The detector itself contaminating double slit? How do we know?

[bahai]

Hi Matt, can you summarize for me here please? I am extremely confused by the claims made in youtube videos regarding the double-slit experiment.

 

[mattjfox]

Sorry Bahai. I am pretty confused as well. As a matter of fact... I have come to the conclusion that no one really knows wtf is going on when it comes to light unfortunately. If they do, I personally haven't met anyone who is able to communicate it well enough for my little brain to understand. Everyone seems to have an opinion on the matter and that makes it even worse.

I don't really buy the observer effect though any longer personally.

 

[bhobba]

Hi Matt, can you summarize for me here please? I am extremely confused by the claims made in youtube videos regarding the double-slit experiment.

Forget them.

Here, using the QM formalism, is the proper answer:
http://arxiv.org/ftp/quant-ph/papers/0703/0703126.pdf

The issue is this stuff can be discussed at various levels of sophistication. At the lay level they are not quite correct - but it the best that can be done.

A much better analysis is the paper above. Note the heavy use of math. That's the issue here. This stuff is written in the language of math. If you want to avoid it you run into difficulties being exact.

But even that link has issues - a more careful analysis shows its not quite the truth either. And that is a big problem in physics - the real answers are often quite subtle, complex and use some really heavy math.

Thanks
Bill

 

[bhobba]

Sorry Bahai. I am pretty confused as well. As a matter of fact... I have come to the conclusion that no one really knows wtf is going on when it comes to light unfortunately. If they do, I personally haven't met anyone who is able to communicate it well enough for my little brain to understand. Everyone seems to have an opinion on the matter and that makes it even worse.

They do - its described by QED.

But its a theory of the highly advanced math of Quantum Field Theory. I quote a guy who posts under the name of Streangerep in my signature - another thing he says - and its equally true - is one finds humility in field theory. It's HARD.

That said you may find the following dirt cheap lay treatment of it valuable:
https://www.amazon.com/dp/B004ULVG9O/?tag=pfamazon01-20

I have a copy and quite like it.

Feynmans well known book is also good:
https://www.amazon.com/dp/B00BR40XJ6/?tag=pfamazon01-20

But the issue you run into here is the varying levels of sophistication these things can be discussed at. At the lay level of sophistication it's not quite correct - but its the best that can be done. Feynman's book is not quite correct. This is really annoying - I know - but it's just the way it is.

Thanks
Bill

 

[e.bar.goum]

As a matter of fact... I have come to the conclusion that no one really knows wtf is going on when it comes to light unfortunately.

+1 to bhobba's comments. In fact, we've had a very good handle on what light does for pretty close to 100 years now. QED is just hard to explain in laypersons terms. Light is easy compared to a lot of problems in physics.

The fact that the (science interested) general public still thinks that "wave-particle duality" is a problem tells us something fairly troubling about science communication. To (poorly) paraphrase Sean Carrol - "Subtatomic particles aren't waves or particles, they're fields".

 

[bhobba]

To (poorly) paraphrase Sean Carrol - "Subtatomic particles aren't waves or particles, they're fields".

Indeed. QM is simply an approximation to QFT.

That's the view of the Colour Of Fields book I linked to. It tries to get that across in lay terms.

A different approach I rather like since it avoids many pitfalls.

Thanks
Bill

 

[e.bar.goum]

That's the view of the Colour Of Fields book I linked to. It tries to get that across in lay terms.

Thanks for the recomendation. I should check it out, my lay-person explanation of QFT is more like a third-year-undergraduate-physics-student explanation, which is rather unsatisfying for anyone without several years of university physics.

 

[atyy]

I think one confusing thing in this thread is that I, bhobba and Nugatory have agreed that the different results for single and double slit experiments are not puzzling, since different experiments give different results, and that the main complaint about the term "messing up" is that it could be misleading. On the other hand DrChinese has insisted that "messing up" is plain wrong, as indeed Feynman himself suggested.

Is the difference between whether contextuality or non-contextuality is adopted? So if we adopt non-contextuality, then we would say the detector is not messing up, and there is quantum weirdness. But if we allow contextuality, then we would say that the detector is "messing up" the interference, and there is no quantum weirdness. ?

 

[bhobba]

On the other hand DrChinese has insisted that "messing up" is plain wrong, as indeed Feynman himself suggested

I wouldn't get too worried.

Its just semantics and different interpretations of such - semantics would have to be one of the silliest things ever to get too worried about.

Thanks
Bill

 

[bhobba]

Thanks for the recomendation. I should check it out, my lay-person explanation of QFT is more like a third-year-undergraduate-physics-student explanation, which is rather unsatisfying for anyone without several years of university physics.

If you are third year undergraduate level in math and/or physics the following would be a better choice:
https://www.amazon.com/dp/0691140340/?tag=pfamazon01-20

Its slow going, and you learn more with each reading - but it is the real deal.

Thanks
Bill

 

[e.bar.goum]

If you are third year undergraduate level in math and/or physics the following would be a better choice:
https://www.amazon.com/dp/0691140340/?tag=pfamazon01-20

Its slow going, and you learn more with each reading - but it is the real deal.

Thanks
Bill

I'm actually a physics PhD student. It's just that my explanation of QED is targeted at about a 3rd year level. (Which may indicate I don't understand it enough since I can't explain it to my grandmother. But that's ok, I'm an experimentalist ;) .) I like to read other peoples lay-person explanations, it helps make mine better.

 

[atyy]

Indeed. QM is simply an approximation to QFT.

That's the view of the Colour Of Fields book I linked to. It tries to get that across in lay terms.

A different approach I rather like since it avoids many pitfalls.

Thanks
Bill

Thanks for the recomendation. I should check it out, my lay-person explanation of QFT is more like a third-year-undergraduate-physics-student explanation, which is rather unsatisfying for anyone without several years of university physics.

Or maybe QFT is an approximation to QM :) http://arxiv.org/abs/hep-lat/0211036 suggests that even QM can get g-2 in principle.

 

[atyy]

I wouldn't get too worried.

Its just semantics and different interpretations of such - semantics would have to be one of the silliest things ever to get too worried about.

Thanks
Bill

Good to know.

 

[bhobba]

I'm actually a physics PhD student. It's just that my explanation of QED is targeted at about a 3rd year level. (Which may indicate I don't understand it enough since I can't explain it to my grandmother. But that's ok, I'm an experimentalist ;) .) I like to read other peoples lay-person explanations, it helps make mine better.

Yea - my background is applied math - which labels me as a theorist I am afraid - not really into experimental stuff. My typical reaction and inclination is what some guy wrote about Landau - Mechanics:
https://www.amazon.com/dp/0750628960/?tag=pfamazon01-20

BTW love your picture of the great Emily Noether - deep respect.

If anyone doesn't know of her, and her famous Noether's Theorem - do check it out.

Thanks
Bill

 

[e.bar.goum]

Yea - my background is applied math - which labels me as a theorist I am afraid - not really into experimental stuff. My typical reaction and inclination is what some guy wrote about Landau - Mechanics:
https://www.amazon.com/dp/0750628960/?tag=pfamazon01-20

BTW love your picture of the great Emily Noether - deep respect.

If anyone doesn't know of her, and her famous Noether's Theorem - do check it out.

Thanks
Bill

Good eye! I think Noether's theorem is absolutely the most beautiful result in theoretical physics. Noether herself is so fantastic. Even for a mathematitian .

 

[bhobba]

Or maybe QFT is an approximation to QM :) http://arxiv.org/abs/hep-lat/0211036 suggests that even QM can get g-2 in principle.

I mean it in the sense you will find on page 18 of Zees book.

QM is simply 0+1 dimensional QFT

Thanks
Bill

 

[bhobba]

Good eye! I think Noether's theorem is absolutely the most beautiful result in theoretical physics. Noether herself is so fantastic. Even for a mathematitian .

So do I.

As I said deep respect for both her and her work.

As well as the discusting way she was treated that led Hilbert to say 'After all, we are a university, not a bath house'

Thanks
Bill

 

[DrChinese]

Also I don't really get the validity of the polarization experiments. When you turn the lens 90 degrees from the other lens aren't you are essentially just blocking light from going through one of the holes. If I am understanding that correctly I am not sure how this method validates anything except recreating a 1 slit experiment where you get a blob of light instead of interference pattern again.

No, neither hole is blocked. The total amount of light passing through remains unchanged if the source is suitably situated.

 

[DrChinese]

Sorry Bahai. I am pretty confused as well. As a matter of fact... I have come to the conclusion that no one really knows wtf is going on when it comes to light unfortunately. If they do, I personally haven't met anyone who is able to communicate it well enough for my little brain to understand. Everyone seems to have an opinion on the matter and that makes it even worse.

This is a complete misconception. The knowledge of light is extensive and what you have read does not even begin to scratch the surface. The reason it is difficult to communicate is because there are so many situations in which subtle changes in the setup leads to different outcomes. Once it is agreed what the setup actually is, the scientific prediction can be determined.

 

[atyy]

@DrChinese, if I understand correctly, the basic intuition that mattjfox has is that different experiments have different results. The experiment with two polarizers vertical, and the experiment with one polarizer vertical and one polarizer horizontal, are different experiments. So why should we expect them to give the same interference pattern?

 

[DrChinese]

@DrChinese, if I understand correctly, the basic intuition that mattjfox has is that different experiments have different results. The experiment with two polarizers vertical, and the experiment with one polarizer vertical and one polarizer horizontal, are different experiments. So why should we expect them to give the same interference pattern?

You are right, that part is fine.

But in my example, there is still the same "detector" mechanism in place regardless of the polarizer settings themselves. Note the title of the thread: "The detector itself contaminating double slit? How do we know?" So we do know!

And the answer is: It is NOT the detector's presence itself changing the outcome. It is ONLY the relative setting of the polarizers.

 

[atyy]

You are right, that part is fine.

But in my example, there is still the same "detector" mechanism in place regardless of the polarizer settings themselves. Note the title of the thread: "The detector itself contaminating double slit? How do we know?" So we do know!

And the answer is: It is NOT the detector's presence itself changing the outcome. It is ONLY the relative setting of the polarizers.

How's these: ?

In the case where the detector is placed at one of the slits, the detector is changing the interference pattern (compared to the setup without the detector).

More generally, an experiment with a different operation on the system (whether it is [STRIKE]unitary like[/STRIKE] a polarizer, or non-unitary like a detector) would change the interference pattern (compared to the setup without the operation).

This is not puzzling in general because we do expect different operations to have different results.

If there is a mystery in quantum mechanics, a Bell test expresses it better than the double slit, because the Bell tests prove that any realistic hidden variables cannot be local.

Edit: Not sure if a polarizer is unitary, it looks like a projector, see references in post #68.

 

[DrChinese]

How's these: ?

In the case where the detector is placed at one of the slits, the detector is changing the interference pattern (compared to the setup without the detector).

The only "detector(s)" in my setup are the polarizers in front of each slit. Nothing is added or removed. Let's assume the source polarization is at 0 degrees. When the polarizers are aligned parallel (both at 45 degrees) there is interference. If one of the polarizers is changed to -45 degrees, there is no interference. Intensity does not change in this example.

The only thing that changes is that when we have the *possibility* of determining which-slit information, interference disappears. Note that in actuality, nothing at all is detected. Yes, the different angle settings lead to different setups and therefore different results. But it is NOT because a detector is "contaminating" the double slit!

 

[atyy]

The only "detector(s)" in my setup are the polarizers in front of each slit. Nothing is added or removed. Let's assume the source polarization is at 0 degrees. When the polarizers are aligned parallel (both at 45 degrees) there is interference. If one of the polarizers is changed to -45 degrees, there is no interference. Intensity does not change in this example.

The only thing that changes is that when we have the *possibility* of determining which-slit information, interference disappears. Note that in actuality, nothing at all is detected. Yes, the different angle settings lead to different setups and therefore different results. But it is NOT because a detector is "contaminating" the double slit!

Sure, but one can also set it up with a detector. In that case, the detector is messing up the interference pattern.

 

[DrChinese]

Sure, but one can also set it up with a detector. In that case, the detector is messing up the interference pattern.

In the OP's sense, sure. But that is obviously not the general case. Generally, the detector is NOT the cause and that is what I wanted to make clear.

My case is the general case, and that makes it clear that there is something else at work. The OP is on the wrong track.

 

[atyy]

In the OP's sense, sure. But that is obviously not the general case. Generally, the detector is NOT the cause and that is what I wanted to make clear.

My case is the general case, and that makes it clear that there is something else at work. The OP is on the wrong track.

I think we all agree that some concepts in the initial posts were not quite correct. But couldn't the OP's idea be generalized to your case by saying that the different polarizer setting messed up the interference pattern?

 

[DrChinese]

I think we all agree that some concepts in the initial posts were not quite correct. But couldn't the OP's idea be generalized to your case by saying that the different polarizer setting messed up the interference pattern?

In my semantics, no. The original implication was that the detector's presence alone actively disturbs the interference.

In my storyline, it is the relationship between the polarizers that is central to the experiment. That is all that changes. So that should be the starting point of further discussion about when and why interference patterns emerge.

 

[Nugatory]

I think we all agree that some concepts in the initial posts were not quite correct. But couldn't the OP's idea be generalized to your case by saying that the different polarizer setting messed up the interference pattern?

Depends on whether you think of the polarizer setting as a controlled variable whose effect on the interference pattern is what you're trying to investigate, or an accidental uncontrolled input...

Yes, this is something of a quibble about wording, especially because when it comes to designing an experiment, it's natural enough to think of the interference pattern as something that is changed by the action of ("messed up by") the polarizer setting. But I still feel that that if we encourage people to start out thinking in those terms, they'll just have to unlearn them at some point.

 

[atyy]

In my semantics, no. The original implication was that the detector's presence alone actively disturbs the interference.

In my storyline, it is the relationship between the polarizers that is central to the experiment. That is all that changes. So that should be the starting point of further discussion about when and why interference patterns emerge.

I'm not so familiar with your set-up. Concretely, how would the polarizer settings be used to infer which path information?

 

[DrChinese]

I'm not so familiar with your set-up. Concretely, how would the polarizer settings be used to infer which path information?

That's a good question!

1. If you added an extra polarizer between the slits and the screen where the pattern appears, and orient it at a 45 degree angle: any photon arriving at the screen must have gone through the slit oriented at 45 degrees as well. You have gained which slit information.

2. If you next oriented that extra polarizer at a -45 degree angle: any photon arriving at the screen must have gone through the slit oriented at -45 degrees as well. You have gained which slit information.

3. Importantly: no extra polarizer is actually required for the interference to disappear when the slit polarizers are crossed per above. It is enough that you could have done it, regardless of whether you actually did.

 

[DrChinese]

Depends on whether you think of the polarizer setting as a controlled variable whose effect on the interference pattern is what you're trying to investigate, or an accidental uncontrolled input...

Yes, this is something of a quibble about wording, especially because when it comes to designing an experiment, it's natural enough to think of the interference pattern as something that is changed by the action of ("messed up by") the polarizer setting. But I still feel that that if we encourage people to start out thinking in those terms, they'll just have to unlearn them at some point.

I so agree!

 

[atyy]

That's a good question!

1. If you added an extra polarizer between the slits and the screen where the pattern appears, and orient it at a 45 degree angle: any photon arriving at the screen must have gone through the slit oriented at 45 degrees as well. You have gained which slit information.

2. If you next oriented that extra polarizer at a -45 degree angle: any photon arriving at the screen must have gone through the slit oriented at -45 degrees as well. You have gained which slit information.

3. Importantly: no extra polarizer is actually required for the interference to disappear when the slit polarizers are crossed per above. It is enough that you could have done it, regardless of whether you actually did.

Thanks!

Depends on whether you think of the polarizer setting as a controlled variable whose effect on the interference pattern is what you're trying to investigate, or an accidental uncontrolled input...

Yes, this is something of a quibble about wording, especially because when it comes to designing an experiment, it's natural enough to think of the interference pattern as something that is changed by the action of ("messed up by") the polarizer setting. But I still feel that that if we encourage people to start out thinking in those terms, they'll just have to unlearn them at some point.

I so agree!

What is the actual mathematical expression for this principle that the interference pattern depends on potential which path information? (I confess I have heard it, but I never use it, nor have I seen it as an equation. I always think the detector or polarizer is messing up the experiment. In the delayed choice quantum eraser experiment, I simply think different experimnents have different results, and different observables have different distributions, and it is no mystery to me.)

 

[atyy]

Also, looking up the polarizer action, it seems to be a projector, which is not so different from a measurement and collapse?

These notes all agree the polarizer is a projector:
http://www.physics.metu.edu.tr/~sturgut/p507/pol.pdf
http://ocw.mit.edu/courses/nuclear-...s-fall-2012/lecture-notes/MIT22_51F12_Ch3.pdf
http://web.stanford.edu/~rsasaki/AP226/text1.pdf
http://www.math.ru.nl/~maassen/lectures/Trieste.pdf

http://books.google.com/books?id=l-l0L8YInA0C&vq=polarizer&source=gbs_navlinks_s (p415) even makes an analogy between the polarizer and a measurement device, associating an observable to it, and eigenvectors.

 

[Cthugha]

Also, looking up the polarizer action, it seems to be a projector, which is not so different from a measurement and collapse?

Yes, a polarizer is a projector. However, in these measurements one usually places a lambda-half waveplate at the slits. It just rotates the polarization of the light beam passing through it. This is a completely reversible interaction.

 

[atyy]

Yes, a polarizer is a projector. However, in these measurements one usually places a lambda-half waveplate at the slits. It just rotates the polarization of the light beam passing through it. This is a completely reversible interaction.

Thanks! So there is a unitary operation which is just rotation to realize DrChinese's experiment.

Looking at the delayed choice quantum eraser the OP linked to earlier, it doesn't seem mysterious because the observable for which interference is observed or not observed is not even the same observable. So I would explain the "mysteries" by saying different setups yield different results, different observables yield different results. But is there a principle beyond that, a mathematical expression that formalizes the idea that potential which path information destroys the interference? I assume it's not just the [x,p] commutation relation?