Double slit experiment exposed?

[Nav]

http://www.pnas.org/content/109/24/9314.full
According to the experimenters they have found out which path the photon took and still observed the interference pattern. So they know that the particle went through the left or right slit but at the same time saw the interference pattern built up on the back wall.
What do you guys think?
Do they really know which slit the photon went through?

 

[DrChinese]

Some of the interference features of entangled photons have been known for some time. Usually they do not even self-interfere, for a number of reasons. In this experiment, the interference requires coincidence counting. To my eyes, it looks like a very clean version of some delayed choice experiments which have been discussed around here.

So my point is that the result is not really "exposing" anything more than would be expected, given the limits of the Heisenberg Uncertainty Principle.

 

[Nav]

Some of the interference features of entangled photons have been known for some time. Usually they do not even self-interfere, for a number of reasons. In this experiment, the interference requires coincidence counting. To my eyes, it looks like a very clean version of some delayed choice experiments which have been discussed around here.

So my point is that the result is not really "exposing" anything more than would be expected, given the limits of the Heisenberg Uncertainty Principle.

So your saying that in the experiment, they don't know which slit the photon went through to create the interference pattern?

 

[bhobba]

So your saying that in the experiment, they don't know which slit the photon went through to create the interference pattern?

What I think he is saying is what struck me when I read it, and what often strikes me when I read about 'weird' experiments in QM, is the truth of QM lies in the math. Beginning textbook intuitions like if you don't know the path you get interference, wave particle duality etc, can come unstuck when applied to complex situations. I forget the exact experiment, but there was this experiment that on the surface seemed to contradict basic tenants of QM. It, like this, was rather complex and left you scratching your head about what was going on. But someone did a numerical simulation and showed its results were exactly what QM predicted. It just on the surface looked a problem - in reality it wasn't.

I think its the same here. QM isn't at fault - its simply our intuition becoming unstuck. Indeed at the beginning of the paper they say it likely represents a further modification of some of these early ideas like complementarity just as the quantum eraser did. My personal opinion is I don't particularly worry about ideas from the early days of QM going by the wayside. A lot of water has passed under the bridge since then and we understand QM a lot better without those early days crux's. For example our knowledge of decoherence is now much much better and explains the quantum eraser with ease. Its simply that in simple cases decoherence can be undone.

Bottom line here, and I think Dr Chinese's point, is this is a variant of the delayed choice eraser experiment, and simply shows, like that famous experiment, our intuition can be led astray.

Thanks
Bill

 

[ZapperZ]

Why is this any different than the other weak measurement experiments, even the one claiming to be able to track the trajectory of a particle? None of these are done with just ONE single measurement, are they? They had to do multiple/many identical measurements and then deduce from the entire statistics.

Even the authors in this paper are claiming that nothing they are doing violates the standard QM.

Zz.

 

[Nav]

What I think he is saying is what struck me when I read it, and what often strikes me when I read about 'weird' experiments in QM, is the truth of QM lies in the math. Beginning textbook intuitions like if you don't know the path you get interference, wave particle duality etc, can come unstuck when applied to complex situations. I forget the exact experiment, but there was this experiment that on the surface seemed to contradict basic tenants of QM. It, like this, was rather complex and left you scratching your head about what was going on. But someone did a numerical simulation and showed its results were exactly what QM predicted. It just on the surface looked a problem - in reality it wasn't.

I think its the same here. QM isn't at fault - its simply our intuition becoming unstuck. Indeed at the beginning of the paper they say it likely represents a further modification of some of these early ideas like complementarity just as the quantum eraser did. My personal opinion is I don't particularly worry about ideas from the early days of QM going by the wayside. A lot of water has passed under the bridge since then and we understand QM a lot better without those early days crux's. For example our knowledge of decoherence is now much much better and explains the quantum eraser with ease. Its simply that in simple cases decoherence can be undone.

Bottom line here, and I think Dr Chinese's point, is this is a variant of the delayed choice eraser experiment, and simply shows, like that famous experiment, our intuition can be led astray.

Thanks
Bill

But they claim to know which slit the photon passed through to create the interference pattern

 

[bhobba]

But they claim to know which slit the photon passed through to create the interference pattern

And your issue with that is?

As I said the truth of QM lies in the math - not in what texts say to give beginner students an intuitive feel. Another example is the wave-particle duality which says that quantum objects can show particle behaviour or wave behaviour - but not both. But experiments have been done where exactly that happens. The resolution of that is the wave-particle duality is a myth:
http://arxiv.org/abs/quant-ph/0609163

Personally I have no problems with an experiment that shows you can get interference even when knowing what path it went. What is known as weak measurements are rather strange beasts that can seem to contradict basic principles of QM:
https://en.wikipedia.org/wiki/Weak_measurement

The same way weak measurements can allow both position and momentum to be known at the same time then its quite reasonable they can show interference and which path information.

Thanks
Bill

 

[vanhees71]

I'll read the paper this evening, but already the title contains two sins, namely the buzz words "wave-particle duality" (which doesn't exist at all) and "complementarity" which is an unsharp philosophical idea by Bohr, but we'll see. Often quantum-theory papers by experts in the field are better than their titles and abstracts suggest ;-)).

 

[DarioC]

Bill, thanks for the reference to the arxiv paper. Looks like it will be a good one for my level of digging for a better understanding of QM.
DC

 

[jknm]

When a photon/matter particle departs a double slit apparatus, it has (whether we know, or not, what slit it has gone through) changed its momentum. A component is added across the direction of travel, in the plane of the slits. If you plot that momentum pattern, it has the form of a Fourier transform of the slit pattern, scaled by Planck's constant.
If you use quantum mechanics to calculate the momentum of a particle 'trapped' between two wells, then the expectation function for the transverse momentum is exactly the same as the pattern of momenta added to the incident particles in the double slit experiment.
The problem as to whether the incident particle physically 'goes' through both slits is, in my view, best addressed by noting that in QED 'virtual'/transient photons mediate in interactions. If a transient photon, acting in the plane of the slits, mediates the exchange of momenta, then its possible spectrum of momenta will be filtered by the slit structure.
Interestingly, this spectrum is the same as the frequency pattern that classical electrodynamics predicts if you treat the 'inverse' of the slits as an antenna.
The incident particle does not have to propagate through all holes. This is why the spectrum of transverse momentum matches the diffraction pattern, and is independent of the value of the incident momentum, and the type of particle involved.
As for finding out (by intervening with some interaction) which slit a particle goes through, the momentum of a photon that could sample where a particle is, would need to have a wavelength shorter than the slit separation, which is many times larger than the momentum change that is required to deflect a particle from one bright line of the diffraction pattern to another.
The experiment above, is consistent with Alfred Lande's interpretation of quantum theory. Furthermore, the idea that 'knowledge' of paths is important is actually an ad-hoc device to prop-up interpretations that seek to explain QM by having the incident particles bring a classical 'wave' phenomenon to a passive slit barrier. Such interpretations try to jam everything into the incident particle as if it is a magic 'swiss army knife' that adapts according to what 'knowledge' is available, I think that instead, explanations that break the scattering down into more than one particle type and interaction can work better, and actally fit better with the quantum math.

 

[Derek Potter]

Why is this any different than the other weak measurement experiments, even the one claiming to be able to track the trajectory of a particle? None of these are done with just ONE single measurement, are they? They had to do multiple/many identical measurements and then deduce from the entire statistics.

Of course the experiment needs multiple measurements - how else would a pattern or graph emerge? Unless I'm missing something HUGE, the comparison with weak measurements is a bit misleading. This experiment uses multiple measurements for entirely classical reasons - in order to see the probability distribution as an actualized pattern.

 

[Derek Potter]

Personally I have no problems with an experiment that shows you can get interference even when knowing what path it went. What is known as weak measurements are rather strange beasts that can seem to contradict basic principles of QM:
https://en.wikipedia.org/wiki/Weak_measurement
The same way weak measurements can allow both position and momentum to be known at the same time then its quite reasonable they can show interference and which path information.

Bill, have you read Lubos Motl's scathing comments on weak measurements?

 

[bhobba]

Bill, have you read Lubos Motl's scathing comments on weak measurements?

So?

All that's going on with weak measurement is since its weakly coupled to what you are measuring you can measure things you might think QM forbids. But since they are weakly coupled they are not accurate so do not violate QM principles such as the uncertainty relations or measuring the path and getting an interference pattern.

As Lubos says - 'The "weak value" doesn't actually generalize the "measured value". Instead, it generalizes the "expectation value".'

Thanks
Bill

 

[Derek Potter]

As Lubos says - 'The "weak value" doesn't actually generalize the "measured value". Instead, it generalizes the "expectation value".'

I'm not arguing, I was just curious to know what you thought of Motl's impatient dismissal of weak measurements as no measurements at all.

 

[bhobba]

I'm not arguing, I was just curious to know what you thought of Motl's impatient dismissal of weak measurements as no measurements at all as you drew the comparison with them.

They are what they are. Semantics about what you call them IMHO is pointless. Call them Jaberwockeys - it makes no difference.

As I said at the start the truth lies in the math - our intuition can easily be fooled.

Thanks
Bill

 

[Derek Potter]

They are what they are. Semantics about what you call them IMHO is pointless. Call them Jaberwockeys - it makes no difference.

As I said at the start the truth lies in the math - our intuition can easily be fooled.

Thanks
Bill

Then the answer is to train our intuition correctly.

 

[atyy]

Cthugha commented on (a paper that commented on) the paper in https://www.physicsforums.com/threads/duality-principle-is-safe-and-sound.767696/#post-4833650

That is an incredibly annoying way of overselling.

The 2012 paper (R. Menzel, D. Puhlmann, A. Heuer, and W. P. Schleich.
Wave-particle dualism and complementarity unraveled by
a dierent mode. Proceedings of the National Academy of
Sciences, 109(24):9314{9319, 2012., http://www.pnas.org/content/109/24/9314.abstract?sid=af26123d-7825-4b52-bdae-beab95cf46f9) never actually claimed that it breaks complementarity or stuff like that. They had two modes of different wave vector present at the double slit, which means that even if you only have one slit, you still have two probability amplitudes for getting a photon to the detector that can interfere, if it is intrinsically indistinguishable which mode each photon will end up in.

The current paper (http://arxiv.org/abs/1402.6487) now shows that the previous paper does not prove a claim that was never made in that paper. I need to remember that trick for "selling" my next manuscripts. I must admit that their analysis of interference visibility variations is nice. I am still somewhat puzzled why the paper made it to PNAS, though.

 

[bhobba]

Then the answer is to train our intuition correctly.

Good luck.

Thanks
Bill

 

[Derek Potter]

Good luck.

No luck involved, but thanks anyway.

 

[gjonesy]

But they claim to know which slit the photon passed through to create the interference pattern

And your issue with that is?

I think what Nav is getting at is similar to the discussion Dr Chinese and I had on the subject...(ie to what degree you gain "which slit information you lose the interference pattern"

What it appears to suggest is that you can some how know which slit and make a measurement and still get interference.

 

[Nugatory]

I think what Nav is getting at is similar to the discussion Dr Chinese and I had on the subject...(ie to what degree you gain "which slit information you lose the interference pattern"

What it appears to suggest is that you can some how know which slit and make a measurement and still get interference.

Whether you get an interference pattern or not depends on what the wave function at the screen is. You can calculate that from the wave function at the barrier with the slits. The classic form of the two-slit experiment, in which there is no contribution at all from one of the slits, is the only case in which you should trust simple English-language descriptions of the problem that start "if you know which slit..."

 

[Skim Halo]

What I don't understand about this experiment (and every show I've seen describing it) is how they can say that the single photon they think they are firing is a particle. Yes it's spooky that an interference pattern appears when not observed and only two lines when observed but if you are truly firing a particle at the solid center of a plate containing two slits, why is it passing through any slit at all? It should behave like a bullet and hit the center between the slits and nothing should appear on the back wall.

If you think you're firing a single particle that's in no way a wave and you are aiming precisely at a solid substrate why are you not questioning whether you've truly isolated a single particle out of your firing device?

 

[bhobba]

What I don't understand about this experiment (and every show I've seen describing it) is how they can say that the single photon they think they are firing is a particle.

It isn't a classical particle - its something entirely different - a quantum particle.

Thanks
Bill