Will detecting 1 of the 3 slits break the wave patterns?

[mousheng xu]

In a double slit experiment (say, of electrons), when putting a detector to examine the passing of electrons through one slit, the wave pattern of the electrons disappears and instead the particle pattern appears. The classical explanation is that an electron exists both as a wave and as a particle the same time, but the wave feature of an electron disappears when it is observed.

Several questions come to me for such an explanation.

1) The so called "wave pattern disappears when particles are observed" is not exactly true, because we observed the wave pattern anyway, be it observed by a film or something similar.

2) Let's assume the wave pattern was observed by a film in the back of the two slits. If we use multiple layers of semi-transparent films, will we be able to see the wave pattern on the first layer and other layers? If yes, the belief that observing an particle destroys superposition (of multiple possibilities) or wave feature does not hold.

3) If we use 3 slits instead of 2 slits, and we use an electron detector to observe one slit, what will happen? Has anyone actually done such an experiment? I've seen people reasoning and debating what could happen but have not seen any experimental results. It is a big surprise that such an experiment has not actually been carried out. Such an experiment would be very interesting.

Thanks for any discussion or information.

 

[Nugatory]

1) The so called "wave pattern disappears when particles are observed" is not exactly true, because we observed the wave pattern anyway, be it observed by a film or something similar.

it's not true at all, and no one has ever said that it might be. Instead, what is being said is that the pattern disappears if we observe (by any means) which slit the particle goes through; that remains true.

2) Let's assume the wave pattern was observed by a film in the back of the two slits. If we use multiple layers of semi-transparent films, will we be able to see the wave pattern on the first layer and other layers? If yes, the belief that observing an particle destroys superposition (of multiple possibilities) or wave feature does not hold.

Are you running the experiment with individual particles so each produces a single visible dot on the film, and the dots build up to form the interference pattern? Every particle will leave its dot on one layer and that interaction is the first and only interaction for that particle, the one that collapses the wave function and ends the superposition.

3) If we use 3 slits instead of 2 slits, and we use an electron detector to observe one slit, what will happen? Has anyone actually done such an experiment? I've seen people reasoning and debating what could happen but have not seen any experimental results. It is a big surprise that such an experiment has not actually been carried out. Such an experiment would be very interesting.

There's no doubt about what quantum mechanics says will happen. I don't know if anyone has actually done the exact experiment you suggest, but I don't find it so surprising that probably no one has. Experiments are exprnsive and hugely time-consuming so we cannot do them all; instead we prioritize the ones that have good chances of telling us something new.

Experiments in which there are more than two paths, but the paths aren't realized by physical slits as you're asking, certainly have been done, many times.

 

[Justice Hunter]

Let's take for example, an electron gun that shoots one electron at a time towards a wall with 3 slits. and behind it is a photographic plate. if the detector is turned off you'll get a diffraction pattern, that looks pretty much the same as a double slit.

But let's say, we do turn the detector on, which is like a low energy photon emitter. In a perfect world we can't turn on the detector, and only have it detect just one slit, but, regardless the result would be the same. We would have a high intensity region closest to the detector, and further away from it, the interference pattern shows. the reason is the nature of the detector, which is just photons hitting the oncoming electrons, which knocks them out of their superposition and gives them a defined trajectory.

Now don't be fooled by the depiction. Those electrons represented as wavy things, aren't really waves nor are they really particles with definite trajectories. Only until the detector has hit them, do they gain a definite trajectory. The electrons exist in a special state where they have the potential to be anywhere in the picture, with a given probability of being detected at a given location. It's only when the photons from the photon emitter interact with an electron, do they gain a trajectory.

Now diving into how the photographic plate actually works, keep in mind, that the photographic plate itself is also a detector. is just a solid surface with indicators, or light switches that flick on when hit with the force of that electron. If one were to create such a plate, where the electron does not stop but still interacts with the plate, then what happens is the plate is like a secondary detector. The plate would collapse the superposition of the incoming electrons, and any electron passing beyond the plate would have a definite trajectory.

However, if we were to only have one electron for example, and once it passes through the first plate, the plate itself changes the trajectory of the passing electron with an unknown momentum, and so when the passing electron were to hit a 2nd plate, the position it lands will be different then when it passed through the first. Given the amount of interaction from the plate, i don't exactly know what it would look like, but i would imagine it being a jumbled mess, basically dots everywhere dispersed randomly.

Now I'm not sure if this is the actual case or not, and could be up for debate, but i assume that this conclusion is solely based on the nature of the plate itself, and how it would actually work.

In addition, a little side note. Here is a link that i found that shows multiple slit diffraction, although not shown to have a detector involved, shows what the patterns look like for an increasing number of slits.

http://hyperphysics.phy-astr.gsu.edu/hbase/phyopt/mulslidi.html

 

[StevieTNZ]

2) Let's assume the wave pattern was observed by a film in the back of the two slits. If we use multiple layers of semi-transparent films, will we be able to see the wave pattern on the first layer and other layers? If yes, the belief that observing an particle destroys superposition (of multiple possibilities) or wave feature does not hold.

I'll chime in on this question.

I wouldn't say that the particle hitting the first semi-transparent is an observation in the sense that it collapses the particle to one of the many possible states it could take on. We cannot tell from that interference pattern which one slit the particle took, because it "took all at the same time" (not in the classical sense).

 

[mousheng xu]

1) The so called "wave pattern disappears when particles are observed" is not exactly true, because we observed the wave pattern anyway, be it observed by a film or something similar.
it's not true at all, and no one has ever said that it might be. Instead, what is being said is that the pattern disappears if we observe (by any means) which slit the particle goes through; that remains true.
2) Let's assume the wave pattern was observed by a film in the back of the two slits. If we use multiple layers of semi-transparent films, will we be able to see the wave pattern on the first layer and other layers? If yes, the belief that observing an particle destroys superposition (of multiple possibilities) or wave feature does not hold.
Are you running the experiment with individual particles so each produces a single visible dot on the film, and the dots build up to form the interference pattern? Every particle will leave its dot on one layer and that interaction is the first and only interaction for that particle, the one that collapses the wave function and ends the superposition. --------------------------------- Reply starts here --------------------------------
Thank you, https://www.physicsforums.com/threads/will-detecting-1-of-the-3-slits-break-the-wave-patterns.840890/members/nugatory.382138/ . First thanks for pointing out the error I had in the 1) question. If the wave pattern disappears after we observe which slit the particles go through, I have to rethink for a while for question 1) & 2).

As for question 3), without doing experiments, what would be your answer? The pattern would be that of shooting particles through the slits?

 

[Simon Phoenix]

Mousheng,

you ask what would happen if we did the experiment. OK - so let's think about putting a detector - say at slit 1. What is the result of this? Well, we're either going to get a 'ping' or 'no ping'. If we get a ping then we know the state of the outputs of slits 2 and 3 to be |0>, i.e., nothing or no occupation of these output modes. If we get no ping at slit 1 - then we know the state of slits 2 and 3 to be just the same as if we were doing a 2-slit experiment.

So we have a statistical mixture of 1/3 nothing hitting the screen at all - and 2/3 normal 2-slit result for slits 2 and 3 (assuming we've set it up to give an equal probability of detecting the electron in any of the outputs).

So loosely:
Initial state (non-normalized) goes as |001> + |010> + |100>
where the 1 and 0 here indicate the occupation of an output mode.

After the measurement at slit 1 we have a mixed state ρ for the output slits 2 and 3 where
3 ρ = |00><00| + 2 |ψ><ψ|

where |ψ> = (|01> + |10>)/√2

OK - that's something of an abuse of Dirac notation here - but I hope it's sufficient to give you the idea of what's going on. The first measurement projects the system into one of 2 possibilities.

 

[Nugatory]

As for question 3), without doing experiments, what would be your answer? The pattern would be that of shooting particles through the slits?

Yes, you get the two-slit pattern from interference between the paths through slits 2 and 3.

 

[franklinhu]

i

Experiments in which there are more than two paths, but the paths aren't realized by physical slits as you're asking, certainly have been done, many times.

Has the slit experiment ever been done with observed electrons? I'm looking for a web reference for such a foundational experiment.

I have found this experiment:
Controlled double-slit electron diffraction
Roger Bach 1,3 , Damian Pope 2 , Sy-Hwang Liou 1
and Herman Batelaan
New Journal of Physics 15 (2013) 033018 (7pp)

This experiment uses slits and electrons but does not do the observation part. How would you explain that on page 4 they show what happens with only a single slit open and it clearly shows a diffraction pattern and not just a diffuse particle pattern? Doesn't that just completely blow away the entire premise of the experiment that the particle view shows up with only a single slit?

 

[bhobba]

Doesn't that just completely blow away the entire premise of the experiment that the particle view shows up with only a single slit?

This wave particle stuff is a myth:
http://arxiv.org/abs/quant-ph/0609163

I gave an analysis from the principles of QM, not the wave particle duality, in another thread where you asked the same question, but here it is again:
http://arxiv.org/abs/quant-ph/0609163

Basically it requires the uncertainty and superposition principle.

Thanks
Bill

 

[Nugatory]

How would you explain that on page 4 they show what happens with only a single slit open and it clearly shows a diffraction pattern and not just a diffuse particle pattern? Doesn't that just completely blow away the entire premise of the experiment that the particle view shows up with only a single slit?

You're thinking that the premise is "wave-particle duality", that with one slot open we should get particle behavior and with two slits open we get wave behavior. Although you'll often hear it stated that way, that's not what quantum mechanics says.

In the modern formulation of QM, quantum objects are neither waves nor particles. Instead they are quantum objects that always display some wave-like behavior (interference, diffraction) and some particle-like behavior (interact with matter at a single point, leave trails in cloud chambers).

The essence of the double slit experiment is that:
1) Whether one or two slits are open, you can always arrange things so that the pattern is built up out of individual dots on the screen.
2) If there is one path possible, you see diffraction.
3) If there are two paths, you also see interference.
4) #1-#3 work the same for both electrons and light (as well as other quantum objects).

 

[jtbell]

How would you explain that on page 4 they show what happens with only a single slit open and it clearly shows a diffraction pattern and not just a diffuse particle pattern?

I don't have access to that paper, but even in classical optics a single slit produces a diffraction pattern:

http://hyperphysics.phy-astr.gsu.edu/hbase/phyopt/sinslit.html#c1

The details of the pattern are of course different from what you get with two slits, but it's not a simple diffuse blob.

 

[mousheng xu]

Yes, you get the two-slit pattern from interference between the paths through slits 2 and 3.

Q1: What happens to slit 1? Particle pattern?

Q2: Well, the particle being a wave exists the same time at each slit with a probability. I assume observing the "yes/no" at one slit collapses the wave to a particle? That said, it seems we won't see a two-slit wave pattern at slits 2 & 3, but a three-slit particle pattern?

Thanks.

 

[mousheng xu]

Let's take for example, an electron gun that shoots one electron at a time towards a wall with 3 slits. and behind it is a photographic plate. if the detector is turned off you'll get a diffraction pattern, that looks pretty much the same as a double slit.

But let's say, we do turn the detector on, which is like a low energy photon emitter. In a perfect world we can't turn on the detector, and only have it detect just one slit, but, regardless the result would be the same. We would have a high intensity region closest to the detector, and further away from it, the interference pattern shows. the reason is the nature of the detector, which is just photons hitting the oncoming electrons, which knocks them out of their superposition and gives them a defined trajectory.

View attachment 91232

Now don't be fooled by the depiction. Those electrons represented as wavy things, aren't really waves nor are they really particles with definite trajectories. Only until the detector has hit them, do they gain a definite trajectory. The electrons exist in a special state where they have the potential to be anywhere in the picture, with a given probability of being detected at a given location. It's only when the photons from the photon emitter interact with an electron, do they gain a trajectory.

Now diving into how the photographic plate actually works, keep in mind, that the photographic plate itself is also a detector. is just a solid surface with indicators, or light switches that flick on when hit with the force of that electron. If one were to create such a plate, where the electron does not stop but still interacts with the plate, then what happens is the plate is like a secondary detector. The plate would collapse the superposition of the incoming electrons, and any electron passing beyond the plate would have a definite trajectory.

However, if we were to only have one electron for example, and once it passes through the first plate, the plate itself changes the trajectory of the passing electron with an unknown momentum, and so when the passing electron were to hit a 2nd plate, the position it lands will be different then when it passed through the first. Given the amount of interaction from the plate, i don't exactly know what it would look like, but i would imagine it being a jumbled mess, basically dots everywhere dispersed randomly.
View attachment 91233

Now I'm not sure if this is the actual case or not, and could be up for debate, but i assume that this conclusion is solely based on the nature of the plate itself, and how it would actually work.

In addition, a little side note. Here is a link that i found that shows multiple slit diffraction, although not shown to have a detector involved, shows what the patterns look like for an increasing number of slits.

http://hyperphysics.phy-astr.gsu.edu/hbase/phyopt/mulslidi.html

OK, thanks for the reply in details. Basically you hold the same view as Dr. Nugatory. Please see my reply to him a moment ago.

Two more questions:

1) I guess the reason I ask the semi-transparent film question is trying to see if the nature of the interaction between the detector & the particle has something to do with the "collapse" of wave to particle. We may be able to find an observer that does not collapse the wave pattern to particle pattern.

2) This question might have no answer than "that's the way it is, don't ask me why": why observing the particle collapse the wave feature into particle?

Thanks.

 

[Nugatory]

Q1: What happens to slit 1? Particle pattern?

If we block slit one, then of course nothing comes through it. If we leave it open with a detector there, then any time the detector clicks a dot will appear on the film, and these dots will be distributed in the single-slit pattern. Because slots two and three are open without detectors, from time to time a dot will appear on the film even though the detector in slot one has not clicked; these will be from particles that didn't go through slot one, and they will be distributed in the slot-two/slot-three interference pattern. After we've accumulated enough dots, our film will show an interference pattern and a bright splotch behind slit one.

Q2: Well, the particle being a wave exists the same time at each slit with a probability. I assume observing the "yes/no" at one slit collapses the wave to a particle? That said, it seems we won't see a two-slit wave pattern at slits 2 & 3, but a three-slit particle pattern?

You're thinking that we'd have a wave until we measure it and then we have a particle. That's not how it works - see Simon Phoenix's post #6 and my reply to franklinhu in post #10. The measurement collapses the wave function down to whatever state is consistent with the measurement. In this case the collapse is to either "went through slot one" if the detector did click or a superimposition of "went through slot two" and "went through slot three" if the detector did not click.

The idea that the measurement turns a wave into a particle is part of the old idea of "wave-particle duality" which was abandoned long ago.