Single-photon double-slit with multiple apparatus?

[aletheia]

TL;DR

Double-slit experiment: has anyone tested single photons with multiple identical apparatuses?

Has anyone tested the double-slit experiment with single photons using multiple identical apparatuses — each with its own emission history —, and then formed the histogram from these “zillions” photons?

Standard QM predicts the outcome as only statistical.

If the histogram formed from different apparatuses do not reproduce the interference pattern, it could point out to a locally influence from the superposition of the previous collapsed photons.

If this hasn't been done, would it be feasible?

 

[Matterwave]

What do you mean by "multiple identical apparatuses"?

Can you clarify your experiment and what you are expecting to see?

 

[anuttarasammyak]

My favorite single photon Young double slit experiment by Hamamatsu Photonics in 1982 is here. https://photonterrace.net/en/photon/duality/

 

[aletheia]

What do you mean by "multiple identical apparatuses"?

Can you clarify your experiment and what you are expecting to see?

My question is pretty straightforward: according to mainstream QM framework, the interference pattern is a statistical consequence of multiple photons — theoretically, so, if you shoot one photon at a time in the same apparatus or if you shoot one photon at an identical apparatus at every single shoot, the histogram formed by the “zillions” photons should not differ, therefore, if the outcome was be different, the history left behind by the photons that had already collapsed into the screen could have influence in the statistical outcome.
What could be it?
Many things (idn).
That’s why I ask here, since I couldn’t find any peer reviewed studies on the double slits testing this hypothesis, I’d like to know if it has already been tested (if so, where?), if not, would it be possible to be done?

 

[PeroK]

My question is pretty straightforward: according to mainstream QM framework, the interference pattern is a statistical consequence of multiple photons

I guess your idea is to do the experiment with one photon. Record the impact location on the detector screen. Replace the detector screen with a new one. Repeat?

If you do a physics degree, you can propose this as an experiment. QM predicts this will make no difference to the overall pattern of impact locations.

 

[Dale]

Double slit tests have been done by many different labs using a variety of apparatuses. Are you just asking if any of those experiments have been done with the same make and model of equipment?

 

[PeterDonis]

according to mainstream QM framework, the interference pattern is a statistical consequence of multiple photons

I'm not sure what you mean by this. The interference pattern is a consequence of the wave function of a single photon. The double slit experiment has been run with a light source of low enough intensity that only one photon is passing through the apparatus at a time, and the points of impact of each individual photon still build up an interference pattern over time.

if you shoot one photon at a time in the same apparatus or if you shoot one photon at an identical apparatus at every single shoot, the histogram formed by the “zillions” photons should not differ

I'm not sure what you mean by "an identical apparatus". If you want to see the interference pattern build up over time when you're doing the experiment one photon at a time, you have to have a single detector screen that records the impacts of all the photons.

 

[PeterDonis]

QM predicts this will make no difference to the overall pattern of impact locations.

But it will be more difficult to obtain the pattern, since you'll have to combine measurements of photon impacts on multiple detector screens.

 

[PeterDonis]

f the outcome was be different, the history left behind by the photons that had already collapsed into the screen could have influence in the statistical outcome

This reasoning would only apply if you ran all the photons through the same apparatus, i.e., recorded all their impacts on the same detector screen. And this experiment has been done, as I said in post #7, and showed the expected interference pattern.

If, OTOH, you run the experiment with a new apparatus and a new detector screen for each photon, you're not even testing whether "the history left behind" by previous photons affects the results for future photons. So I don't see why you'd even want to do this.

 

[PeterDonis]

Here is a paper describing a typical single-photon double slit experiment:

https://sciencedemonstrations.fas.h...-demonstrations/files/single_photon_paper.pdf

This experiment also included other elements testing other QM predictions.

 

[PeterDonis]

Here is a paper describing a typical single-photon double slit experiment

And for historical context, here is the journal containing the paper (from 1909) describing the first such experiment, by G. I. Taylor:

https://archive.org/details/proceedingsofcam15190810camb/page/114/mode/2up

The paper starts on page 114.

 

[aletheia]

I guess your idea is to do the experiment with one photon. Record the impact location on the detector screen. Replace the detector screen with a new one. Repeat?

If you do a physics degree, you can propose this as an experiment. QM predicts this will make no difference to the overall pattern of impact locations.

That’s exactly what I was proposing.
It would just be an experiment to test this prediction.
Most likely the outcome would be the same, but if not there will be something more to think of.
I don’t have an exact idea nor possible explanation other than a possible ontological correlation between the “history” from previous photons and the statistical pattern.
I’m not actually a physicist, I’m a physician, I work at ER, but between shifts I have been self-studying physical ontology for the past 10+ years, reason why I’m more interested in the logical aspects of the experiments rather than the mathematical framework, which I can only take for granted as accurate — I have no reason for not thinking it likewise.

 

[aletheia]

This reasoning would only apply if you ran all the photons through the same apparatus, i.e., recorded all their impacts on the same detector screen. And this experiment has been done, as I said in post #7, and showed the expected interference pattern.

If, OTOH, you run the experiment with a new apparatus and a new detector screen for each photon, you're not even testing whether "the history left behind" by previous photons affects the results for future photons. So I don't see why you'd even want to do this.

Actually, I’m proposing that theoretically it shouldn’t change the outcome if you shoot one photon at a time in the same apparatus or one photon per apparatus.
In the same apparatus we get the interference pattern — with multiple apparatus, theoretically, the histogram from all separated photons should be the same.
If so, nothing new, if not, something new.

 

[aletheia]

But it will be more difficult to obtain the pattern, since you'll have to combine measurements of photon impacts on multiple detector screens.

I have no doubt that it would be terribly difficult — that’s why I ask if it would be feasible at all.
But it surely won’t hurt anyone if it’d be done.
Probably it wouldn’t produce something new but perhaps it would be worth testing.

 

[anuttarasammyak]

If so, nothing new, if not, something new.

Experiments including one in post #3 have shown so and nothing new.
Photons are independent. A photon interferes with only itself.

 

[aletheia]

Double slit tests have been done by many different labs using a variety of apparatuses. Are you just asking if any of those experiments have been done with the same make and model of equipment?

No.
I’m asking if anyone tested using one photon for each new device — a single photon per single apparatus —, and then combine the single outcome of each photon from each apparatus into a histogram to compare if the exact same interference pattern arises statistically in the same way.
Since the interference pattern is predicted statistically, the outcomes shouldn’t vary.
If there would be any difference, this difference would have to be studied.
Nevertheless, any possible explanation could only be formulated, first, if it indeed existed and through what would be its difference.
I wouldn’t expect any difference at all — the idea would, if feasible, understand if the interference pattern only probabilistic or if perhaps each previous photon inside the same apparatus could influence the statistical outcome.
Another way to think about it would be, if possible, “reset” the apparatus before each emission.
To test if the “apparatus history” could influence the statistical outcome is as far as I can go in my reasoning.
My motivation is to find out if anyone has already done it, ‘cause if so I can erase this hypothesis from my reasoning, if feasible, if it would be sound at all as a valid hypothesis to be tested.
I’m used to reasoning with observations + logic. I don’t have deep mathematical understanding from the QM theoretical framework, which I obviously don’t question, although I have self-studied modern physics for a little more than a decade.
I’m not at a position for questioning the validity of QM — I just like to use my brain for things other than evidence based medicine.

 

[aletheia]

A photon interferes with only itself.

Ok, I’m not even also questioning it, however, each photon leaves only one mark at the screen and not two half-marks.
The interference pattern comes after several photons reach the screen.
So, if the photon interferes with the very itself, but leaves one discrete impression, and, if the photon can’t be divided in two, how does it happen?
I’m thinking about the physical ontology of the experiment.
Feynman said that the double slit experiment hides the great mystery of QM.
Modern Physics doesn’t have an ontological framework that can describe the physical reality (as far as I know).
I’d like to state it clearly: I’m not questioning the validity of modern physics theoretical and mathematical framework — I’m only trying to understand the physical ontology, and I do agree that the double slit experiment is excellent for it.
I appreciate your attention and I will take a deeper look in all the kind suggestions I received here so far, but it will take me a couple of days (or weeks).
Cheers.

 

[Roberto Pavani]

I've found this that can be related to the discussion here:
https://www.pnas.org/doi/full/10.1073/pnas.1201271109

 

[Dale]

To test if the “apparatus history” could influence the statistical outcome

OK, so the test you are proposing is actually not a strong test of this. If you want to test for “apparatus history” you would want the maximum possible history effect, not the minimum. So you would want to use a single detector as long as possible.

Then you would bin the collected counts into two or more bins based on collection time. Finally, you would statistically test if the late bins are conditionally independent from the early bins, conditioned on the wavefunction.

This would be the far stronger test for “apparatus history” effects. If the counts are already conditionally independent with the largest possible history effect then removing the history effect could not do anything besides increase the conditional dependence, and that increased conditional dependence could not be attributed to removing the history effect.

Modern Physics doesn’t have an ontological framework that can describe the physical reality

On the contrary, it has several. They are called “interpretations”. Experiments cannot distinguish them. You just pick the one you like and then shout at everyone else.

 

[anuttarasammyak]

The interference pattern comes after several photons reach the screen.
So, if the photon interferes with the very itself, but leaves one discrete impression, and, if the photon can’t be divided in two, how does it happen?
I’m thinking about the physical ontology of the experiment.

Probability is essential in QM. A same exact condition produces not unique but varied results which are under law of probability.

In classical statistic mechanics probability comes from many particle ensemble, but QM applies probability to single particle.

QM tells us this probability and it is visualized by accumulated many- time independent experiments as well as by many particles at a time experiments.

A particle is associated with space expansion entity usually called wave function which describes the probability. Divided wave function meet again and interfere after one pass the right and the other pass the left slits.

 

[PeterDonis]

Actually, I’m proposing that theoretically it shouldn’t change the outcome if you shoot one photon at a time in the same apparatus or one photon per apparatus.

That is correct, QM predicts the same outcome with both methods (though in practical terms doing it the second way, a new detector screen for each photon and then combining the results in a single histogram, is so much more cumbersome that I would not expect anyone to be willing to spend the resources to do it).

 

[PeterDonis]

if perhaps each previous photon inside the same apparatus could influence the statistical outcome.

But we've already tested this--the "standard" double slit at low intensity, where one photon at a time is inside the same apparatus, has been done, and confirms the QM prediction, which is based on no influence of previous photons on any given photon.

This is why I (and apparently I'm not the only one) am confused about why you would even care about doing the experiment the other way, with a new apparatus for each photon.

 

[PeterDonis]

it surely won’t hurt anyone if it’d be done.

But it would cost time and effort that could be spent on something else. And since, as I've already pointed out, the hypothesis you say you want to test--whether previous photons leave any "imprint" on the apparatus that could affect the next photon--has already been tested and ruled out, it seems reasonable that nobody would want to spend time and effort on what you propose.

 

[PeterDonis]

if the photon interferes with the very itself, but leaves one discrete impression, and, if the photon can’t be divided in two, how does it happen?

Mathematically, the way QM arrives at its prediction is clear: each individual photon's wave function contains inteference terms, and the wave function is what predicts the probabilities of a photon hitting the detector as a function of position on the detector.

I’m thinking about the physical ontology of the experiment.

That depends on which QM interpretation you adopt. Discussion of QM interpretations would be off topic for this thread; it would need to be discussed in a separate thread in the QM interpretations subforum.