How Are Slits Made for the Double Slit Experiment?

[rawtrax]

Hello all,

I have some questions in regards to the double slit experiment. Now I apologise if some of these seem very basic, likewise if they are very hard to explain, however it is something I would like to get a better grasp on.

First, you are firing out of said ‘gun’; let’s say a proton, through a system of just one slit or two or more. The question instantly that springs to my mind is if you are firing something so small as to be at the molecular level how is it you can build a screen with a slit small enough for the proton to pass through? What is the screen made out of? How big are the slits? How big is the screen?

Secondly, the main result of this experiment seems to be that if you observe the results they will be different from results not observed. My question is this; If you can see the results of the experiment you have not observed, then surely you have observed them, just later in time?

This follows to my next question: What is it that records the experiment? As I gather measuring devises are a no-go as they interfere with the particles that fire from the gun. So it’s a screen at the tail end of the experiment environment that records where each particle has landed, surely is this not another method of measurement, if not, why not?

I hope that made sense, I am no quantum physicist by a long shot, so if I have misunderstood things here then please let me know.

Thanks in advance!

 

[bdavlin]

I believe the observer alters the particle moving through the slit. I have always wondered if this is the mere act of observing (in the Cartesian sense of the observer naturally warping reality), or if the act of observing is taking a measurement by injecting energy at the passing particle. I would love clarification.

 

[DrChinese]

1. ...The question instantly that springs to my mind is if you are firing something so small as to be at the molecular level how is it you can build a screen with a slit small enough for the proton to pass through? What is the screen made out of? How big are the slits? How big is the screen?

2. Secondly, the main result of this experiment seems to be that if you observe the results they will be different from results not observed.

Welcome to PhysicsForums, rawtrax!

1. Usually, the slits are on the order of a wavelength apart and a wavelength wide. That is for whatever you are sending through. This is on the rough order of maybe nanometers in typical experiments.

2. The results change according to whether or not you know which slit the particle passes through. There are ways to accomplish this which might be called non-destructive and allow the particle to still make contact with the screen. There is an observation either way.

 

[Drakkith]

Don't get confused about the observing part. There are two different observations going on. The first is when you observe where the particles hit the detector. If you observe only this then you will see the classic interference patterns that result from the experiment. However, when you try to make the 2nd observation, which is observing which slit the particle goes through, the act of observing that will cause the results of the 1st observation to be different. In that case the interference dissapears.

(I might be wrong on this next part, so someone correct me if i am.)
If you want to be more correct, the concept of a "particle" is misleading. All matter can be described as a wavepacket, which removes the "duality" and explains everything as waves.

 

[lowing99]

Hi Rawtrax

I don't know if you watched the recent BBC Horizon programme on the double slit experiment, it was excellent, you can still view it online at You tube.

One of the chaps was explaining how his quantum computer worked and remarked even a single electron passing through the quantum electron fields seemed to act as an observer and caused the binary quantum hard drive from working.

It was well worth watching in my opinion.

Best

Colin

 

[lowing99]

Hi rawtrax again,

Forgot to mention the programme also said in experiment that when the photon is observed and recorded, it does the usual thing but it actually reverts to a wave pattern if you observe the photon and record it's slit position but then delete the information before the photon hits the plate.

 

[morrobay]

Don't get confused about the observing part. There are two different observations going on. The first is when you observe where the particles hit the detector. If you observe only this then you will see the classic interference patterns that result from the experiment. However, when you try to make the 2nd observation, which is observing which slit the particle goes through, the act of observing that will cause the results of the 1st observation to be different. In that case the interference dissapears.

Does the 2nd observation of detection involve using photons or other particles that can
interact with the electron - and alter the path of the electron ?

https://www.physicsforums.com/showthread.php?t=418579

 

[Drakkith]

Don't get confused about the observing part. There are two different observations going on. The first is when you observe where the particles hit the detector. If you observe only this then you will see the classic interference patterns that result from the experiment. However, when you try to make the 2nd observation, which is observing which slit the particle goes through, the act of observing that will cause the results of the 1st observation to be different. In that case the interference dissapears.

Does the 2nd observation of detection involve using photons or other particles that can
interact with the electron - and alter the path of the electron ?

https://www.physicsforums.com/showthread.php?t=418579

I don't believe there is a way of observing the electron without interacting and changing it at least slightly.

 

[morrobay]

I don't believe there is a way of observing the electron without interacting and changing it at least slightly.

Can this interacting and changing of the electron (disregarding observation)
during detection process destroy the interference pattern ?

 

[Chopin]

Secondly, the main result of this experiment seems to be that if you observe the results they will be different from results not observed. My question is this; If you can see the results of the experiment you have not observed, then surely you have observed them, just later in time?

This follows to my next question: What is it that records the experiment? As I gather measuring devises are a no-go as they interfere with the particles that fire from the gun. So it’s a screen at the tail end of the experiment environment that records where each particle has landed, surely is this not another method of measurement, if not, why not?

There always has to be some kind of 'observation'--that's the only way you can find out anything about what happened in your experiment. The thing that matters is exactly what you're observing. In the two-slit experiment, there are two different things you could measure. One measurement you can make is the strength of the particle field on the screen behind the slits. The other is the number of particles going through the detectors on the slits. The difference between these two measurements is the type of information that you get by making them.

For the screen, the only piece of information it gives you is where the particle was when it finally hit the screen. It doesn't tell you anything about what the particle was doing before it hit the screen--most notably, it doesn't tell you which slit the particle went through on the way there. Whether the particle went through Slit A or Slit B, the pattern on the screen will be exactly the same. The detectors on the slits, on the other hand, will tell you which slit the particle went through. If you use the detectors, it gives you more information than you would have if you just looked at the screen. The fact that you're performing a measurement which makes this piece of information known is what changes the outcome of the experiment.

 

[bdavlin]

But does the act of observation without a detector machine alter the reality? In essence is the Cartesian view of the observer warping reality.

 

[Chopin]

I believe the observer alters the particle moving through the slit. I have always wondered if this is the mere act of observing (in the Cartesian sense of the observer naturally warping reality), or if the act of observing is taking a measurement by injecting energy at the passing particle. I would love clarification.

But does the act of observation without a detector machine alter the reality? In essence is the Cartesian view of the observer warping reality.

I'm not sure what you mean by 'Cartesian' in this context.

The idea that the observer injects some energy into the system and nudges the particle around is an old idea that was introduced early in the development of quantum mechanics (by Dirac, I believe), to try to explain this bizarre concept of the measurement affecting the outcome. However, as quantum theory has progressed, it has become clear that this idea is not actually necessary for the behavior of quantum mechanics to take place.

It's probably true that most real measurements you can do will nudge the system around a bit, and inject some energy into it, but even if you could construct a perfect measuring tool which could measure the system without doing that, the results of quantum mechanics would still take place. The important thing is what pieces of information become known, and what pieces of information stay unknown. If a piece of information becomes known, by any means at all, it will collapse the wavefunction. If the piece of information stays unknown, then the wavefunction continues to interfere with itself.

 

[DrChinese]

The idea that the observer injects some energy into the system and nudges the particle around is an old idea that was introduced early in the development of quantum mechanics (by Dirac, I believe), to try to explain this bizarre concept of the measurement affecting the outcome. However, as quantum theory has progressed, it has become clear that this idea is not actually necessary for the behavior of quantum mechanics to take place.

It's probably true that most real measurements you can do will nudge the system around a bit, and inject some energy into it, but even if you could construct a perfect measuring tool which could measure the system without doing that, the results of quantum mechanics would still take place. The important thing is what pieces of information become known, and what pieces of information stay unknown. If a piece of information becomes known, by any means at all, it will collapse the wavefunction. If the piece of information stays unknown, then the wavefunction continues to interfere with itself.

As Chopin says: the idea that the electron is physically disturbed by observing "which slit" - and that causes the interference to disappear - is somewhat misleading and unnecessary to the explanation. It is all about knowledge, because you can construct scenarios whereby the hypothesized physical disturbance must violate special relativity.

 

[DarioC]

"It's probably true that most real measurements you can do will nudge the system around a bit, and inject some energy into it, but even if you could construct a perfect measuring tool which could measure the system without doing that--

If no one can do that experiment, then we don't know for sure, right?

"construct scenarios"

Scenarios are interesting to do, but they don't prove anything, right also?

DC

 

[DrChinese]

"construct scenarios"

Scenarios are interesting to do, but they don't prove anything, right also?

Well, they are actual experiments, so...

For example, switch to photons and use polarizers or PBSs. Send light randomly polarized either V or H through a double slit and you get interference. Add a polarizer at each slit (V at one, H at the other) and the interference disappears. Now, how does a V polarizer change a V polarized photon (or H and H)? Answer: it doesn't!

 

[DarioC]

...Randomly polarized either v or h... bothers me a little bit. Does that mean either polarized one or the other or changing randomly with time? Could you be more specific or give me a reference link to that experiment. I would like to read it.

Pretty obviously I need to study more as I have just recently had an old interest additionally piqued by the Italian experiment of "slightly" interfering with one slit and then increasing the interference to see the results.

http://apl.aip.org/resource/1/applab/v97/i26/p263101_s1?isAuthorized=no

DC

 

[DrChinese]

...Randomly polarized either v or h... bothers me a little bit. Could you be more specific or give me a reference link to that experiment. I would like to read it.

Pretty obviously I need to study more as I have just recently had an old interest additionally piqued by the Italian experiment of "slightly" interfering with one slit and then increasing the interference to see the results.

DC

Also, you can see the issue even better when you vary the angle theta between the L and R slits. When the theta=0, there is interference regardless of the particular polarization of the photon source. When theta=90 degrees, there is none. So obviously the polarizer itself is doing nothing. It must be the polarizer in conjunction with something else: namely the other polarizer, which makes no sense if you are postulating a physical interaction with the polarizer.

As to a reference: Debil's Avocado has a cool link to a simulation of the above. I thimk.

 

[Chopin]

"It's probably true that most real measurements you can do will nudge the system around a bit, and inject some energy into it, but even if you could construct a perfect measuring tool which could measure the system without doing that--

If no one can do that experiment, then we don't know for sure, right?

"construct scenarios"

Scenarios are interesting to do, but they don't prove anything, right also?

DC

Idealized experiments serve as useful limiting cases to what we can do in the real world. This is even the case in classical mechanics--we will never be able to make a perfectly round sphere and roll it around on a 100% frictionless table, so by your logic there's no use trying to discuss what would happen in those cases. However, the laws of motion that we derive by thinking about such idealized objects allow us to make approximations about what real spheres rolling around on real tables will do.

And, crucially, the closer we make our spheres to being perfectly round, and the closer we make the table to being 100% frictionless, the closer the results of our actual experiment will agree with the results of our idealized one. In many cases the discrepancy between the two is so small that it doesn't matter. So the ideal experiments allow us to examine the basic physical principles at work, and ignore all the imperfections.

In the same way, we know enough about quantum mechanics to say that an "ideal" measurement instrument, which doesn't physically interact with the system at all, will still lead to wavefunction collapse for the reasons that have been stated. We will never be able to construct such a perfect instrument, but it serves as a useful limiting case for real experiments.

 

[DarioC]

I suppose then that I will have to do considerably more searching for informative reading on the subject to see just how "close" experimenters have come to a totally non-interacting detector/sensor, whatever they are using.

I am seriously short of knowledge, at this time, on exactly what type of device would (in a real experiment) have a negligible effect on a particle's passage. From what I have read recently I am sure there is lots of good information available out there.

Once a hardware guy, always a hardware guy, I guess.

DC

 

[DarioC]

Proposed experiment of which-way detection by longitudinal
momentum transfer in Young's double slit experiment
Masanori Sato
Honda Electronics Co., Ltd.,

Direct quote from first paragraph in paper:

"Heisenberg's uncertainty relations and
complementarity have been discussed previously
[1-12] as being the reasons why it has not been
possible to determine the direction in which a
photon travels in Young's double slit experiments
[1-3]."

Most recent work referenced in paper.

[14] H. M. Wiseman, "Bohmian analysis of
momentum transfer in welcher Weg
measurements" Phys. Rev. A, 58, (1998),
1740-1756.

Two weeks of searching the Internet has produced zero examples of the use of a real physical detector/monitor successfully applied to determining the "which way" path in the double slit experiment.

 

[WoodenBadger]

but even if you could construct a perfect measuring tool which could measure the system without doing that, the results of quantum mechanics would still take place. The important thing is what pieces of information become known, and what pieces of information stay unknown. If a piece of information becomes known, by any means at all, it will collapse the wavefunction. If the piece of information stays unknown, then the wavefunction continues to interfere with itself.

This does not make much sense to me.

What do you mean 'if we could construct a perfect measuring tool which could measure the system without doint that'? I mean, how could it work, the detector showing measurings about the electorns(that is, the state of the electrons affects the state of the detector), but the state of the detector does not affect the electrons? I believe the old action-reaction principle holds true for QM systems too, right? I believe that, by the knowledge we have on physics, we can't. This is theoretically explained, as you pointed it out, by information theory, but in nature, in order to obtain information you must exchange energy(remember Maxwell's demon and entropy?). I believe that, if you are to obtain information with the whole system about which slit the electron goes through, you exchange energy/information with it in such a way that causes the result to change anyway.

I mean, think about it. You set the 'slit detector' at some point in 3D space, right? This detector doesn't just 'interact' with the system in any random way. Its purpose is to spit out information about which slit the electron will go through, or went through. So it must be structured in such a way in order to predict the position/momentum of the electron, at the point where it is measured, with accuracy enough to make us say we do, in fact, hold that information. At any point we do the measurement, we must know the position/momentum with such accuracy in order to allow us to say if the electron wave have gone through another point in space(the slit). I imagine that, at the point where we place the detector, we can obtain, as information, a "cone" that shows us where the particle is probable to "go"(or have gone). That cone should be narrow enough in order for us to say we have the information about which slit the wave has gone through, otherwise the information just isn't correct. How is it possible to make any detector that does not become part of the system, if it extracts such information about the system? They exchange information, thus they interact, thus exchange energy. It's not just "any" interaction, it's an interaction purposed to extract certain information. Can that really be done without exchanging energy? I believe the correct phrase is that, no matter how advanced the technology of the physical detector is, the uncertainty in QM is always a limit.

I have as a goal to try to make some simple simulation of this as a computer program, seeing as a program computer games in my spare time, to see how a hypothetical detector that extracts "just" enough information would change the behaviour of electron waves in a double slit setup, but I haven't found enough info on the internet about how that exp. is usually setup anyway.

For example, switch to photons and use polarizers or PBSs. Send light randomly polarized either V or H through a double slit and you get interference. Add a polarizer at each slit (V at one, H at the other) and the interference disappears. Now, how does a V polarizer change a V polarized photon (or H and H)? Answer: it doesn't!

The H polarizer at one slit which change the V photons passed from that slit and the V polarizer at the other slit will change the H photons passed from that slit, won't it? Eventually, you will have only V photons go out of one slit and H photons of the other exactly due to the polarizers. If you didn't have the polarizers, you would have V or H photons go through the slits randomly. Am I wrong about this?

 

[Chopin]

Richard Feynman gives a very good example of how this measurement can be accomplished without touching the particle in The Feynman Lectures (Vol. III Ch. 5, if you have access to a copy). It's a little different than the double slit experiment, but still illustrates the crucial principle, that measuring a particle can change the outcome.

The thought experiment involves a piece of equipment called a Stern-Gerlach apparatus. This is essentially just a strong magnetic field, through which you fire a beam of particles. For particles with spin, like electrons, this will split the beam into two beams--one with positive spin, one with negative spin, since they respond oppositely to the magnetic field. An opposite magnetic field can then merge these split beams back together again into a single beam.

The experiment consists of putting a series of these machines together. Imagine firing particles into one machine, which splits them apart, and then blocking one of the two beams before they can merge back together. The resulting beam has half the intensity of the incoming beam (since we blocked half of the particles), but we now know that all of the particles have the same spin. This can be verified by running this beam through another Stern-Gerlach apparatus. If we attempt to block the same half of the beam in the second apparatus, we will find that the output intensity doesn't change at all--there were no particles in this beam. Similarly, if we block the other beam in the second apparatus, the entire output beam goes away--all of the particles were in this beam.

This experiment shows us that the particles "remember" which state they were in--since the first apparatus filtered the particles, the second apparatus will only produce one beam. However, now imagine placing a third apparatus in between the first two, which has a magnetic field that is 90 degrees off from the other two. This will produce a pair of beams as well, but the split is now due to their spin on a different axis. Thus there will be two beams, not just one. Now imagine blocking one of the two beams in this apparatus. We have now produced a beam of particles which are filtered to this new spin state--in essence, we have observed its spin on that axis.

The crucial point is that now, if we put this filtered beam through the third apparatus, we will discover that it will produce both beams. The act of observing the spin on the second axis has erased the particles' knowledge of what their original spin was. However, if we do not block either of the beams in the second apparatus, then the third apparatus will only produce one beam again. Thus it isn't the fact that there was a second apparatus in the way that changed the behavior, it was the fact that we filtered the beam in that state, forcing the particles into eigenstates of that spin axis.

It's difficult to describe this experiment without using pictures--if you can find a copy of the Feynman lectures it will be much clearer than the preceding description. But the basic idea is that putting the particles through a magnetic field doesn't change anything about their spin, it just deflects them in one direction or another. But by using this field as a way to "observe" the particles, the same quantum mechanical effects are observed.

 

[DarioC]

"The thought experiment..."
DC

 

[Chopin]

The point is, it's a mechanism of measuring which doesn't involve touching the particle. If you want an experiment you can actually perform, you can easily convert the logic of this experiment into one with polarizing filters, which you can perform in your own garage.

Take a polarizing filter, and use it to filter incoming light. The resulting light is 50% of the intensity of the incident light, which can be explained by viewing the light as a combination of vertically polarized and horizontally polarized light. The filter blocked (say) the horizontal light, and transmitted the vertical light. You can verify this by putting a second filter after it, oriented in the same direction. There will be no drop in the intensity of the light, because it is all going in the same direction as the filter. Similarly, if you place the second filter horizontally, it will block all light, because all of it was vertical.

Now put a third filter in between the other two, this time at 45 degrees. It will again block part of the light and let part through, this time polarized at 45 degrees. However, now you will find that part of this light goes through the third filter if it is vertical, but part of it will also go through if the third filter is horizontal. The presence of this intermediate filter, and the measurement we made of its 45 degree polarization, caused it to "forget" that it was vertically polarized, and so part of it goes through the horizontal filter.

This is exactly the process that liquid crystals use to twist polarized light to a different orientation--i.e. the monitor you are reading this on right now is using this process to produce the image that you are looking at.

 

[WoodenBadger]

The point is, it's a mechanism of measuring which doesn't involve touching the particle.

I just have one question...what do you mean 'not touching the particle'? This is a mechanism which applies magnetic force to the particles, splits beams using those forces, blocking them...how is it not 'touching' it? That's what 'touching' means in respect to particles(and in general really), applying forces to them. What am I missing?

 

[Chopin]

But it does so in a predictable way...instead of the particle knocking into another particle and flying off in a random direction, it is just smoothly deflected. People usually worry about the measurement process introducing some degree of randomness because the particle interacts with the detector in a random way. This shows that it's possible to become aware of the particle's state in a way that doesn't involve destroying that state.

 

[DarioC]

Seems there are all types of interesting effects are found when using multiple polarizing filters and if you say that it shows that the which way experiment works exactly as advertised without actually doing the experiment with a detector who am I to say otherwise.

Thought experiments are a very powerful way of understanding different processes, but in the end, still only exist in the minds of the creators and the readers.

That my very first attempt to delve into a supposedly "well proven" light phenomena ran immediately into a brick wall showing that the experiment had never been successfully done is disconcerting to say the least.

Even more discouraging is the representation, as fact, of using non-existing detectors to do an experiment that has not been done successfully, on a forum that is saturated with young people asking questions. Not using proper qualifications such as, "most evidence points toward, this has not yet been proven by experiment but similar experiments puts it on solid ground etc., is, in my opinion something that should not be acceptable. Unfortunately it seem to be a very common occurrence in every aspect of life today.

Personally I have learned a lot in this little endeavor and I hereby leave it to your imagination to speculate on the most important point realized.

I see the sun has finally come out here so I am off to deal with the Classical world and see if I can get my antique airplane in the air once again.

DC

 

[PhysDrew]

That my very first attempt to delve into a supposedly "well proven" light phenomena ran immediately into a brick wall showing that the experiment had never been successfully done is disconcerting to say the least.

I've had the same thing. I've been trying to track down some papers on the topic with no success. I have to agree with Dario on this one. Even though the result may be a foregone conclusion, I'd like to read the experimental verified results on this one.

 

[Chopin]

The variation on the drugstore sunglasses "trick" is always interesting, but the fact that the actual "which way' experiment has not been done is not changed. Thought experiments are a very powerful way of understanding different processes, but in the end, still only exist in the minds of the creators and the readers.

Your main topic in this post is that the experiments commonly held up to explain these phenomena have not actually been performed. However, when provided with an actual experiment that can be performed, you dismiss it out of hand?

That my very first attempt to delve into a supposedly "well proven" light phenomena ran immediately into a brick wall showing that the experiment had never been successfully done is disconcerting to say the least.

Even more discouraging is the representation, as fact, of using non-existing detectors to do an experiment that has not been done successfully, on a forum that is saturated with young people asking questions. Not using proper qualifications such as, "most evidence points toward, this has not yet been proven by experiment but similar experiments puts it on solid ground etc., is, in my opinion something that should not be acceptable. Unfortunately it seem to be a very common occurrence in every aspect of life today.

I think this is quite often the case in science. The actual experiments that demonstrate these phenomena often require so much interpretation that the central idea is lost. These simpler idealized experiments serve to show how the theory works in a way that doesn't get lost in the real-world details. Their purpose is to put a clear, enduring idea into your mind of how things work that is easy to remember. For instance, people will say "if you drop a hammer and a feather on the moon, they'll hit the ground at the same time", as an easy way to discuss the concept of air resistance and gravity. Prior to the mid-20th century, nobody had actually done this, but when they finally did go to the moon and do it, they knew what would happen from countless other simple, boring, less obvious experiments back home. But the reason people used that example, even before it had actually been done, is that it's something that sticks in your head and makes it easy to remember a fundamental bit of physics.

That said, it certainly would be nice if people made it clearer which experiments have actually been done, and which are thought experiments only. To his credit, Feynman says right up front that nobody has performed the Stern-Gerlach experiment I mentioned in my post. Unfortunately the thought experiments often get misattributed as real experiments over time (Galileo dropping weights off of the Leaning Tower of Pisa is a famous example), simply because they stick in your head so well.

At the end of the day, though, what exactly is your point in all of this? The fact that the double slit experiment has not been performed in exactly this way cannot be used as proof that its predictions are not true. Enough other experiments have confirmed its results, including the simple one I've mentioned above, and the one that DrChinese outlined earlier. All that's really been revealed after all of this discussion is that history often becomes distorted over time through people mistakenly repeating half-truths, which is a pretty universal human trait, even outside of science. Did George Washington really chop down his parents' cherry tree? Was Napoleon really unusually short? Did people in Columbus's time really think the Earth was flat? No, but the images persist, precisely because of how simple and evocative they are.

 

[DarioC]

My comment on the polarizing filters has been removed from my reply as it was snide and uncalled for; perhaps a result of frustration and too much coffee.

My judgement on the misrepresentation in reference to "detectors" will remain the same until my lights go out.

DC