Newbie Question - Double Slit Experiment - 'Interference Patterns'

[SonOfBruze]

I am relatively new to QM and so have a couple of basic questions for you all.

I understand that the double slit experiment creates an ‘interference pattern’. What bothers me, is that it seems like an assumption to me, that the pattern created could *only* have been made by two waves interfering. Is it proven, or is it just an assumption?

Has there been any other research that someone could kindly point me to, that attempts to (I imagine unsuccessfully) look for other ways to create such a pattern? Obviously there has been a lot of work done based upon this assumption, and therefore I believe a lot of work should be done to prove the assumption that it must be wave based.

I could imagine a similar pattern being created by a particle bouncing repeatedly between the double slit wall and the back wall. Has anyone done the experiment and also placed a photographic plate (or whatever it is that picks up where they arrive)…on the back side of the double slit wall? It will probably be a waste of time, but it probably needs to be tried once doesn’t it? I have imagined doing a similar experiment with chalk covered tennis balls, they would bounce and I think create a similar pattern...particularly as they would always arrive on the back wall on a bit of an angle.

I imagine there could be a number of ways that particles could build up such a pattern. Particularly if they have any magnetic qualities at all. Perhaps others with far more understanding of what these objects actually are have thought up other ideas?

I also question the assumptions made about the measuring devices put in place to observe which slit the electron (or whatever was fired) went through. I can imagine that the choice of measuring device used so far, changed the behaviour of the electron …because of some, so far unknown, behavioural quality of the electron. I believe it is an assumption to say that *any* way of observing it will change its behaviour. We would just need to find another way of measuring it (which perhaps is not yet technically possible). It seems to me that all we know is that the way we have currently tried to observe it, has changed its behaviour, nothing more.

Looking forward to being thoroughly flamed for my ignorance :-).

 

[Ken G]

I understand that the double slit experiment creates an ‘interference pattern’. What bothers me, is that it seems like an assumption to me, that the pattern created could *only* have been made by two waves interfering. Is it proven, or is it just an assumption?

I would say it is neither proven, nor an assumption. It is just the pattern that interference makes, and so is called an interference pattern. There is no need to invert the direction of that argument-- our goal is to find a process that explains the outcome, not to assert that the outcome must come from that process. Consider, for example, Newtonian gravity-- Newton was thrilled to find a description that explained the orbits of the planets, he never made any attempt to prove that no other description could also explain the orbits of the planets (indeed such a proof would be impossible-- Einstein's general relativity also explains the orbits of the planets).

Has there been any other research that someone could kindly point me to, that attempts to (I imagine unsuccessfully) look for other ways to create such a pattern? Obviously there has been a lot of work done based upon this assumption, and therefore I believe a lot of work should be done to prove the assumption that it must be wave based.

It's not really clear what it means to be "wave based." There are two equations that I know of that will give you such an interference pattern, one is the "wave equation" of classical fields, and the other is the Schroedinger equation applied to the "wave function." So the word "wave" appears in both, and both have important similarities, but they are not exactly the same equations. Are they both wave equations even though they are different? I'm not sure there is any precise meaning of a "wave" that could answer this, it's just a term that gets used to stress the similarities. I certainly wouldn't be surprised if yet another equation is someday found (if it hasn't been already) that would also give that pattern, and I don't know if it would be called a "wave equation" or not.

I could imagine a similar pattern being created by a particle bouncing repeatedly between the double slit wall and the back wall. Has anyone done the experiment and also placed a photographic plate (or whatever it is that picks up where they arrive)…on the back side of the double slit wall? It will probably be a waste of time, but it probably needs to be tried once doesn’t it?

There are lots of experiments that need to be tried once, the trick is to get someone who is willing to make the effort to do it-- knowing that it is probably a waste of time. People don't like to waste time, that's the issue.

I have imagined doing a similar experiment with chalk covered tennis balls, they would bounce and I think create a similar pattern...particularly as they would always arrive on the back wall on a bit of an angle.

I don't see why you think tennis balls would do that, but you may be assured that if there is anything you can do with tennis balls that would make that pattern, it would be quite important to be able to say what that is.

I can imagine that the choice of measuring device used so far, changed the behaviour of the electron …because of some, so far unknown, behavioural quality of the electron.

People have tried very hard to "trick" the electron into revealing which slit it went through and still get the pattern, but no one has ever succeeded. A good thing to look at is called "delayed choice quantum eraser" experiments-- if you study those and understand them, it might well change your mind that there should be some way to get the electron to give up that information without spoiling the interference pattern.

It seems to me that all we know is that the way we have currently tried to observe it, has changed its behaviour, nothing more.

No one would say anything different. All we know is that the ways we have currently tried, there's no way to avoid conservation of energy, there's no way to go faster than light, and there's no way to get an electron to tell you which slit it went through without ruining the interference pattern. None of that means any of those things are impossible, but physics must account for what we have found to be the case-- not what we have proven must be the case.

 

[DrChinese]

... I understand that the double slit experiment creates an ‘interference pattern’. What bothers me, is that it seems like an assumption to me, that the pattern created could *only* have been made by two waves interfering. Is it proven, or is it just an assumption?

Has there been any other research that someone could kindly point me to, that attempts to (I imagine unsuccessfully) look for other ways to create such a pattern? Obviously there has been a lot of work done based upon this assumption, and therefore I believe a lot of work should be done to prove the assumption that it must be wave based.
...

Welcome to PhysicsForums, SonOfBruze!

I would say it is proven, although each person might have their own threshold for proof.

The only variable that enters into the equation is the amount of knowledge you have as to which slit the particles go through. Suppose I place polarizing filters at each slit. If they are set to be parallel (0 degrees difference), then I get no knowledge of which slit the particle goes through. There IS an interference pattern. If they are set to be crossed (90 degrees difference), then I potentially get complete knowledge of which slit the particle goes through. Of course, now there is NO interference pattern. And as I vary the relative angle setting of the polarizers between 0 and 90 degrees, the pattern morphs from one to the other.

So clearly, bouncing off one screen to the other has nothing to do with anything as this element doesn't change in this setup! The ONLY thing changing is our potential knowledge of which slit. Note that we don't need to actually KNOW which slit the particle went through for the pattern to change. Merely that we could obtain it with the setup is enough.

 

[Ken G]

I would say it is proven, although each person might have their own threshold for proof.

What I meant that we cannot say it is proven is that descriptions that work are rarely unique. We can prove that the interference description works, and works very well. That can never be regarded as a proof that interference "actually causes the pattern", because we don't get to know what actually causes the pattern, and any proof that we do must include a proof that no other possible explanation could work. This distinction becomes important when some other description does equally well without invoking any language of interference.

I don't know exactly how that might be done with the two-slit pattern, but I would not rule out that it could (though I agree with your doubts that bouncing around between walls could work), because there are many examples of this type of thing in other situations. For example, in a delayed choice quantum eraser experiment, as you well know, you get a non-interference pattern from the two slits on the detector screen. So you might be tempted to conclude "this proves that when we know which slits the photons went through, they do not interfere." Yet if we erase that knowledge and correlate hits with an entangled pair, we find that the original non-interference pattern can equally well we interpreted as two double-slit interference patterns that are slightly offset and cancel each other's interference in the net. Thus we could equally well embrace language like "two-slit interference does occur even when we know which slit the photons go through, but that knowledge causes the way that interference occurs to be offset in such a way that washes out the interference in the net pattern." Washing out interference that does occur, and having the interference not occur, are two very different ways of talking about the same outcome-- so we cannot prove what actually occurred there.

 

[SonOfBruze]

Oooh busy Christmas period finally over…..OK back to my question.

I can see firstly that I have an awful lot of reading to do, as I don’t yet fully understand some of the comments made above and why they so clearly prove that the bouncing idea is impossible. (as mentioned before I am a newbie)…..so I will hit the books at my earliest convenience. I really do appreciate your responses…they all help me learn.

I don't see why you think tennis balls would do that, but you may be assured that if there is anything you can do with tennis balls that would make that pattern, it would be quite important to be able to say what that is.

Alright, I will try :uhh:.

My understanding has always been that the central band of the interference pattern, the one between the two slits, was the really impossible one to create right?

I have not yet found a graphic displaying the interference pattern from a double slit experiment with the slits superimposed on the pattern itself, to get an idea of scale…..so I really don’t know where the slits are found in relation to the interference pattern. A number of possibilities come to mind though, and some I imagine would be much easier to create with tennis balls than others. I have drawn up two possibilities in the attachments (SlitsInFirstDarkBand.png and LongSlitsInMiddle.png).

Which one or neither is correct…there are obviously many other possibilities here. Is there a graphic in a journal article somewhere I could look up?

My thought was, that to a photon traveling through one of the slits….that the slits would appear as large cavernous long tunnels rather than short slits, due to the size difference between the two. I therefore can see the photons bouncing off the outer inside walls of the slits, and then landing on the back wall in or around the centre. Obviously if the photon was to bounce too early in the tunnel – it would possibly just bounce around inside the tunnel until it loses momentum. However if it hit say in the last ½ of the tunnel on the outer wall, it could bounce back to the centre of the back wall….of course….assuming they bounce .

I therefore could imagine replicating this with tennis balls…..with not so much two slits but two short tunnels …that would allow the tennis balls to bounce off the inside of the tunnel and ideally hit the back wall in the very middle, in-line with the ‘gun’.

Please see attachment 'Bouncing.png' to better explain the above.

Obviously the angles in this image are not right, but it was just to help explain how I thought the pattern could be created with tennis balls.

I imagine I can only see that tennis balls would create such a pattern as I do not as yet fully understand some key aspects of the interference pattern, or some key aspects of how tennis balls bounce …. I really need to read real journal articles rather than just popular science books eh?

 

[bohm2]

It's not exactly QM but some of the behaviour of these macroscopic objects (silicon droplets) seem quantum mechanical, like path memory and diffraction pattern (see end of video for double-slit experiment):


Yves Couder . Explains Wave/Particle Duality via Silicon Droplets
https://www.youtube.com/watch?v=W9yWv5dqSKk

 

[SonOfBruze]

Ahhh I attempted in my previous post to use the 'attachment' option, but cannot seem to locate the attachments after the posting, so have linked to them below.

http://users.tpg.com.au/hreeves/LongSlitsInMiddle.png
http://users.tpg.com.au/hreeves/SlitsInFirstDarkBand.png
http://users.tpg.com.au/hreeves/Bouncing.png

Thanks for the video link bohm2. It was very interesting.

 

[Strafespar]

It's great that you are trying to make what is called an interpretation, but from my understanding your model is flawed overall because it is too provincial (you are just looking at one experiment). First off if you think about a photon as a particle, which you are, the photon could eventually go through at all possible angles of entry into the two slits and what you would see on the back screen would be an entirely lit up screen. But it's even more convoluted than that (trying not to be Pauli)...try to think about all the implications of what you are saying. I am not trying to discourage you of interpretations because QM is stuck in the middle between a couple right now.

 

[SonOfBruze]

Thankyou so much for your feedback - I know I am such a newbie that I shouldn't even be commenting, but I do find this all very interesting. I did originally think that the whole back screen would be entirely lit up, but eventually decided I didnt think it would be. Certainly the middle band would be somewhat wide to account for all those that bounced off the 2nd half of the 'tunnel'...in all random spots, but any that passed straight through the tunnel would be on a slight angle when they hit the back wall...and therefore I thought would bounce neatly in bands. ie we would still have all the outer bands clearly defined...and one central band that was perhaps wider than the others.

I did worry about all those that bounced in the initial part of the tunnel, but felt that they would bounce back inside the tunnel again and hopefully would lose momentum and therefore never hit the back screen.

I agree I am just looking at one experiment only...and I imagine there are lots of others that support or perhaps don't support the current theories, I was just initially interested in the assumptions I felt I saw being drawn primarily from this experiment only. At least that is the impression I got from the articles I have thus far read...which is perhaps the only problem here :-).

 

[SonOfBruze]

I decided to try out my idea and built a double slit experiment to use with golf ball sized foam balls. I was able to create what looks to me to be an 'interference pattern'. It is loaded on youtube now, if you are interested in pointing out the mistakes I have surely made, you can access it at .

 

[yuiop]

First, on the bouncing on the back of the slit screen thing, it would be very easy to experimentally test that idea by placing obstacles, deflectors and or light absorbing materials on the back of the screen. I suspect it will make very little difference to the interference pattern as long as the direct path is not obscured.

I decided to try out my idea and built a double slit experiment to use with golf ball sized foam balls. I was able to create what looks to me to be an 'interference pattern'. It is loaded on youtube now, if you are interested in pointing out the mistakes I have surely made, you can access it at .

I had a quick look at your interesting practical simulation. The first thing that is obvious is that the pattern you achieve on the screen has very little to do with the bouncing off the back of the slit screen but is mainly due to bouncing off the slit tunnels. Quite often slits are no more than etches in a film coating on glass and perhaps a lot less tunnel like than your simulation suggests, but *maybe* in terms of wavelength even a thin film may appear tunnel like. Another thing to consider is the fact that an interference pattern can be produced by placing a single needle or wire in the path of a laser shone at a screen. In that case there is no outside edge of a slit tunnel for particles to rebound off, to create an interference pattern.

 

[Petyab]

Your experiment is perhaps more of a showing of particle interference. Between the three groups of marks made there is the possibility that a more sticky chalk would create a pattern between the larger groupings. This would be more like a wave interaction showing. The notion of the double slit experiment is that the electron can pass through either slit or both at the same time. Now here's where I may be slightly wrong. The electrons shot at either slit can excite electrons on the way and the observation of an interference pattern may be partially from these excited electrons. I have to think about that more but it seems a possibility.

 

[SonOfBruze]

Thanks for your response yuiop.

The first thing that is obvious is that the pattern you achieve on the screen has very little to do with the bouncing off the back of the slit screen but is mainly due to bouncing off the slit tunnels.

It is really only the central band in my experiment that was created by the bouncing inside the tunnels. The back of the slit wall showed where some balls landed that went straight through the tunnels (without touching the sides) and then hit the back wall directly before bouncing back to the back of the slit wall.

I felt that if I had more wood, bouncier balls or a more powerful gun, the balls would have then hit the back wall again and made extra bands. There were some marks on the back wall in the very bottom right and left that were created from balls that went completely through the tunnel (without touching the sides), hit the back wall, bounced back to the back of the slits, and then again hit the back wall low to the ground.

I had thought that if the interference pattern we observe in real double slit experiments is created by waves, that we should see nothing on the back side of the slit wall - as the two waves perhaps wouldn't be interfering there. At least the below videos, don't appear to have much interference on the back side of the slit. The wave from the closest slit would have to be too powerful there to experience much effect from the other wave?


http://www.youtube.com/watch?v=dNx70orCPnA&feature=related

Quite often slits are no more than etches in a film coating on glass and perhaps a lot less tunnel like than your simulation suggests, but *maybe* in terms of wavelength even a thin film may appear tunnel like.

I would have thought that even the smallest etch in a film coating on glass would have to contain billions of atoms (?) and therefore would be substantially larger in size than an electron?

Again though, I am such a newbie that I am happy for you to point out my faults here.

 

[scijeebus]

I am relatively new to QM and so have a couple of basic questions for you all.

I understand that the double slit experiment creates an ‘interference pattern’. What bothers me, is that it seems like an assumption to me, that the pattern created could *only* have been made by two waves interfering. Is it proven, or is it just an assumption?

There's math you can use to describe the double slit experiment which is essentially just probability matrices and does not use any wave mechanics, and it achieves the same results, but Schrodinger argued that you need some kind of visual component to explain the "why".

Has there been any other research that someone could kindly point me to, that attempts to (I imagine unsuccessfully) look for other ways to create such a pattern? Obviously there has been a lot of work done based upon this assumption, and therefore I believe a lot of work should be done to prove the assumption that it must be wave based.

You can do it by putting a wave through two slits. You can use light or water or w/e.

I could imagine a similar pattern being created by a particle bouncing repeatedly between the double slit wall and the back wall. Has anyone done the experiment and also placed a photographic plate (or whatever it is that picks up where they arrive)…on the back side of the double slit wall? It will probably be a waste of time, but it probably needs to be tried once doesn’t it? I have imagined doing a similar experiment with chalk covered tennis balls, they would bounce and I think create a similar pattern...particularly as they would always arrive on the back wall on a bit of an angle.

Would such a random process create a perfect pattern that perfectly matches other predicted mechanics?

I imagine there could be a number of ways that particles could build up such a pattern. Particularly if they have any magnetic qualities at all. Perhaps others with far more understanding of what these objects actually are have thought up other ideas?

If you have some math and some new theories, try patenting them.

I also question the assumptions made about the measuring devices put in place to observe which slit the electron (or whatever was fired) went through. I can imagine that the choice of measuring device used so far, changed the behaviour of the electron …because of some, so far unknown, behavioural quality of the electron. I believe it is an assumption to say that *any* way of observing it will change its behaviour. We would just need to find another way of measuring it (which perhaps is not yet technically possible). It seems to me that all we know is that the way we have currently tried to observe it, has changed its behaviour, nothing more.

It's not really "change", it's more of "correlate". When an electron is observed, it's probability simultaneously correlates to that of a specific point because the act of measuring creates the equation that the electron can only be at a specific point, which time is not a part of. Or more precisely, the photon we observe does.