It's interesting to see the lack of HN consensus on what's going on, and even lack of satisfaction with explanations on normally-lauded YT channels such as 3B1B.
This only works if you forget that light is quantized. The place the weirdness really comes in is that if you shoot single photons at a time, you observe the same effects.
You can consider a wave passing through a filter as a sum of two orthogonal waves, rSin(θ) + rCos(θ), θ being the angle between the light-wave and the filtered angle, r being the amplitude of the wave.
One wave gets eliminated, and whatever exits exits at the only angle it can, the angle orthogonal to the filtered angle.
The part that still needs explaining is how the magnitude can be reduced. IIRC single photons can be polarized by these things, and AFAIK their wavelength is not changed so their energy is unchanged as well.
I have always thought (how I got there I don't know) that the polarizer did something weird like rotate the photon to the correct phase angle AND passed it through with probability based on the angle / or didn't let it pass. This would give a similar reduction in intensity for a desktop experiment while having similar but different details when looking at the photon level. Is this correct?
The polarization of a single photon is a quantum property, so it's essentially a probability distribution. Passing a photon through a polarizer modifies the probability distribution such that "more perpendicular" polarizations are now less likely and "more parallel" ones more likely. (Polarization is a superposition (ie. a linear combination) of two orthogonal basis vectors, and a polarizer projects a polarization vector onto one of the basis vectors.)
The explanation given is at the level of classical electromagnetism, and is sufficient to explain and predict experiments with regular light.
If you want an explanation and prediction of what happens at the level of single photons, you need more structure from the theory of quantum optics. But briefly the filter at angle T does a measurement on the photon in the basis {T, T+pi/2}, and you end up seeing the photon on the other side of the filter only with whatever probability the photon has for being in state |T>, as opposed to state |T+pi/2>.
So, filters are inherently destructive and fewer and fewer photons pass through each subsequent filter. And a photon that makes it through a filter at angle T, now has a new state |T>.
The thing is that there is no actual evidence that “single photons” exist in the electromagnetic field. That last bit is important: photons are a mathematical shorthand for dealing with emission and absorption by atomic orbitals.
They’re not an explanation for continuous waves in between, but the mathematics largely works anyway because photons are very similar to how one would do a Monte Carlo numerical simulations of continuous wave phenomena.
This has resulted in an unbelievable amount of confusion…
If you make the energy in a mode of the EM field continuous, you get the famous 'ultraviolet catastrophe.' This holds whether the matter involved has a continuous emission spectrum or not.
I'm not so sure: 'black body' are made of atoms, atoms emits lights only on certain frequency due to the way electrons orbitals are structured, with discrete energy levels, but I don't see how this is related to EM fields themselves..
Free electrons also emit blackbody radiation. Photon quantization has nothing to do with the energy levels of atoms. It is a very fundamental property of the universe - you can tell because the only relevant constant is planck's constant.
I didn't say that light isn't quantized, I just said that for me blackbody radiation of ordinary matter isn't a convincing proof that light is quantized..
As for free electrons emitting blackbody radiations can you give me a reference?
If you're talking about a wiggle (https://en.wikipedia.org/wiki/Wiggler_(synchrotron)) I don't see how this is related to the issue.
So there are two things we should identify separately that go under the name "quantization".
One is that when we send out a single photon worth of energy, even in an isotropic or other multi-path emission pattern, we only ever seem to detect a single photon. You can prove this with a half mirror and two detectors. This is kind of weird - not clear how this works, and the core question driving different quantum interpretations.
Second is that the energy level of a single photon detection is based on the constraint that Et=h (t is period), also written E=hf. This is not related to atomic properties of the detector. Why? Equation does not depend on any variable of the detector.
Same goes for Px=h. (Momentum/wavelength relationship.) You can prove that one with two razor blades close together. Same relationship, but along space axes instead of time.
Also for uncertainty principle - reinterpret EPtx as the variance of their distributions.
All of this is some sort of fundamental constraint of the universe around quantities of dimension force*time*distance (action). Not related to atoms per se. Not that I understand why this constraint exists. I know some ways to think about it mathematically, but not intuitively.
Fundamentally, you have to believe in nonsense such as kilometer-sized particles that are also infinitesimal if your point of view is that photons are particles.
Many of the "mysteries" of Quantum Mechanics vanish if keep in the back of your mind that QM is just a bag of mathematical tricks that "work" in specific scenarios to solve practical problems, but aren't a model of "what actually occurs".
Practically all of it is restricted to statistical predictions related to matter. Remove matter or remove statistics and then the models are no longer valid.
I'm not making this up. Go dig up any of the original QM papers and these assumptions and/or constraints are called out in the first few paragraphs.
> Many of the "mysteries" of Quantum Mechanics vanish if keep in the back of your mind that QM is just a bag of mathematical tricks that "work" in specific scenarios to solve practical problems, but aren't a model of "what actually occurs".
They also tend to start vanishing once you get to studying Quantum Field Theory, which based on your comments I'm not convinced you have. Field Quanta very much "exist" (not getting into a debate on platonism here), and the debate between paticle/wave doesn't.
Yes, some crazy YouTube video logger named Schrödinger.
I wouldn't trust him either. What does he know?
You're basically arguing that the assumptions of QM theory that is in many of the older papers for some reason stopped mattering. That they've been invalidated... somehow, instead of simply being dropped for brevity in newer papers, assuming that everyone knows them already.
All I'm saying is that if you go back to the original version of the theory, they were quite clear about the scope: a statistical theory that applies to ensembles of atoms.
In theoretical Physics, especially when dealing with phenomena outside of normal human experience, you have to be very careful with your assumptions otherwise you can make yourself believe all sorts of nonsense.
Such as wave function collapse. Or spooky action at a distance. Both of which stem from a misunderstanding of the limits of the scope of a statistical theory that only applies to some types of physical objects.
This point of view is interesting. It's also inconsistent with the actual physics. (I'm calling the mainstream view of basic QM 'actual physics' because it has been checked obsessively against experiment for 100 years, not because mainstreamy ideas are automagically correct.)
> The thing is that there is no actual evidence that “single photons” exist in the electromagnetic field. That last bit is important: photons are a mathematical shorthand for dealing with emission and absorption by atomic orbitals
That is not true, unless I completely misunderstood what you are trying to say. Photons exist in contexts without any atom involved, e.g. particle/antiparticle pair creation or annihilation. The gamma photons emitted during radioactive decay have nothing to do with orbitals either.
Besides, single photons have been observed. It was an important part of the Young slits experiment. Alain Aspect’s experiments also depend on the existence of single photon pairs. How would they work if photons were not a thing?
The double-slit experiment is no proof for the existence of single photons. It can be fully explained (at any light intensity) by assuming a classical electric field. The "clicks" of a "single-photon detector", such as a photomultiplier, only require the existence of quantized charges (electrons).
No, not really. If the amplitude of the EM wave is enough to make an electron jump, all the electrons with an available energy level would jump across the whole detector. To have a single detection in a given region of the detector, you need a localised energy transfer. At this point we’re just reinventing photons.
Or is it another case of "wave function collapse"? Remember that diffraction grating work too, and they rely on the wave nature of light. It is sometimes said (and can be modeled) that light takes every path from source to destination and the diffraction grating causes many paths to cancel out so it appears to go specific ways. Mirror reflection can also be modeled this way, and making periodic scratches will turn it into a diffraction grating. In the end, a photon detection is like a "measurement" that collapses the reflected wave function.
Photons are never observed "in flight" they are only detected when they hit something. In between they may as well BE a wave function.
This is not how quantum mechanics works. One standard textbook example is to calculate the dynamics of a two-level atom that is perturbed by an oscillating (classical) electric field. The result is that there is a certain probability per unit time for the atom to transition to the excited state. In a solid, the two levels are replaced by a band structure, and the excitation leads to the generation of an electron-hole pair. The calculation then immediately gives quantized absorption events that follow a Poisson distribution. No photons required.
One could equivalently say that the thing that really exists is the photon, that is, light is a particle and not a wave, and the weird stuff comes from how photons move through space. That was the view of Feynman at least, espoused in his pop science book "QED: The Strange Theory of Light and Matter"
Actually I just checked, on page 15:
> I want to emphasize that light comes in this form - particles. It is very important to know that light behaves like particles, especially for those of you who have gone to school, where you were probably told something about light behaving like waves.
Anyway I say "equivalently" because this is just another way to conceptualize the same theory (and make the same predictions), it's not new physics
You're suggesting that the EM field doesn't need to be quantized if everything it interacts with is quantized, so that the non-quantized parts can never be brought to bear and observed?
I think that although light is a quantum phenomenon and we can detect single photons, many people overlook that photons still behave similarly to Maxwell's (wave) equations. In particular, the average behavior is that of the electromagnetic wave (up to some extremes like extremely high energies). The 'bullet' model people think of when the word photon (particle) is mentioned is inadequate. This becomes clear in field theory (QED and QFT), where there's a more complete description based of this phenomena solely based on field (wavelike) behavior. It's believe any (small) system follows QFT exactly (there's still uncertainty around gravity).
The exact nature of the relationship of quantumness and fields (i.e. how the single-particle behavior arises from QFT) is still unclear, which is why there are many competing interpretations of quantum mechanics. In the Copenhagen interpretation, which is the most "easy" one, the behavior of photons is just (almost) that of Maxwell's equations, on average, s.t. a single photon will be measured with probability equal to the average light intensity anywhere (they are said to "collapse" at the moment of measurement, which is surely a simplification of a more complete underlying theory).
The magnitue drop is reasonably simple to understand in terms of fields. The oscillating optical field might be less effective at exciting material oscillations in the middle filter due to a mismatch in polarization, but it still does so at the same frequency. You can think of it as multiple photons (incoming field) collectively exciting the same electron on the same frequency but with reduced efficiency. The electron then re-emits fewer photons (outgoing field) of the same wavelength, leading to a lower light intensity detected after the filter.
I don't think photons are absorbed and re-emitted by electrons. At least that argument does not hold when discussing light slowing down in glass or water. Light is affected by the electromagnetic field of the material it is going trough, is slowed down, or absorbed based on some of its property, but photons that go trough are going trough without collision. Photons that get absorbed and re-emitted are scattered in all directions, and are mostly lost. You would not see a consistent image trough a polarising sunglass, if the photons you were seeing were re-emitted photons.
I think using the term "re-emit" is a mistake on my part. I fully agree that the photons are not absorbed and re-emitted.
However, their speed of propagation in the material, and their properties can be altered by the interaction with dipole moments of the material. Therefore, photons do indeed "go trough without collision" but not without interaction. The dipoles in a dielectric material such as glass need to take the energy to oscillate from the field, and then they give it back to the field by radiating it out again.
I agree that this is not "re-emission" in the absorption/emission diagram sense, and I understand that the field in a material cannot be separated into the material portion and the vacuum portion, since it is both at any given time, but what I have tried to outline in the previous comment is still a useful representation of the role of the material which the original article chose to leave out. I should have used "radiate" and "dipole" instead of "emit" and "electron" to make it a bit clearer.
At the single photon level, the photon that goes into the polarizer and the photon that goes out of the polarizer are not the same photon, so it's not right to say that a photon is changed or transformed.
For a photon coming in at 45 degrees to the polarization angle, the probability that another photon will be emitted is sin(45deg) =~ 70% and the probability that it will be absorbed is 1/sin(45deg) =~ 30%.
(This is also a simplification; polarization angle is similarly quantum in nature, and I have assumed it to be collapsed here)
Easier to think about the polarizer's effect on a non-quantum EM wave, and then understand that our probability of observing a photon is based on the magnitude of the resultant EM wave.
> single photons can be polarized by these things, and AFAIK their wavelength is not changed so their energy is unchanged as well.
This is correct. The linked article doesn't explain how the single-photon behaves. It's the same issue that the two-slit experiment has if you try to ignore quantum behavior: You still can't say which slit a single photon passes through.
> These results can be verified by performing the experiment with an actual light meter — the meter should show about twice as strong a reading in the Figure 1 arrangement as it does in the Figure 3 arrangement.
Quantum mechanics predicts that the difference is a factor of 4, not a factor of 2.
While this explanation is very nice, it still does not actually explain what is happening on a material level.
The light does not "pass" through the middle filter, but it excites oscillations in the material, which effectively re-emits the light with different properties. The incoming light polarized at 0° induces oscillations in electrons which are "bound to a rail" in the material, which allows them to only oscillate in the direction of 45° (and all oscillations in the direction of -45° are absorbed). Therefore, a portion of the incoming field essentially gets re-emitted by the middle filter linearly polarized at 45°.
This representation is much less helpful if you think of the light in terms of individual photons rather than fields of course, but it is not worse than the article in this regard either.
It’s both and neither since photons are particles and waves and focusing on one or the other to build intuition can be useful in some cases and not other.
How light actually behaves is probably beyond the ability of human cognition (since so much happens in a billionth of a second)
> How light actually behaves is probably beyond the ability of human cognition (since so much happens in a billionth of a second)
This is not remotely true. The behavior of light is very well understood and relatively simple to model compared to other, less linear physical processes.
We have some great models but what actually happens doesn't quite fit into any one model. For example all photons are constantly redshifted as they travel through space because space is expanding. That’s not really relevant on human timescales but it is an effect that takes place between your monitor and your eyes.
When you really dig into this stuff your realise stuff like the density of air is really an abstraction that doesn't quite fit what is actually going on.
It may not be possible to meaningfully answer this question in this case.
If we're talking about something like fluorescence, there's a fairy clear point where one photon disappears and another appears.
In a linear process like this, photons are not really absorbed by the material. In fact, the quantum behavior of photons is not relevant to the process, so you can just treat it purely as a wave phenomenon.
In cases like this, I would generally say that it is the "same" photon, but again, not really appropriate to think in terms of photons when there is nothing about the process that depends on quantization.
In my opinion, this question is impossible to properly answer in the framework of Maxwell's equations / field representation, since it cannot be defined in those terms.
If I were to answer this question in terms of photons as small amounts of field oscillations, I would argue that these are "new" photons, due to the fact that the "old" ones induced oscillations in the dipole moment of the material, which then in turn radiated energy out as the "new" photons.
But you can just as easily think of it as the material "suggesting" a better direction to the field propagating through it, and thus reorienting it. This is just very difficult to imagine and describe, at least for me.
If the material is being excited into oscillations that then re-emit "new" light, how is the color and direction preserved? Polarization filters tend to pass the full spectrum (or nearly so) of visible light, but my understanding of photon absorption and emittance is that the wavelengths are dependent on the electron energy levels. (I'm thinking of the same mechanism that produces lines on a spectrometer, indicating which elements are present in a sample.)
I guarantee I've misused a term or two above. Hopefully you get what I'm asking.
Taking a stab at my own question, the "rails" are field lines within the material, and not electrons themselves that interact. Is that close?
It’s because the “re-emission” is coherent in the sense that it’s in the same phase as the incoming light. As a decent analogy: when you sing a pure note, it “excites” (vibrates) air molecules as it travels, and those air molecules in turn bump into other molecules, all at random, but still all in phase so that whoever is listening hears the original note. Similarly, when light goes through ordinary glass, it wiggles the electrons in the glass, which in turn change the way the light propagates, refracting it while still preserving an image.
Any textbook on electricity and magnetism will cover this in a section called something like “Maxwell’s equations in materials”.
It's very likely an off-resonant, non-quantum excitation. If the incident light is at the resonance frequency of the atom, it will excite electrons which will spontaneously decay and throw off the energy. Far from resonance, though, the atoms will wiggle (classically) like a harmonic oscillator from forces due to the fields (similar to the answer here: https://physics.stackexchange.com/a/474/23322).
I have used the word "re-emit" in the sense of the Maxwell equations description of electromagnetic fields. You are right that the light definitely does not get absorbed and then re-emitted in the sense you mean. In such a case, everything you wrote would be correct. (I probably should have used a term like "radiate" and perhaps dipoles instead of electrons to be more clear.)
I was trying to describe the propagation of light in a material, where the optical field induces oscillations in the dipoles of the material, and these dipoles in turn excite the optical field. This happens constantly in every nanometer of the material, and it is difficult to experimentally separate the field into the "material" portion and the "vacuum" portion, because it exists as an everchanging mixture as long as there are dipoles around.
As for the "rails", the way I've had it explained to me is that in one direction of a polarizer, electrons are free to move, so they fully absorb the light polarized in that direction. In the perpendicular direction, they are bound, and the best they can do in reaction to a field is oscillate back and forth a tiny bit. These oscillations excite an optical field again and it propagates further until it finds another dipole to excite. I like to crudely imagine a polarizer as the grid of a nanoscopic egg slicer :D. Field oscillations will get absorbed along the metal wires, but in the perpendicular direction, it will just excite vibrations in the wires, which will radiate them out again, sort of like a guitar string.
Let me know if this was helpful, or if I've made a mistake somewhere :).
The concern that the article presents - that the middle filter influences the light and thus allows it to pass through the third filter - is actually addressed in popular quantum mechanics explanations that use the 3 filter experiment.
They say that if we use two entangled photons and let them fly far apart, then pass one of them through two filters, and the second photon through the middle filter, the first photon will be affected - it will get a chance to pass though the pair of filters.
That they say is "spooky action at distance" - the second photon will influence behaviour of the first photon at the remote site of the experiment and the "influence" is faster then the speed of light.
Example here by MinutePhysics and 3Blue1Brown: https://youtu.be/zcqZHYo7ONs
Explanation about entanglement starts at around 8:50.
But even with that addressed, to me personally this video is not satisfying.
If the spooky action at distance can be observed so trivially - choosing a filter at one site site affects what happens at the remote site - we don't need a mathematical inequality (the Bell's inequality), it's already so obviously spooky.
I am not sure the real Bell experiments are really done using 3 polarizing filters and will the effect really be observed in experiment with two remote sites.
My conclusion, it's problematic to rely on "pupular science" explanations, even by good channels like MinutePhysics and 3Blue1Brown.
> They say that if we use two entangled photons and let them fly far apart, then pass one of them through two filters, and the second photon through the middle filter, the first photon will be affected - it will get a chance to pass though the pair of filters.
I don't think that can be correct. This would allow FTL communication of information. I could put the filter in the path of the second photon when I wanted to send a 1 and leave it out of the path of the second photon when I wanted to send a 0. In the 1 case the other side would sometimes see the photon pass both filters. In the 0 case the other side would never see the photon pass both filters. Combine this with a little error correction and you have a channel for transmitting arbitrary information faster than light which would violate the no-communication theorem. https://en.wikipedia.org/wiki/No-communication_theorem
The MinutePhysics video you link describes a different setup with only one filter in the path of each photon. This gives you some information about the probability that other photon got past its filter but it doesn't let you transmit anything.
Yes, thank you, in distributed setting they really use only two filters.
But my main point is that the video addresses the issue from the article.
The video lays out intuition for the Bell's theorem using the 3 filter experiment.
And then, at 8:50 they say: "what if the act of passing through one filter changes how the photon will later interact with other filters? Then you can easily explain the results of the experiment" and continue explaining that the real spookiness can be proven in a distributed experiment.
The quantum nature of light is extremely difficult to observe. Almost all laboratory experiments can be explained using Maxwell's equations and the quantization of the electric charge (this explains why photodetectors "click"). Photons usually only show up when higher-order correlation functions are analyzed.
Except for the whole quantisation thing, and what happens in single-photon experiments.
This thread is very weird, there are a lot of people saying nonsense about photons not existing and absorption and re-emission (things that are detailed in any quantum physics textbook in existence). As if the fact that we can also explain the movement of planets around the sun somehow makes general relativity unnecessary or somewhat invalid. Quite disturbing.
I have a PhD in physics and I am on my first postdoc in the field of quantum optics so I'm well aware of the quantisation of the EM field. The point is that a lot of people think this is a demonstration of a purely quantum effect, but it is not, it is a classical effect. Sure, quantum mechanics is the full theory for describing it, but that's like saying that the increase in pressure when you heat up a gas is a "quantum effect".
It's basically misinformation, especially in the twitter thread that others mentioned. People are coming out thinking their mind has been blown with some crazy quantum effect, but actually this is described by boring classical EM.
I thought that experiment was something easy to do that required quantum theory to explain and the article proved me wrong. I'm the exact sort of curious, but ignorant and impressionable demographic that videos about that sort of effect target. I think that text really improved my understanding of reality.
Yeah it is a common misconception. There is also a minutephysics video on it (https://youtu.be/zcqZHYo7ONs) that portrays the effect of polarising filters in a misleading/wrong way, leading to further confusion.
Polarisation is used in many books and courses as a familiar classical system, which we then use to springboard into Stern-Gerlach, which actually is an inherently quantum effect.
Polarisation already appears when writing down the full solution to the classical EM wave equation. On the other hand, the similar effect seen in the Stern-Gerlach experiment with the angular momentum measurements is not: the non-commutivity of the angular momentum operator for different measurement axes is a purely quantum phenomenon.
This is not why “people want to cast it as something non quantum”. If it were the case, we would see “nice, classical EM and QM are consistent here”. Instead we have “classical EM works, therefore photons are bullshit and QM is not real”.
The fact that we can understand a macroscopic phenomenon using classical EM theory is thoroughly uninteresting, as the vast majority of what we see every day is a combination of quantum objects that we can understand macroscopically with classical physics. Of course people had EM theory before Planck. It’s still just a bunch of photons.
It is as not a personal attack of your knowledge, just that I don’t think your conclusion about people’s behaviour was right. My Physics PhD is as relevant as yours here :)
> Instead we have “classical EM works, therefore photons are bullshit and QM is not real”.
Anyone who is saying that is obviously wrong and of course I am not claiming that (although I do think that photons are probably bullshit)
> The fact that we can understand a macroscopic phenomenon using classical EM theory is thoroughly uninteresting
Why is it uninteresting? To me as a physicist this is thoroughly interesting!
> It’s still just a bunch of photons.
This is somewhat misleading, the wave description of light is complete and still totally preserved in quantum optics. The only deficiency is that the light's energy density should be quantised. You absolutely should not think of light as infinitely small particles that move at the speed of light, that is an inaccurate and outdated picture.
> Anyone who is saying that is obviously wrong and of course I am not claiming that
This thread has fine examples of people doing just that, I did not invent it.
> Why is it uninteresting? To me as a physicist this is thoroughly interesting!
It is interesting in that it’s an example of QM agreeing with classical EM theory. The opposite would be quite an issue actually, because it would mean that QM does not behave at the classical limit. There are lots of other examples of QM agreeing with classical mechanics, there is nothing magical about polarisers.
> the wave description of light is complete and still totally preserved in quantum optics. The only deficiency is that the light's energy density should be quantised.
In a narrow range of quantum optics maybe. I work on radiation/matter interaction, and you definitely cannot account for a lot of things if you posit that light is a wave. Unless you add some particle-like characteristics to your wave, in which case it’s not much better than assigning arbitrary wave-like properties to a particle.
> You absolutely should not think of light as infinitely small particles that move at the speed of light
I don’t.
> that is an inaccurate and outdated picture
It is not a wave either. The problem is when we try to pigeonhole it in either character. I mean, really, are we arguing the whole wave/particle duality concept again? That was novel a century ago.
According to the article, the "spookiness" comes from a misunderstanding of what a polarizer does. It doesn't "block" all light polarized on axes different from the polarizer. We know this is true because otherwise sunglasses would transmit much less light than they do. Imagine sunglasses could block any photon within +/- 1 degree of the polarization plane. That means that just 1/180th of the light would get through. But the observed transmission is much higher.
Instead, the polarizer does two things. First, it emits light polarized parallel to its axis. But, and this is the key, all incident light gets effectively passed. Along the way the intensity (amplitude, or "magnitude" in the article) is attenuated based on deviation from the polarizer's plane. The attenuation is 0% for light polarized in parallel and 100% for light polarized perpendicularly.
Now we can understand the experiment with a new mental model. Three filters are placed in series (A, B, and C). However, we can disregard A for the most part and treat this as a two-filter system (B, C), where the light exiting B is attenuated relative to the light entering A and polarized along B's axis. This model explains all of the observations.
And if you add more and more polarizers angled the right way, more and more light passes through, until in the limit you reach 1/2 the initial intensity, see https://aip.scitation.org/doi/pdf/10.1063/1.4943751
This could be demonstrated at the macro scale with a rope passing through slits of different alignments. You start shaking the rope in a circle, and observe what happens after it passes through each slit (the “polarizers”).
I'm a bit confused here. This link reads to me like "if we ignore any observations that can only be explained by quantum behaviors, then we don't need quantum behaviors to explain the observed behaviors". What am I missing?
The article grossly simplifies what is happening in the experiment and completely ignores the physical properties of EM-waves and polarizing materials. And there seems to be a misconception of how the polarizing filters relate to each other. Don't bother reading, just open a physics book from college or something.
I really like quantum flytrap. Happy to see you added a bunch of tools to do basic classical processing of the events, so I don't have to hack around that anymore by like making the bombs conditionally transparent.
One feature request: the setups are small enough that you could fit a description of them into the URL. This would allow people to much more easily share setups on forums and twitter and etc. This is what I did in Quirk and it's been one of the more useful features https://algassert.com/quirk#circuit=%7B%22cols%22%3A%5B%5B%2...
And I do like Quirk (we mention it in the intro to our recent paper https://doi.org/10.1117/1.OE.61.8.081808). And in the (not yet released) interactive element description for CNOT (along with Quantum Odyssey).
When it comes to links - a good point. I do like the StackOverflow style. I will discuss it with the team.
Fun fact: if you use more and more polarizing filters adjusted by finer and finer algorithms, and assume they are ideal (no loss of correctly oriented photons), that block of filters acts like a polarizing filter that filters to one angle but outputs at an angle rotated by 90 degrees.
This is an aspect of the quantum zeno effect. Normally it means to measure quickly while rotating an object to pin the object against the rotation. But here you instead measure a static object quickly while rotating the measuring device to force the object to rotate.
Incidentally, this is how LCDs work: two orthogonal polarizers (so no light should pass through), and sandwiched between them liquid crystals which rotate the light so it can pass through. When a current is applied, the crystals align, effectively turning off the rotation and thus the light.
It's also a nice exercise to see the polarizer as a linear transformation on the horizontal and vertical direction of the light: if the polarizer is oriented at direction θ, it's equivalent to rotating the light by -θ, keeping only the horizontal component, and rotating back. If you work out the matrixes for rotations 0, π/4, and π/2, you can see that multiplying 0 and π/2 gives the zero matrix, while multiplying 0, π/4, and π/2 keeps a nonzero component.
The article claims to explain this without reference to the "spookyness" of QM, but in fact it uses an analogous projection postulate as the QM version. After the polarizer, the photon state is projected onto the "measurement" of the polarizer, and in the article's figures the author does the same which leads to the same conclusion (kind of obviously).
What QM brings to the table is the explanation for what happens particle-by-particle.
88 comments
[ 4.8 ms ] story [ 156 ms ] threadSeveral replies explained how the effect can be explained without quantum mechanics.
This article (linked in one of those comments) is so clear, and I'm amazed I haven't seen it on HN before.
You can consider a wave passing through a filter as a sum of two orthogonal waves, rSin(θ) + rCos(θ), θ being the angle between the light-wave and the filtered angle, r being the amplitude of the wave.
One wave gets eliminated, and whatever exits exits at the only angle it can, the angle orthogonal to the filtered angle.
The reason, for example, that the 45 degree filter leaves 1/(2^.5) ~= .7
It's just the trigonometric break down of the right-angle isosceles.
I have always thought (how I got there I don't know) that the polarizer did something weird like rotate the photon to the correct phase angle AND passed it through with probability based on the angle / or didn't let it pass. This would give a similar reduction in intensity for a desktop experiment while having similar but different details when looking at the photon level. Is this correct?
If you want an explanation and prediction of what happens at the level of single photons, you need more structure from the theory of quantum optics. But briefly the filter at angle T does a measurement on the photon in the basis {T, T+pi/2}, and you end up seeing the photon on the other side of the filter only with whatever probability the photon has for being in state |T>, as opposed to state |T+pi/2>.
So, filters are inherently destructive and fewer and fewer photons pass through each subsequent filter. And a photon that makes it through a filter at angle T, now has a new state |T>.
They’re not an explanation for continuous waves in between, but the mathematics largely works anyway because photons are very similar to how one would do a Monte Carlo numerical simulations of continuous wave phenomena.
This has resulted in an unbelievable amount of confusion…
If you make the energy in a mode of the EM field continuous, you get the famous 'ultraviolet catastrophe.' This holds whether the matter involved has a continuous emission spectrum or not.
One is that when we send out a single photon worth of energy, even in an isotropic or other multi-path emission pattern, we only ever seem to detect a single photon. You can prove this with a half mirror and two detectors. This is kind of weird - not clear how this works, and the core question driving different quantum interpretations.
Second is that the energy level of a single photon detection is based on the constraint that Et=h (t is period), also written E=hf. This is not related to atomic properties of the detector. Why? Equation does not depend on any variable of the detector.
Same goes for Px=h. (Momentum/wavelength relationship.) You can prove that one with two razor blades close together. Same relationship, but along space axes instead of time.
Also for uncertainty principle - reinterpret EPtx as the variance of their distributions.
All of this is some sort of fundamental constraint of the universe around quantities of dimension force*time*distance (action). Not related to atoms per se. Not that I understand why this constraint exists. I know some ways to think about it mathematically, but not intuitively.
Fundamentally, you have to believe in nonsense such as kilometer-sized particles that are also infinitesimal if your point of view is that photons are particles.
Many of the "mysteries" of Quantum Mechanics vanish if keep in the back of your mind that QM is just a bag of mathematical tricks that "work" in specific scenarios to solve practical problems, but aren't a model of "what actually occurs".
Practically all of it is restricted to statistical predictions related to matter. Remove matter or remove statistics and then the models are no longer valid.
I'm not making this up. Go dig up any of the original QM papers and these assumptions and/or constraints are called out in the first few paragraphs.
They also tend to start vanishing once you get to studying Quantum Field Theory, which based on your comments I'm not convinced you have. Field Quanta very much "exist" (not getting into a debate on platonism here), and the debate between paticle/wave doesn't.
> I'm not making this up.
No, somebody else is, and you're buying it.
I wouldn't trust him either. What does he know?
You're basically arguing that the assumptions of QM theory that is in many of the older papers for some reason stopped mattering. That they've been invalidated... somehow, instead of simply being dropped for brevity in newer papers, assuming that everyone knows them already.
All I'm saying is that if you go back to the original version of the theory, they were quite clear about the scope: a statistical theory that applies to ensembles of atoms.
In theoretical Physics, especially when dealing with phenomena outside of normal human experience, you have to be very careful with your assumptions otherwise you can make yourself believe all sorts of nonsense.
Such as wave function collapse. Or spooky action at a distance. Both of which stem from a misunderstanding of the limits of the scope of a statistical theory that only applies to some types of physical objects.
That is not true, unless I completely misunderstood what you are trying to say. Photons exist in contexts without any atom involved, e.g. particle/antiparticle pair creation or annihilation. The gamma photons emitted during radioactive decay have nothing to do with orbitals either.
Besides, single photons have been observed. It was an important part of the Young slits experiment. Alain Aspect’s experiments also depend on the existence of single photon pairs. How would they work if photons were not a thing?
Or is it another case of "wave function collapse"? Remember that diffraction grating work too, and they rely on the wave nature of light. It is sometimes said (and can be modeled) that light takes every path from source to destination and the diffraction grating causes many paths to cancel out so it appears to go specific ways. Mirror reflection can also be modeled this way, and making periodic scratches will turn it into a diffraction grating. In the end, a photon detection is like a "measurement" that collapses the reflected wave function.
Photons are never observed "in flight" they are only detected when they hit something. In between they may as well BE a wave function.
Actually I just checked, on page 15:
> I want to emphasize that light comes in this form - particles. It is very important to know that light behaves like particles, especially for those of you who have gone to school, where you were probably told something about light behaving like waves.
Anyway I say "equivalently" because this is just another way to conceptualize the same theory (and make the same predictions), it's not new physics
The exact nature of the relationship of quantumness and fields (i.e. how the single-particle behavior arises from QFT) is still unclear, which is why there are many competing interpretations of quantum mechanics. In the Copenhagen interpretation, which is the most "easy" one, the behavior of photons is just (almost) that of Maxwell's equations, on average, s.t. a single photon will be measured with probability equal to the average light intensity anywhere (they are said to "collapse" at the moment of measurement, which is surely a simplification of a more complete underlying theory).
However, their speed of propagation in the material, and their properties can be altered by the interaction with dipole moments of the material. Therefore, photons do indeed "go trough without collision" but not without interaction. The dipoles in a dielectric material such as glass need to take the energy to oscillate from the field, and then they give it back to the field by radiating it out again.
I agree that this is not "re-emission" in the absorption/emission diagram sense, and I understand that the field in a material cannot be separated into the material portion and the vacuum portion, since it is both at any given time, but what I have tried to outline in the previous comment is still a useful representation of the role of the material which the original article chose to leave out. I should have used "radiate" and "dipole" instead of "emit" and "electron" to make it a bit clearer.
For a photon coming in at 45 degrees to the polarization angle, the probability that another photon will be emitted is sin(45deg) =~ 70% and the probability that it will be absorbed is 1/sin(45deg) =~ 30%.
(This is also a simplification; polarization angle is similarly quantum in nature, and I have assumed it to be collapsed here)
This is correct. The linked article doesn't explain how the single-photon behaves. It's the same issue that the two-slit experiment has if you try to ignore quantum behavior: You still can't say which slit a single photon passes through.
Quantum mechanics predicts that the difference is a factor of 4, not a factor of 2.
Power is proportional to the square of the amplitude. I don't know if that's what you're doing here, but it's a familiar pitfall.
The light does not "pass" through the middle filter, but it excites oscillations in the material, which effectively re-emits the light with different properties. The incoming light polarized at 0° induces oscillations in electrons which are "bound to a rail" in the material, which allows them to only oscillate in the direction of 45° (and all oscillations in the direction of -45° are absorbed). Therefore, a portion of the incoming field essentially gets re-emitted by the middle filter linearly polarized at 45°.
This representation is much less helpful if you think of the light in terms of individual photons rather than fields of course, but it is not worse than the article in this regard either.
How light actually behaves is probably beyond the ability of human cognition (since so much happens in a billionth of a second)
This is not remotely true. The behavior of light is very well understood and relatively simple to model compared to other, less linear physical processes.
When you really dig into this stuff your realise stuff like the density of air is really an abstraction that doesn't quite fit what is actually going on.
If we're talking about something like fluorescence, there's a fairy clear point where one photon disappears and another appears.
In a linear process like this, photons are not really absorbed by the material. In fact, the quantum behavior of photons is not relevant to the process, so you can just treat it purely as a wave phenomenon.
In cases like this, I would generally say that it is the "same" photon, but again, not really appropriate to think in terms of photons when there is nothing about the process that depends on quantization.
If I were to answer this question in terms of photons as small amounts of field oscillations, I would argue that these are "new" photons, due to the fact that the "old" ones induced oscillations in the dipole moment of the material, which then in turn radiated energy out as the "new" photons.
But you can just as easily think of it as the material "suggesting" a better direction to the field propagating through it, and thus reorienting it. This is just very difficult to imagine and describe, at least for me.
I guarantee I've misused a term or two above. Hopefully you get what I'm asking.
Taking a stab at my own question, the "rails" are field lines within the material, and not electrons themselves that interact. Is that close?
Any textbook on electricity and magnetism will cover this in a section called something like “Maxwell’s equations in materials”.
I was trying to describe the propagation of light in a material, where the optical field induces oscillations in the dipoles of the material, and these dipoles in turn excite the optical field. This happens constantly in every nanometer of the material, and it is difficult to experimentally separate the field into the "material" portion and the "vacuum" portion, because it exists as an everchanging mixture as long as there are dipoles around.
As for the "rails", the way I've had it explained to me is that in one direction of a polarizer, electrons are free to move, so they fully absorb the light polarized in that direction. In the perpendicular direction, they are bound, and the best they can do in reaction to a field is oscillate back and forth a tiny bit. These oscillations excite an optical field again and it propagates further until it finds another dipole to excite. I like to crudely imagine a polarizer as the grid of a nanoscopic egg slicer :D. Field oscillations will get absorbed along the metal wires, but in the perpendicular direction, it will just excite vibrations in the wires, which will radiate them out again, sort of like a guitar string.
Let me know if this was helpful, or if I've made a mistake somewhere :).
They say that if we use two entangled photons and let them fly far apart, then pass one of them through two filters, and the second photon through the middle filter, the first photon will be affected - it will get a chance to pass though the pair of filters.
That they say is "spooky action at distance" - the second photon will influence behaviour of the first photon at the remote site of the experiment and the "influence" is faster then the speed of light.
Example here by MinutePhysics and 3Blue1Brown: https://youtu.be/zcqZHYo7ONs Explanation about entanglement starts at around 8:50.
But even with that addressed, to me personally this video is not satisfying.
If the spooky action at distance can be observed so trivially - choosing a filter at one site site affects what happens at the remote site - we don't need a mathematical inequality (the Bell's inequality), it's already so obviously spooky.
There are also serious problems with clarity of their explanation, as I commented in https://www.youtube.com/watch?v=zcqZHYo7ONs&lc=Ugz3tzpDP_i1N... and https://www.youtube.com/watch?v=zcqZHYo7ONs&lc=Ugz3tzpDP_i1N...
I am not sure the real Bell experiments are really done using 3 polarizing filters and will the effect really be observed in experiment with two remote sites.
My conclusion, it's problematic to rely on "pupular science" explanations, even by good channels like MinutePhysics and 3Blue1Brown.
I don't think that can be correct. This would allow FTL communication of information. I could put the filter in the path of the second photon when I wanted to send a 1 and leave it out of the path of the second photon when I wanted to send a 0. In the 1 case the other side would sometimes see the photon pass both filters. In the 0 case the other side would never see the photon pass both filters. Combine this with a little error correction and you have a channel for transmitting arbitrary information faster than light which would violate the no-communication theorem. https://en.wikipedia.org/wiki/No-communication_theorem
The MinutePhysics video you link describes a different setup with only one filter in the path of each photon. This gives you some information about the probability that other photon got past its filter but it doesn't let you transmit anything.
But my main point is that the video addresses the issue from the article.
The video lays out intuition for the Bell's theorem using the 3 filter experiment. And then, at 8:50 they say: "what if the act of passing through one filter changes how the photon will later interact with other filters? Then you can easily explain the results of the experiment" and continue explaining that the real spookiness can be proven in a distributed experiment.
I dont get the desire to cast light as something non quantum...
Because you can describe it entirely using classical physics in this situation
This thread is very weird, there are a lot of people saying nonsense about photons not existing and absorption and re-emission (things that are detailed in any quantum physics textbook in existence). As if the fact that we can also explain the movement of planets around the sun somehow makes general relativity unnecessary or somewhat invalid. Quite disturbing.
It's basically misinformation, especially in the twitter thread that others mentioned. People are coming out thinking their mind has been blown with some crazy quantum effect, but actually this is described by boring classical EM.
Polarisation is used in many books and courses as a familiar classical system, which we then use to springboard into Stern-Gerlach, which actually is an inherently quantum effect.
Polarisation already appears when writing down the full solution to the classical EM wave equation. On the other hand, the similar effect seen in the Stern-Gerlach experiment with the angular momentum measurements is not: the non-commutivity of the angular momentum operator for different measurement axes is a purely quantum phenomenon.
The fact that we can understand a macroscopic phenomenon using classical EM theory is thoroughly uninteresting, as the vast majority of what we see every day is a combination of quantum objects that we can understand macroscopically with classical physics. Of course people had EM theory before Planck. It’s still just a bunch of photons.
It is as not a personal attack of your knowledge, just that I don’t think your conclusion about people’s behaviour was right. My Physics PhD is as relevant as yours here :)
Anyone who is saying that is obviously wrong and of course I am not claiming that (although I do think that photons are probably bullshit)
> The fact that we can understand a macroscopic phenomenon using classical EM theory is thoroughly uninteresting
Why is it uninteresting? To me as a physicist this is thoroughly interesting!
> It’s still just a bunch of photons.
This is somewhat misleading, the wave description of light is complete and still totally preserved in quantum optics. The only deficiency is that the light's energy density should be quantised. You absolutely should not think of light as infinitely small particles that move at the speed of light, that is an inaccurate and outdated picture.
This thread has fine examples of people doing just that, I did not invent it.
> Why is it uninteresting? To me as a physicist this is thoroughly interesting!
It is interesting in that it’s an example of QM agreeing with classical EM theory. The opposite would be quite an issue actually, because it would mean that QM does not behave at the classical limit. There are lots of other examples of QM agreeing with classical mechanics, there is nothing magical about polarisers.
> the wave description of light is complete and still totally preserved in quantum optics. The only deficiency is that the light's energy density should be quantised.
In a narrow range of quantum optics maybe. I work on radiation/matter interaction, and you definitely cannot account for a lot of things if you posit that light is a wave. Unless you add some particle-like characteristics to your wave, in which case it’s not much better than assigning arbitrary wave-like properties to a particle.
> You absolutely should not think of light as infinitely small particles that move at the speed of light
I don’t.
> that is an inaccurate and outdated picture
It is not a wave either. The problem is when we try to pigeonhole it in either character. I mean, really, are we arguing the whole wave/particle duality concept again? That was novel a century ago.
Also I have to ask, how do you think of a particle?
According to the article, the "spookiness" comes from a misunderstanding of what a polarizer does. It doesn't "block" all light polarized on axes different from the polarizer. We know this is true because otherwise sunglasses would transmit much less light than they do. Imagine sunglasses could block any photon within +/- 1 degree of the polarization plane. That means that just 1/180th of the light would get through. But the observed transmission is much higher.
Instead, the polarizer does two things. First, it emits light polarized parallel to its axis. But, and this is the key, all incident light gets effectively passed. Along the way the intensity (amplitude, or "magnitude" in the article) is attenuated based on deviation from the polarizer's plane. The attenuation is 0% for light polarized in parallel and 100% for light polarized perpendicularly.
Now we can understand the experiment with a new mental model. Three filters are placed in series (A, B, and C). However, we can disregard A for the most part and treat this as a two-filter system (B, C), where the light exiting B is attenuated relative to the light entering A and polarized along B's axis. This model explains all of the observations.
https://youtu.be/0yCtJAgO5gc
One feature request: the setups are small enough that you could fit a description of them into the URL. This would allow people to much more easily share setups on forums and twitter and etc. This is what I did in Quirk and it's been one of the more useful features https://algassert.com/quirk#circuit=%7B%22cols%22%3A%5B%5B%2...
When it comes to links - a good point. I do like the StackOverflow style. I will discuss it with the team.
This is an aspect of the quantum zeno effect. Normally it means to measure quickly while rotating an object to pin the object against the rotation. But here you instead measure a static object quickly while rotating the measuring device to force the object to rotate.
[1] https://en.wikipedia.org/wiki/Optical_rotation
It's also a nice exercise to see the polarizer as a linear transformation on the horizontal and vertical direction of the light: if the polarizer is oriented at direction θ, it's equivalent to rotating the light by -θ, keeping only the horizontal component, and rotating back. If you work out the matrixes for rotations 0, π/4, and π/2, you can see that multiplying 0 and π/2 gives the zero matrix, while multiplying 0, π/4, and π/2 keeps a nonzero component.
What QM brings to the table is the explanation for what happens particle-by-particle.