Hacker News Reader

Comment by Willingham | Hacker News Reader

I’ve always wondered, has there ever been a definitive experiment where one photon hits a slit and on the other side two photons come out, but then when you add a photon observer, it immediately only comes out on one side? Or has the proof always been mathematical rather than a live experiment?

Edit: Thank you all for the responses, it has been very educational. It appears I was misunderstanding the most important aspect of the double slit experiment. A photon is a wave function when unobserved, it literally goes through both slits and creates an interference pattern like how waves in water would. However, when observed at the slit, or at the detector screen, the wave function collapses and only one photon(billiard like particle) will be detected.

View on HN

Double slit experiment did happen and totally reproducible even then photons/electrons are sent by one at a time.

"two photons come out" part makes no sense though. On a target side, there's always single hit after single photon/electron, but distribution of theses hits as if said electron got through both slits and interfered with itself

P.S. the funny thing is - this works on any small thingy, measured up to 2000 atoms-big, as if it's the property of the universe itself

I would love to try this experiment with something basketball sized out in space. Like we build an enormous basketball detector behind a double slit inside an unobservable black box. If thr basketball started acting like a wave I would be sooo freaked out

The largest double-slit projectile I know of is C-60, a soccer-ball shaped molecule of sixty atoms.

https://iopscience.iop.org/article/10.1088/2058-7058/12/11/4

That website's captcha is horrible!

Or amazing.

I got cats with sunglasses

Me too. And I had to click on bird legs to verify o_O

Wiki, citation 19

> The largest entities for which the double-slit experiment has been performed were molecules that each comprised 2000 atoms (whose total mass was 25,000 atomic mass units).[19]

https://en.m.wikipedia.org/wiki/Double-slit_experiment

If the slits are big enough, isn’t that just gravity?

And Why can't gravity serve as an observer?!

The electron/proton entering a slit is affected by gravity!

The entanglement theory would imply that if you build a detector that turns the gravity interaction into a finite piece of data observable by an (amplified) system, then gravity will act as an observer and collapse the waveform when it reaches that point. That's my take on the whole thing... It's almost like information theory. If the information is lost to the sands of time (below noise floor if you will) then the entanglement can continue.

From my recollection of Hawking - it’s the energy of light that disturbs the particle being measured thus confounding the measurement

Presumably a gravity based detector would have similar issues as these particles are affected by gravity (as can be seen around black holes)

I don’t know what it’s called but I read about a proposed experiment to do it in space with macrosized glass beads.

The experiment working on clusters of atoms is news to me and I loved getting to know about that. But the thing that really breaks my mind is the experiment that proves that the behavior depends on the possibility of getting information from which slit the particle went through. So we can rule out the act of measurement itself interfering with the behavior of the particle.

They did it by splitting a beam of particles into a pair of entangled particles and then setting up a way to measure the polarity of one of them after the point in time where it even hits the final screen. If you measure the polarity then, after the other stream of particles from the beam had already had time to make the pattern, the pattern will be two clusters. If you don't, it goes back to an interference pattern.

That one really cemented the notion in my head that this is just how the Universe is and not some local weirdness with particles and measurements.

Hadn’t hear that, that one is wild

That version is called the quantum eraser experiment. It really is brain bending.

I think Sabine explained this social effect few years ago. I know she's a little controversial, but the key thing in the video (as opposed to all other videos about DSE on the internet) was that you don't get "two clusters" actually. They are both statistical parts of a single [non-]interference pattern. "||" is a lie. I'm not in a physics rebel camp and don't prefer Sabine either, but after that I sort of lost trust in the interpretations that can't even get the resulting picture right. I even suspect that showing dumbed results amplifies "wow" effect and monetizes better.

https://www.youtube.com/watch?v=RQv5CVELG3U

This is the video if you're interested. Again, I'm no physicist and don't know if explanations are legit or statistically correct. But that little || trick that all other popsci videos play on you, that's a true concern.

You are correct, the experiment I referred to does not yield two distinct clusters of collisions at all.

You can check it out here.

Summary https://www.stonybrook.edu/laser/_amarch/eraser/index.html

Paper https://www.stonybrook.edu/laser/_amarch/eraser/Walborn.pdf

My fascination with these experiments has never been due neat clusters of impacts, although popsci depictions have clearly tainted my memory.

Are you referring to the double-slit experiment? If so, yes: It has always been an experiment. The experiment came before any theory explaining the behavior AFAIK. https://en.m.wikipedia.org/wiki/Double-slit_experiment

The double slit experiment has been replicated even with fairly hefty molecules.

https://www.nature.com/articles/s41567-019-0663-9

> Here, we report interference of a molecular library of functionalized oligoporphyrins with masses beyond 25,000 Da and consisting of up to 2,000 atoms, by far the heaviest objects shown to exhibit matter-wave interference to date.

It would be awkward to say that the 2000 atom molecule comes out of both sides... but it does, until you look.

The double slit experiment is not a duplication cheat of reality... it's weirder than that.

Am I misunderstanding the significant of the double slit experiment?

I thought the takeaway wasn’t that the particle comes out both sides, the implication is that the behavior of a single particle is the same as the behavior of multiple particles - that is to say, it appears to be an interference pattern, even when there should be no other particles to interferes with the single one.

No you're understanding correctly (I think), the behaviour of a single detected particle depends on all possible paths it could take to get to the detection.

This is fundamental to 100 years of quantum mechanics and underlies most of physics including all semiconductors, materials science, chemistry, lasers, etc. The double slit experiment is just a very good illustration of the principle boiled down to its essentials, which is why it's everywhere in pop-sci. It makes for more accessible story than describing how a hydrogen atom works.

Live experiments have been done and they get really weird: https://en.wikipedia.org/wiki/Delayed-choice_quantum_eraser

What does "one photon hits a slit and on the other side two photons come out" mean?

There is no photon multiplication happening on the double slit.

One photon hits the slit and one photon comes out. It is only if you repeat the experiment many times that you start to see a strange wawe-like pattern in where the photons hit.

It is as if every photon that went through the slit is somehow aware of all other photons that did so too so each photon can choose the (random) position where it hits on the wall behind the slit such that together they look like as if a WAWE went through the slit.

That is (one reason) why they call it "Quantum Weirdness". God is playing dice with us

> It is as if every photon that went through the slit is somehow aware of all other photons that did so too

Why isn't it just that there's a probability density function that describes the aggregate outcomes of a large number of samples from a random process? Why is "memory" involved?

I think because instead of two clusters like you'd expect from random BBs being shot, you get multiple bands like you'd see with interfered waves. Even when shot one at a time.

No. The same photon is aware of all alternative paths it can take, without creating or interacting with any other photon.

There's no photon multiplication, and no "all other photons" changing their path.

There is some inter-photon interaction because they are bosons. But it's not significant enough to impact the multi-slit experiment. And the experiment works exactly the same way if you send only one photon at a time.

I really don’t understand the topic much but this veritasium video is quite eye opening and goes into further depth than any layman explanation I’ve ever seen in the topic: https://youtu.be/qJZ1Ez28C-A?si=6gSQYcJPpaSIt1x1

> I’ve always wondered, has there ever been a definitive experiment where one photon hits a slit and on the other side two photons come out, but then when you add a photon observer, it immediately only comes out on one side? Or has the proof always been mathematical rather than a live experiment?

Only one photon comes out, but it can interfere with itself if it had the possibility of going through either slit.

That nuance aside, the Quantum Eraser Experiment is a real physical experiment that covers what I think you're asking about. If you send photons through double slits in a setup where you can tell which slit the photon went through, you don't get an interference pattern. If you can't tell, you do get the interference pattern.

The experiment came first, all the rest of it came after to try and figure out why the results are as weird as they are.

RE: your edit

You still do not understand what is happening, please READ the article, it shows that the wave function doesn't go through anything and the it certainly doesn't create the interference pattern.

Perhaps the closest thing would be some nuclear decays that spit out two gamma rays of equal energy in opposite directions. I'm struggling to remember which isotope does this.

It’s a cadmium isotope. Super cool technique, I think perturbed angular correlation.

https://en.m.wikipedia.org/wiki/Perturbed_angular_correlatio...

I haven’t used it for my research, but it’s an incredible local probe of electric and magnetic fields in materials. There’s no other technique that I’m aware of that smuggles information about the chemical structure of a single coordination sphere into such clean, distinct emissions. The brief excited state of the isotope after the first emission event and before the second is sensitive to practically everything. It all shows up in the deconvoluted spectra.

Shame nearly all the isotopes that work for this are not ones that are super interesting for modern quantum materials. Perhaps that will change out of necessity.

If you want to learn more of about current theory and experiments I suggest you pick up “Waves in an impossible sea” by Matt Strassler

I think the problem is in insisting on referring to the photon as a particle.

In fact the photon may not actually exist. and I have questions as to what "single photon experiments" are actually measuring. let me explain.

The EM field is not quantized, or at least not quantized at the level of a photon, what we call a photon is the interaction of the EM field with matter, or more precisely with the electron shell of matter. it is the sound of the wave breaking on the shore, not the wave.

Now none of this actually matters as the only method we have of interacting with the EM field is through matter(electrons really) so we can only measure it in photon sized increments.

Just for sake of argument, when looking at it from this angle, EM particles could exist and we lack the ability to emit a single one? But then why would these "single photon" double slit problems not split the particle bunch further?

I honestly don't know, that is my question as well.

However note that we can only perturb the em field in photon sized energy levels, and we can only pick up disturbances of the em field in photon sized bunches as well. Not sure what this implies for how em field energy is accumulated on electrons in order for us to detect it.

Well, the EM field CAN be quantified. Just look up any textbook on quantum field theory. And the quanta of the EM field is called the photon.

But, to "solve" the wave /particle conundrum, I like to think of it as fields all the way down. A "particle" is then a localized and quantisized interaction of said field with another field.

If you think of particles as small billiard balls flying through space on some ballistic trajectory, you'll soon run into all kinds of trouble and the mental model breaks down.

> I think the problem is in insisting on referring to the photon as a particle.

Or in insisting on referring to the electron as a particle.

“We begin by throwing an ultra-microscopic object — perhaps a photon, or an electron, or a neutrino”

> The EM field is not quantized, or at least not quantized at the level of a photon, what we call a photon is the interaction of the EM field with matter, or more precisely with the electron shell of matter.

I don't agree with this. You can absolutely consider a classical (non-quantized) EM field interacting with quantized matter. This semi-classical model can describe the photoelectric effect, but it cannot describe other experimental observations such as sub-poissonian photo-detections / photon anti-bunching.

Reply on news.ycombinator.com