaps the most complex possible for us to discuss, and not really a good learning tool. So take my analysis with a grain of salt, there is some simplification for sake of clarity. For an event to qualify as one of the 245:

A: 1 possible detection, value is marked according to angle (P0 or P1 selected by the RNG) and whether photon was detected (+1) or not (-1) within a 12 ns time frame.
B: 1 possible detection, value is marked according to angle (P0 or P1 selected by the RNG) and whether photon was detected (+1) or not (-1) within a 12 ns time frame.
C: 2 detections (indicate a swap), one from each detector, within the requisite time window (what they call one early and one late). I was not able to find an exact value. They say: "We choose this window conservatively to optimize the entangled state fidelity at the cost of a reduced data rate." (I admit I don't fully understand Figures 3.a and 3.b., which I believe covers this. However, it is not really necessary for understanding what is going on theoretically.)

Note that the time window is relative to distance. C is located away from A & B, so timing is adjusted accordingly. (I.e. the A, B & C markers won't literally be within the same time window.) But the 2 detections at C (in this experiment) ARE within the same time window for each event.

For those pairs that meet the criteria at C, the corresponding A and B photons are entangled. The Bell test results for those 245 A/B pairs can then be calculated using the usual CHSH formula for S. (The C photon pairs are themselves not entangled in a Bell state.)

 

[vanhees71]

If I understand it right, the point is to use entanglement swapping to select (or post-select, which doesn't really matter, if QT is correct, and there's no reason to doubt it, including the result of this experiment!) entangled electron pairs. Without this selection the electron pairs are not entangled at all!

 

[grzegorzsz830402]

doubt: a feeling of uncertainty or lack of conviction.

feeling: an emotional state or reaction.

reason:(synonyms) rationality, logic, reasoning.

Doubt is an emotional reaction, not rational by definition, needs no conscious reason.

Doubt may be justified or not.

Justified: having, done for, or marked by a good or legitimate reason.

Quantum Physics does that to everyone, induce emotional states, regardless if we admit or not. Denial is one of those emotional states.

Why pioneers in that field, had adverse reaction to conclusions, driven from results of those experiments. I can only Imagine, how many times they have went back and forth, with a fine comb to identify where error is. Before settling in two camps, inescapably.

On, one hand you have your own conclusions, that argue with intuitive, gut felt, observable reality, on other hand best solid science and impeccable experiments.

You find yourself between rock and a hard places.

Either you go against intuitive, gut felt notions about reality.

Or, you say to yourself, that you have to be wrong about something that you can not identify, yet its one of those things, that you are most certain of, to be correct.
From that point there is one thing you can be certain off, that if you accept that, you can not be certain of any thing anymore.

There is no good or bad choice. There is price to pay.

Einstein on one side Bell on the other.

We may rationalize, where we have, eventually ended up, but it won't be any more than that.

rationalize: attempt to explain or justify (behaviour or an attitude) with logical reasons, even if these are not appropriate.

For science as a field, (institution) other option was not available. For quantum theory to be verified or falsified, science had only one way, to go about that. (my point of view)

If you find yourself, on one side or the other, you may observe that opposite views are emotionally challenging. If it is the case.
Why?

What is the difference between science, where we have no emotional attachment whatsoever, for any of proposed theories? We are just curious, which one is it?

And science, where we have some emotional investment in one theory or another.

Quantum physics..... like threading on the edge of knowledge, wasn't difficult enough, loaded with emotional trappings.

For most of the time, I have been in favor of quantum theory explanation, choice was emotionally based, now I am curious to study enemies way's:)

I have theory, that we generate our models, (knowledge) and, in some cases, they are applicable only at resolution level, their were driven from (at). So, they work perfectly at low resolution, yet fail at high resolution. Yet if we go about checking our understanding, it looks impeccable, at different level.

It was my attempt, to justify my doubt.

Provoked by exclamation mark, and his unfair demands for emotional states to be reasonable:)

I have gathered all information I needed, about this experiment, I want to thank everyone again for helping me with that.