Double slit barrier questions

[Agramenauer]

I have some questions about double slit experiment.


1) Sometimes photons or electrons should bounce off the initial barrier and nothing should be detected in the plate.
¿Is this correct? i find a video saying that NOTHING is detected in the barrier, only in the plate.

2) If we put 1 detector in one slit, then it is posible to KNOW the position of an electron passing trough the other slit without interact with it. There is no PHYSICAL INTERACTION but wave function collapses anyway, so we are not speaking of a material phenomena.
¿Is this correct?

3) In the delayed choice quantum eraser the photon is entangled before it hits the plate. The entangled photon follows a path that sometimes erases the "which path" information resulting in the well known retrocausal observations.

My question is: if both photons follow a path that destroy the "Which-path" information then we will observe an interference pattern IN ALL DETECTORS. What will take place if the entangled photon travels trough space, then bounce off a mirror and returns 5 minutes later? The experimenter have had time to observe the interference pattern in the original plate and change the experiment configuration to preserve "which-path" information.
¿Is this possible?

 

[Matterwave]

1) Nothing is detected at the barrier because we don't put any detectors at the barrier (for the regular double slit experiment). The ones that bounce off the barrier and back towards the emitter simply go undetected.

2) "Physical interaction" is hard to define in a quantum mechanical sense. The wave function certainly "interacts" with the detector since the presence of the detector has a measurable effect on the wave function (i.e. "wave function collapse"). Whether the particle itself does or not is...up to interpretation.

3) I am not familiar with the delayed choice quantum eraser. But in quantum mechanics, whether an experimenter "chooses to look at the data" does not affect the physical events. Otherwise you would be saying consciousness causes collapse is the only viable interpretation of QM.

 

[Agramenauer]

Thanks

thank you for your clarification, I'm not an expert and I'm full of misconceptions about quantum mechanics.

About the point 2 I still have a doubt, if there is an interaction it could not be physical because detector is unaffected, obviously it's not related to human observation because detectors producing collapse of the wave function are just material objects, so i interpret that this interaction is not taking place at a physical level but at some other unknown level where "possibility of interaction" is the cause and the collapse of the "wave function" the effect.

Point 3 is just curiosity on the apparent retrocausal effects observed in the delayed choice quantum eraser. My question is in fact "Can we produce an interference pattern and have an entangled photon or electron?" it's all.

I think i mistake and i send an answer to your comment as a "report"... well I'm new here i just make some stupid things, sorry.

 

[Cthugha]

3) In the delayed choice quantum eraser the photon is entangled before it hits the plate. The entangled photon follows a path that sometimes erases the "which path" information resulting in the well known retrocausal observations.

My question is: if both photons follow a path that destroy the "Which-path" information then we will observe an interference pattern IN ALL DETECTORS. What will take place if the entangled photon travels trough space, then bounce off a mirror and returns 5 minutes later? The experimenter have had time to observe the interference pattern in the original plate and change the experiment configuration to preserve "which-path" information.
¿Is this possible?

Two points:

I) One does not need retrocausality to explain delayed choice experiments.

II) You will not see an interference pattern in all detectors. In fact, you will not see an interference pattern on any detector directly. If you have a look at the standard experiments (or the wikipedia article discussing them), you will find that the interference patterns show up in the coincidence count patterns only. So you need the information from both detectors before you get some pattern. Therefore it does not matter, whether you change the setup in one arm after the detection in the other arm of the setup already happened. Loosely speaking, the whole pattern is the superposition of many shifted interference patterns and the coincidence counting acts similar to a filter allowing you to single out one of them if there is no which way information. The temporal order of the detections is irrelevant.

 

[Nugatory]

so i interpret

that's a fortuitous choice of words
Google for "quantum mechanics interpretation" for some context.

that this interaction is not taking place at a physical level but at some other unknown level where "possibility of interaction" is the cause and the collapse of the "wave function" the effect.

Wave function collapse is not part of the formalism of quantum mechanics; it doesn't appear in the math. It's just a mental picture of what might be going on in an interaction, and we use it when it helps us reason about a problem. There's no reason to try to explain things in terms of collapse if trying to do so just adds complications - and that's what's happening here.

Thus, it never makes sense to talk of collapse as an effect that has a cause. An effect is always something that physically happens; the pattern on the screen is an effect, caused by light hitting different parts of the screen with different probabilities.

 

[Agramenauer]

... II) You will not see an interference pattern in all detectors. In fact, you will not see an interference pattern on any detector directly. ...

Ok, this is the point where I'm lost. I thought that the superposition of different states could be canceled if the experiment is modified to ALWAYS destroy the "Which-path" information. If the experiment ALWAYS destroy the "which-path" information then there is no superposition of states because you will always SEE and interference (as observed in a normal double slit experiment).

Imagine we repeat de delayed choice quantum eraser experiment with ONLY two detectors, one for the photon (D0), and one for the entangled photon (D1). This experiment ALWAYS destroy de "which-path" information so we should observe always an interference patter in D0 and an interference pattern in D1.

In this case you don't need a coincidence count to obtain the interference pattern because there is only one possible outcome that is interference. My problem is what if the entangled photon travel trough space for 5 minutes before returning to the experimenter's room, allowing the experimenter to change completely the experiment?

Probably I have not understand something important, thanks for your patience.

 

[DrChinese]

Ok, this is the point where I'm lost. I thought that the superposition of different states could be canceled if the experiment is modified to ALWAYS destroy the "Which-path" information. If the experiment ALWAYS destroy the "which-path" information then there is no superposition of states because you will always SEE and interference (as observed in a normal double slit experiment).

The quantum eraser is itself a complex setup and although there is often the "interference pattern" component to it, it does not easily resolve itself in the manner you describe.

The answer is that an interference pattern ONLY appears when doing coincidence counting for determining a special subset between D0 and D1. The particular setup you describe, where you see an interference pattern at BOTH D0 and D1, is not such a subset and never occurs.

The reason is somewhat complicated, but you can follow this reference for more information. See Zeilinger, page 290, figure 2, there is no direct interference pattern for entangled photons:

Experiment and the foundations of quantum physics (1999)

 

[Cthugha]

If the experiment ALWAYS destroy the "which-path" information then there is no superposition of states because you will always SEE and interference (as observed in a normal double slit experiment).

Imagine we repeat de delayed choice quantum eraser experiment with ONLY two detectors, one for the photon (D0), and one for the entangled photon (D1). This experiment ALWAYS destroy de "which-path" information so we should observe always an interference patter in D0 and an interference pattern in D1.

This is a bit oversimplifying what is happening in the usual ouble slit. Having no which-way information does not assure that you will see an interference pattern. The pattern you see in a usual double slit will depend on two things:

1) Is there which-way information or not.
2) The relative phase between the two paths. Whether you see constructive or destructive interference at some point depends on the optical path length difference from the same initial spot via the two slits to the same final spot. If you change the difference, e.g. by inserting a glass plate at one of the slits, you change the pattern and the peaks and dips on the screen will move. You can check this at home and it is pretty cool. :)

The same effect will occur if you move the light source around. This changes the relative distance covered and therefore also the interference pattern seen. This means that having which way information is not the only way to see no interference pattern. If you have a light source which is so large that the light emitted from each point of the source surface will lead to a different interference pattern, you will end up with a superposition of all of these patterns which looks just like no pattern at all. The latter scenario is important for DCQE and my next reply.

In this case you don't need a coincidence count to obtain the interference pattern because there is only one possible outcome that is interference. My problem is what if the entangled photon travel trough space for 5 minutes before returning to the experimenter's room, allowing the experimenter to change completely the experiment?

It is a prerequisite for the kind of entangled photons used in DCQE to suffer from the problem mentioned above. Entangled photons are emitted in a wide range of angles and each emission angle will correspond to a different interference pattern. So you end up with no pattern at all. However, as you only have a large light source, but no which-way information, you can recover the pattern by just picking a smaller range of angles. This is what is done by coincidence counting. You use a narrow detector which just picks a narrow range of emission angles.

Of course you could also directly filter the light going to the double slits and directly see an interference pattern, e.g. by placing the non-linear crystal that creates the entangle photons further away from the double slit. However, one can show that this breaks entanglement. The few photons still arriving at the slits will not be able to violate Bell inequalities and are therefore not entangled anymore. Therefore the delayed choice on the other photon will not matter.

 

[Agramenauer]

Ok, I see that there is no way to generate an entangled photon without destroy a direct observation of interference pattern. But interference still persists as a "second order" phenomena only appreciable trough a coincidence counting.

 

[DrChinese]

Ok, I see that there is no way to generate an entangled photon without destroy a direct observation of interference pattern. But interference still persists as a "second order" phenomena only appreciable trough a coincidence counting.

You've got it! And by the way, welcome to PhysicsForums.

 

[Cthugha]

Ok, I see that there is no way to generate an entangled photon without destroy a direct observation of interference pattern. But interference still persists as a "second order" phenomena only appreciable trough a coincidence counting.

Wow. This is about the first time that I see someone understanding what is going on in such a short amount of time!

 

[Agramenauer]

Thanks, if I have a more doubts i will ask here, it's an excellent place to solve misconceptions and fallacies.