Double slit experiment with observer

[DoofWarrior]

TL;DR Summary

Academic experiment and general knowledge about double slit experiment

Hi ! This is about the well known experiment using small particles like electrons or photons
- Light/electron beam passes through two slits
- We observe a wave interference pattern on the wall after the slits
https://en.wikipedia.org/wiki/Double-slit_experiment#/media/File:Double-slit.svg

Now let's say we add an observer. I can't produce this experiment at home, but theory say that we won't observe an interference schema anymore. Few questions here
- In case the observer can see only one electron/photon among all passing through the slid, does it cancel all interference schema for all photons/electrons ?
- What type of detectors are usually used as observer for this experiment ? How can we be sure that observer does not interact with the photon/electron ?

Hope the question is clear, I'm an engineer so I'm not experienced with quantum experiments.

Kind regards

 

[PeroK]

TL;DR Summary: Academic experiment and general knowledge about double slit experiment

Hi ! This is about the well known experiment using small particles like electrons or photons
- Light/electron beam passes through two slits
- We observe a wave interference pattern on the wall after the slits
https://en.wikipedia.org/wiki/Double-slit_experiment#/media/File:Double-slit.svg

Now let's say we add an observer. I can't produce this experiment at home, but theory say that we won't observe an interference schema anymore. Few questions here
- In case the observer can see only one electron/photon among all passing through the slid, does it cancel all interference schema for all photons/electrons ?

No. You will end up with a combination of intereference patterns if you have which-way information about some but not all particles.

- What type of detectors are usually used as observer for this experiment ? How can we be sure that observer does not interact with the photon/electron ?

You can't detect a photon without absorbing it. And, in principle, you cannot observe an electron without interacting with it. The simplest test of the double slit experiment is:

Run the experiment with one slit closed and note the pattern.
Run the experiment with the other slit closed and note the pattern.
Run the experiment with both slits open and note the pattern.

The key point is that the third pattern is not the sum of the other two. And that is quantum interference and essentially the heart of QM.

Note that there some clever ways to gain which way information about a particle by observing not the particle itself, but using an entagled pair of particles. See, for example:

https://en.wikipedia.org/wiki/Spontaneous_parametric_down-conversion

 

[DrChinese]

- What type of detectors are usually used as observer for this experiment ? How can we be sure that observer does not interact with the photon/electron ?

Answering this portion specifically, and without intending to contradict any part of PeroK's correct answer:

It is possible to obtain which path information in a double slit setup in which the photons are NOT absorbed as per usual. And this can be done in conjunction with the scenario in which single photons are in the apparatus, so that you are certain there is only self-interference* patterns at play. Further, in this configuration: you don't actually learn the which path information, it is enough that you COULD learn that information for the patterns to change.

The technique is as follows: Place a polarizer over each of the 2 slits (2 polarizers total) that can be oriented either parallel or perpendicular relative to each other. When the polarizers are parallel, there WILL be a traditional interference pattern. When the polarizers are perpendicular (orthogonal), there will NOT be a traditional interference pattern. And if you vary the polarizer settings between 0 and 90 degrees, you get a mixture of the two patterns. The intensity of the pattern is not affected by the polarizer settings, only the pattern formed. So the "detector" in this case is the relative polarizer settings.

Young’s double-slit experiment with single photons and quantum eraser*Using the common meaning of "self-interference" in connection with the double slit experiment.

 

[kered rettop]

Note that there some clever ways to gain which way information about a particle by observing not the particle itself, but using an entagled pair of particles. See, for example:

https://en.wikipedia.org/wiki/Spontaneous_parametric_down-conversion

Unfortunately, the most well-known - and also the most widely misunderstood - use of SPDC is the DCQE by Kim et al. For me, the authors muddy the waters when they describe the down conversion as if it happens at a single atom. This seems to imply that the emission is highly located - which would imply that there is very clear which-slit information. However, the emission must be delocalised over both slits as well or it would not be possible to extract the interference patterns. I'm pretty sure I know how the two are reconciled but I don't want to discuss what will undoubtedly be decried as my own personal theory I'm just saying that it's more complicated, and more confusing than just having a point source emitting a pair of photons. If anything, the delocation over both slits is harder to understand than all the rest of the experiment put together but it tends to be glossed over in the excitement of interpreting the experimental results.

 

[PeterDonis]

I'm pretty sure I know how the two are reconciled but I don't want to discuss what will undoubtedly be decried as my own personal theory

If it has a basis in the literature, it isn't a personal theory. Is there literature on this question? It certainly seems like something that would have had some discussion.

 

[kered rettop]

If it has a basis in the literature, it isn't a personal theory. Is there literature on this question? It certainly seems like something that would have had some discussion

I imagine there is, but I don't know what. Surely one of the mentors will know? I can't imagine that I'm the first person to have noticed.

 

[vanhees71]

Unfortunately, the most well-known - and also the most widely misunderstood - use of SPDC is the DCQE by Kim et al. For me, the authors muddy the waters when they describe the down conversion as if it happens at a single atom. This seems to imply that the emission is highly located - which would imply that there is very clear which-slit information. However, the emission must be delocalised over both slits as well or it would not be possible to extract the interference patterns. I'm pretty sure I know how the two are reconciled but I don't want to discuss what will undoubtedly be decried as my own personal theory I'm just saying that it's more complicated, and more confusing than just having a point source emitting a pair of photons. If anything, the delocation over both slits is harder to understand than all the rest of the experiment put together but it tends to be glossed over in the excitement of interpreting the experimental results.

I do not know a single source in the scientific physics literature that claims that SPDC were not due to a collective effect involving the birefringent crystal as a whole. The quantum-eraser experiment by Kim et al is a real experiment and in full accordance with the predictions of QED (as applied in the quantum-optics literature which includes the description of optical elements also for quantum states of light, including single- or multi-photon Fock states).

It's also well understood that you get an interference pattern with the double slit sufficiently far from a "point source", so that the incoming wave well covers both slits, as well as with the screen sufficiently far from the double slit, such that the waves going out from both slits can overlap and interfere at the screen's location.

That's not different from diffraction for classical electromagnetic waves. It's for sure not a personal theory but can be found in any reasonable textbook about optics.

 

[DrChinese]

Unfortunately, the most well-known - and also the most widely misunderstood - use of SPDC is the DCQE by Kim et al. For me, the authors muddy the waters when they describe the down conversion as if it happens at a single atom. This seems to imply that the emission is highly located - which would imply that there is very clear which-slit information. However, the emission must be delocalised over both slits as well or it would not be possible to extract the interference patterns. I'm pretty sure I know how the two are reconciled but I don't want to discuss what will undoubtedly be decried as my own personal theory I'm just saying that it's more complicated, and more confusing than just having a point source emitting a pair of photons. If anything, the delocation over both slits is harder to understand than all the rest of the experiment put together but it tends to be glossed over in the excitement of interpreting the experimental results.

I would point out that SPDC is only occasionally used for the DCQE format. There are literally hundreds of other entanglement applications for down conversion. But I would agree that DCQE is one of the most complex and difficult to understand or discuss. I try to stay away for exactly that reason. There are plenty of other delayed choice experiments that demonstrate that concept much more readily than the quantum eraser.

Entangled pairs are not emitted from what you might call a well-localized point (agreeing with @vanhees71). If you reduce the size of the crystal, you simply get fewer useful pairs. More importantly, you cannot use information from one photon of an entangled pair to predict which slit information for the other, if you also expect there to be visible self-interference on one side. Essentially, entanglement and self-interference are mutually exclusive (that's an over-simplification of course).

You can, of course, force one side to become coherent. That would produce the interference pattern on that side. But that photon will no longer be entangled on the desired basis with its partner.

 

[vanhees71]

In my habilitation colloquium (where I had to give a lecture-like talk on a topic that's not related to my own reseearch) I discussed another delayed-choice quantum-eraser experiment by Walborn et al (that's just by chance, I could as well have chosen the Kim et al experiment; the principles are the same). Maybe the transparencies (English version) are of some value since I really tried to keep it to the minimum of formalism:

https://itp.uni-frankfurt.de/~hees/publ/habil-coll-talk-en.pdf

If find the delayed-choice experiments among the most fascinating applications of quantum entanglement.

 

[kered rettop]

I do not know a single source in the scientific physics literature that claims that SPDC were not due to a collective effect involving the birefringent crystal as a whole.

The quantum-eraser experiment by Kim et al is a real experiment and in full accordance with the predictions of QED ...

The Kim et al DCQE paper describes two processes to create entangled pairs:
1 "A pair of entangled photons, photon 1 and photon 2, is then emitted from either atom A or atom B by atomic cascade decay".
2 "A pair of 702.2nm orthogonally polarized signal-idler photon is generated either from A or B region."

In both cases the process is described as either/or. Obviously the first localises the emission to an individual atom but the second also localises it - to one or other of the two regions. Emission from a single region categorically rules out two-slit interference just as surely as emission from a single atom does.

 

[vanhees71]

Which paper are you referring to? Of course atomic cascades where the first sources used by Clauser et al to perform the first Bell tests. The sources were indeed atomic transitions of single atoms.

 

[kered rettop]

I would point out that SPDC is only occasionally used for the DCQE format. There are literally hundreds of other entanglement applications for down conversion. But I would agree that DCQE is one of the most complex and difficult to understand or discuss. I try to stay away for exactly that reason. There are plenty of other delayed choice experiments that demonstrate that concept much more readily than the quantum eraser.

Entangled pairs are not emitted from what you might call a well-localized point (agreeing with @vanhees71). If you reduce the size of the crystal, you simply get fewer useful pairs. More importantly, you cannot use information from one photon of an entangled pair to predict which slit information for the other, if you also expect there to be visible self-interference on one side. Essentially, entanglement and self-interference are mutually exclusive (that's an over-simplification of course).

You can, of course, force one side to become coherent. That would produce the interference pattern on that side. But that photon will no longer be entangled on the desired basis with its partner.

Exactly. I have no problem at all with delocalised emission even if the entity emitted is an entangled pair. What I have a problem with is describing it as emission "either from A or B".

 

[kered rettop]

Which paper are you referring to? Of course atomic cascades where the first sources used by Clauser et al to perform the first Bell tests. The sources were indeed atomic transitions of single atoms.

Well, I was thinking of the Aspect experiments, but I'm sure there were many more. Does it make any difference?

 

[kered rettop]

Unfortunately, the most well-known - and also the most widely misunderstood - use of SPDC is the DCQE by Kim et al. For me, the authors muddy the waters when they describe the down conversion as if it happens at a single atom. This seems to imply that the emission is highly located - which would imply that there is very clear which-slit information. However, the emission must be delocalised over both slits as well or it would not be possible to extract the interference patterns. I'm pretty sure I know how the two are reconciled but I don't want to discuss what will undoubtedly be decried as my own personal theory I'm just saying that it's more complicated, and more confusing than just having a point source emitting a pair of photons. If anything, the delocation over both slits is harder to understand than all the rest of the experiment put together but it tends to be glossed over in the excitement of interpreting the experimental results.

Correction - they do mention emission from a single atom, but that's in a paragraph explaining the background. They describe the emission in their experiment as coming "either from A or B region", which still implies that it is localised.

 

[kered rettop]

In my habilitation colloquium (where I had to give a lecture-like talk on a topic that's not related to my own reseearch) I discussed another delayed-choice quantum-eraser experiment by Walborn et al (that's just by chance, I could as well have chosen the Kim et al experiment; the principles are the same). Maybe the transparencies (English version) are of some value since I really tried to keep it to the minimum of formalism:

https://itp.uni-frankfurt.de/~hees/publ/habil-coll-talk-en.pdf

If find the delayed-choice experiments among the most fascinating applications of quantum entanglement.

I used to, if only because it gets misunderstood so much. Thanks for the slide show!