Talk:Double-slit experiment

From Wikipedia, the free encyclopedia

WikiProject Physics This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.
Start This article has been rated as Start-Class on the assessment scale.
High This article is on a subject of High importance within physics.

Help with this template This article has been rated but has no comments. If appropriate, please review the article and leave comments here to identify the strengths and weaknesses of the article and what work it will need.

Contents

[edit] Notice

Please put new communications at the bottom of this talk page. If you put new stuff up here with the stuff from 2004 it is likely to get ignored.

(Please archive anything else not needed to archive 3. The others are full. P0M (talk) 01:46, 21 November 2007 (UTC)}


[edit] cleanup and citations tags

I have repeatedly asked for guidance from whomever put those tags up. Nothing has been received. I just checked the article against Brian Greene's The Elegant Universe, and this article now has at least as thorough and well grounded an account as that well-received book does. So I will remove those tags if nobody objects in the next week or so. P0M (talk) 03:10, 9 December 2007 (UTC)

[edit] A simple yes/no question

The more I read about this the more vague it becomes. I understand almost everything, but I have a simple question, that I would like to be answered only by someone who is entirely sure of the answer.

Will there still be an interference pattern when
a) There are detectors, but the data is "destroyed" before the electron arrives.
b) The same, but the data is destroyed after all electrons arrived, but before anyone/anything has looked at the wall.

I think this is a reasonable question, as it is physically testable and objective. The article states that (a) is true, but several people in the discussion apparantly don't agree. And the article states nothing about (b), but I have seen some videos (example) who make that statement. I know this isn't the place to ask questions, but I've already done alot of research about this and I'm lost in an endless sea of contradiction. I would just like to know the answer without devoting my life to it. Certainly because it's such a simple and objective yes/no question.
Greetings, Apography (talk) 19:47, 9 December 2007 (UTC)

The whole point about science is that one should never be "sure" about any positive statement. My feeling, so far, is that nothing unexpected is being shown by the so-called erasure experiments described in Wikipedia articles. But that doesn't mean that somebody won't do an experiment that really is astounding. And there is one amazing claim that I've read about. I will be happier when I have found the reference to it again. In that case there is said to be a real physical interaction, electrons are knocked galley-west by strong light, interference disappears, but then if the light that has been dispersed by exchanges with electrons is gathered to a focus by a lens the interference pattern returns. The argument would appear to be that if light bounces off electrons that are on their way out of a double slit barrier, then the way the light disperses gives information about the electrons and which path each one has taken through the apparatus. However, if one takes a lens big enough to gather all the dispersed light and focuses it to a point -- equivalent to the point at the light's source -- then all the information about electron position and velocity is lost to the observer. At that point, since nobody knows what path the electrons took, the interference pattern will return. If the experiment can be reproduced then it seems to me that it will take a lot of explaining to explain it away. P0M (talk) 07:18, 10 December 2007 (UTC)
The experiment was described in the Scientific American article on "making your own quantum eraser." I wrote to one author about it but didn't get a responsive answer. I couldn't contact the other author. I tried contacting the magazine and that didn't work either. I suspect that it is not a real experiment but just somebody's idea of what might work. P0M (talk) 03:48, 13 April 2008 (UTC)

[edit] Double split strange question.

For a theoretical mind experiment i wondered.

Imagine this famous double split experiment. But now the wall is just simply divided into 2 sides left and right. And one human made a peeking hole at each side. standing in the middle behind the wall his left eye looks into the left peeking hole. While his right Eye peeks into the right side peek hole.

Know we do this experiment, in quantum mechanics both paths are traveled, so which eye saw a photon hitting his eye?. There is only one outcome is into our current world. So by performing this did we erased one 'side'of our test person ? Or as long as we dont ask him what side he saw, will he be be in a superposition, both sides have seen it?.

No, the man is not in superposition. What happens is that a wave-function (or you could call it a "probability function") gets split between the left slit and the right slit. When that wave function gets propagated to the point where it intersects something like a wall or a human retina, the probability function's value at that point tells how likely it is that the wave-function will "collapse" at that point, and a photon will be detected. So if the wave-function gets split between the left slit and the right slit and then the two parts of the wave-function move forward and one part goes in the left eye and one part goes in the right eye, then there will be a certain probability of the wave-function collapsing in the left eye, a certain probility that the wave-function will get collapsed in the right eye, and a certain probability that it will get collapsed somewhere else.
Of course if you believe in multiple universes (and, to make me happy, if you can account for the matter and energy needed to make a new one every nanosecond or so) there is one universe in which the left eye sees a flash of light and another universe in which the right eye gets to see a flash of light. P0M (talk) 08:32, 9 March 2008 (UTC)

Perhaps it's intresting to wonder this, soon there will be QM computers. I wondered could they make our reality less real / or make us perform all kind of misstakes but our world only behaves like the world with the solution?.. strange quetsion i know but i wonder what the results are for the mind the human mental spirit in these cases. Somehow my idea of logic seams to be verry unlogic in QM world.. There must be something in between to gleu it all together.

Oh and btw a nice link here about this experiment drQuantum lnk —Preceding unsigned comment added by 82.217.143.153 (talk) 00:00, 14 December 2007 (UTC)

[edit] Japanese film?

"The results of the 1974 experiment were published and even made into a short film, but did not receive wide attention."

Reference? 202.135.231.49 (talk) 21:58, 20 December 2007 (UTC)

http://physicsworld.com/cws/article/indepth/9745 may have what you want. P0M (talk) 20:36, 19 January 2008 (UTC)

[edit] Size of apparatus?

On a first reading I got no clear impression of the sizes of the devices being discussed. Are these slits an inch apart? A millimeter wide? Or are we working at such a small scale that the slit might be on the order of several atoms wide?

I guess this matters (to me) because if you imagine an electron as a cloud rather than a billiard ball, it makes sense that "part" of the cloud could pass through one slit and the rest of the cloud through the other. But only if the slits are at a near atomic scale. If the slits are (to take the other extreme) a foot apart, it would seem to be clear from the design of the equipment that the electron beam could be aimed at one slit or the other removing all doubt about which slit was used. Somewhere between these extremes, the truth must lie. 70.162.156.229 (talk) 14:35, 19 January 2008 (UTC)

Would it help to add an image of the apparatus used to provide the diffraction and interference patterns (red) in the article?
From outside post to outside post is approximately 1 inch. The brads used to make the slit sides are the thinnest, #18, i.e., about .04 inch in diameter
From outside post to outside post is approximately 1 inch. The brads used to make the slit sides are the thinnest, #18, i.e., about .04 inch in diameter
Greene suggests using something like a 35 mm film slide of a totally black image (unexposed film that has been developed would do it) and then scratching two parallel lines very close together. I found it difficult to get the lines scratched properly. The brads work with my carpenter's laser but not very well with a very weak laser I got for $10 from Staples. There are math formulas that will let you calculate the spread of the interference fringes if you know the frequency of the laser light and the separation of the slits, so all your questions can be answered if you want to do the calculations. (I think I have seen on-line simulations that let you move the slits, change the frequency, etc.)
This apparatus would not work with much longer wavelength radiation. Microwaves would even require wider slits I believe. Whatever "wavelength" is, it has real world consequences. P0M (talk) 21:46, 19 January 2008 (UTC)
Distance between slits = 1.2 mm, slits = .4 mm -- narrower slits would be better assuming an intense enough source of laser radiation (or a sensitive detector). P0M (talk) 21:58, 19 January 2008 (UTC)

I've found one of the simulations. The link goes to a "demo run" that moves too fast to be able to really absorb details like how far the slits are separated, but one can download the Mathematica "player" and the particular demo and run it under your own control. Since the math is reliable, you get to see what you would see with a real apparatus minus the defects that might be caused by rough edges of one kind or another. P0M (talk) 18:59, 10 May 2008 (UTC)

[edit] One photon (or electron) at a time?

I guess we can take it for granted that the experimenters have become clever enough that they can generate individual particles at will. But a brief discussion or reference to how this is done would be helpful. 70.162.156.229 (talk) 14:35, 19 January 2008 (UTC)

Good point. It is often assumed that one-photon experiments were not done until the high tech era arrived. Actually one-photon experiments were done very early on by the simple expedient of filtering more and more light out until they got to the point that they had to wait a minute or two before they would get any exposure at all to their photographic emulsions. The method had an element of statistics and inference involved, but seems to have been widely accepted as valid.Currently there are very high-tech ways of doing things with quantum physics, but the results are the same. A decade or two ago they did the experiment with electrons and could make a movie that showed the arrival of individual electrons as glowing spots on the equivalent of a television screen. And, actually, the occurrence of electron interference was occasionally noticed when people were first developing TV and other CRT tubes.
I probably can scare up a citation somewhere. P0M (talk) 18:39, 19 January 2008 (UTC)

[edit] Copenhagen and the present tense

``The Copenhagen interpretation is a consensus among some of the pioneers in the field of quantum mechanics." since pretty much every pioneer of Copenhagen (ie Bohr) are (is) dead perhaps this sentence should be put in the past tense. Particularly since anyone thinking about this subject nowadays wants to modify or refute Copenhagen in some way. It is a consensus only among popular science writers and physicists not directly involved in this sort of science. 137.222.230.13 (talk) 01:15, 22 January 2008 (UTC)

The sentence says that the pioneers formed a consensus. That consensus is called the "Copenhagen interpretation." That's what it was in the early 1900s and that's what it is today. It's like the Constitution is the law of the land -- even though all the founding fathers are long dead.
Also, where is the research to say whether the Copenhagen interpretation is or is not still widely accepted? All it really says is that it is pointless to argue over the supposed existence of things for which there can be no evidence. Better to be clear what we have evidence of, what we can predict, etc., and let it go at that. True, there are people who try to justify "hidden variables," but the trouble is that the variables are so well hidden that there is no evidence for them. P0M (talk) 08:00, 9 March 2008 (UTC)

[edit] The experimentation method is not clear

The double-slit experiment with electrons is famous now, but the details are missing. One could ask how to observe the photons ... The idea is this experiment is not fully described and because of that many third-party sides took it and tranformed it is some strange theories. Some of them are rather hillarious. I don't deny them, but as I said, it looks like noone knows exactly what was done. For me, this one is like many other hoaxes from internet about free energy, vacuum energy, etc. As I said, I don't deny them but concrete proofs must be offered, otherwise they are nothing. —Preceding unsigned comment added by 192.35.17.29 (talk) 14:11, 6 March 2008 (UTC)

The article has links to the research on electrons.
The photons are not "observed" in the sense that we bounce light off something and see it. The photons are "observed" in the same way that bullets are observed when they leave tunnels through bodies or walls.
The experiment is very well described, and anybody with a moderately powerful "pointer" laser can do it at home. The photos at the top of the article were ones that I made using the laser from a "laser level," a double slit made with the thinnest brads I could find, and a little plastic frame to hold them in.
It may well be that some people have tried to draw weird conclusions on the basis of what is weird enough to begin with, but that's on them, not on the scientists or even the humble experimenters who regularly reproduce the same phenomenon. P0M (talk) 08:09, 9 March 2008 (UTC)

[edit] Fringes

what are fringes? where is it defined / explained on this page? —Preceding unsigned comment added by 129.11.76.215 (talk) 13:25, 31 March 2008 (UTC)

Good point. I'll fix the problem. P0M (talk) 14:21, 31 March 2008 (UTC)

[edit] Something else to try

Would it be possible, instead of "watching" the slits, to accurately measure when the particle hit the wall. Then you can calculate the time the particle needed to get to the wall, and knowing its speed you can calculate the distance it travelled. You can then calculate which slit the particle went through.

This should create a "regular" pattern, but here we know only the way it went after it has hit. Kind of paradoxical to then still switch to an interference pattern.

I searched, but couldn't find any experiment that has tried this before. Maybe any of you do, and could include it in the article? Apography (talk) 18:19, 11 April 2008 (UTC)

I think that using a "macro" view of the world according to which there is "a particle" that goes by way of slit A but not by way of slit B, or vice-versa, is a problem. I know that people often speak that way. But we have to keep several things in mind: (A) it may be difficult to tell exactly when the laser or other device that emits a single photon actually fires. There is an element of inference and supposition. We say that we supply power to the laser at time t "and then" the laser emits a photon. Anything that we do that involves a physical process takes some time. But we would get into infinite regress if we tried to clock the time between when the laser was turned on and when the laser emitted a photon. (B) "Something" has to go through slit A and "something" has to go through slit B. It makes a quantifiable difference how far the slits are apart, and, moreover, anything done to vary the length of either the path through slit A or slit B will affect the interference fringes produced -- or even prevent them from forming. So what appears to happen is that the wave-function is split by the barrier between the two slits, and the two parts of it arrive at the detection screen where they will interfere because they are out of phase. They are out of phase because they have taken paths of different lengths to the saem point on the detection screen, and that means that they have taken different lengths of time to arrive at the screen. If what was going through the slits was a two-dimensional "wave front," then interference could not happen because (as your question implies) the wave front from slit A would reach the screen in the center directly across from that slit and then spread out from there. It would reach point x (let's say to its left) a very short time later. But the wave front from slit B (to the right of slit A) would reach the screen first at a point directly across from that slit at the same time the wave front from slit A reached its opposition point. Yet it would take longer for the wave front from slit B to reach point x because it is farther away. So the wave front from slit A would be gone by the time the wave front from slit B got there, and nothing could happen.
So the wave-function that goes through each slit does not have a 0 extension in the third and fourth dimensions. (See http://www.falstad.com/qm1d/ for some simulations that may help with this idea.) What must be happening is that wherever a photon is detected on the screen the tail end of one segment of the original wave-function is meeting up with the leading end of the other segment of the original wave-function (that came via the other slit). Their probabilities added up to a positive value and the "god of probability" decided to trigger wave-function collapse at that point on that run of the experiment.
Now what that means is that the time it takes for the intersecting portion of the wave-function segment that comes via slit A is the time it takes for the intersecting portion of the wave-function segment that comes via slit B. So if we arbitrarily assume that the "photon" "traveled" by slit A, it will take time period t'-t to get there, but if we assume that "it" "traveled" via slit B, it will still take the time period t'-t to get there.
One other consideration is that the actual path length differences are extremely small because the separation between the two slits is necessarily very small. (Perhaps the most difficult thing about making the Young apparatus is getting the divider between the two slits narrow enough to permit a good interference pattern to form. There is an article in Scientific American on how to make your own quantum eraser and their instructions for making the double-slit part of the apparatus is to straighten out a staple from a desk stapler and put that in the center of the laser beam. If you calculate the difference in path length L between a straight shot to the detection screen from the left side of the staple, and a diagonal shot length R to the same point from the right side of the slit, and if the staple is .01 mm thick, you can see that R will be the square root of the sum of L squared and .01 squared. That's going to be a very tiny difference in lengths, and the speed of light is so very fast that the time differences between travel via the two paths is going to be extremely small. In order to get a credible result, your experimental error would have to be still smaller than that extremely small number.)
One experiment that has been done involves interference effects that result from a time gap between two parts of a wave-function. It would be hard to explain without diagrams. I'll see whether I can scare up a citation to that experiment. It is relevant to your question, and it shows that if you can pinpoint the time when the active part of a wave-function arrives at a detection screen then it can't interfere with itself and no interference effects will be detected. But if the experimental apparatus is adjusted so that two high-probability segments of the same wave-function could both be the ones that "delivered" a photon, then interference effects will be detected. In the case of the regular Young experiment, the spacial arrangements are varied so as to permit interference between different segments of the same wave-function, and in the new experiment the temporal arrangements are varied so as to permit interference between different segments of the wave-function(s) used in that experiment.
For the above reasons I suspect that physicists have already considered the "time of arrival" measurements and what can be learned from them. P0M (talk) 18:02, 12 April 2008 (UTC)
Take a look at this:http://physicsworld.com/cws/article/news/21623 P0M (talk) 03:41, 13 April 2008 (UTC)
Oh, thanks for taking the time to write such a long reply. It takes some time to understand your reasons and arguments, because they're quite advanced. At least for me ;) Apography (talk) 21:23, 13 April 2008 (UTC)

Image:Attosecond Double-Slit Experiment.svg

I'm not sure that I fully understand this experiment. The articles I have found are either too summary or they jump in with lots of things that are implicit. So take the diagram I've made above with a grain of salt. The purple star represents a laser that zaps a container of argon gas (the green circle). The laser can be controlled in such a way that the whole dose of energy needed to kick off an electron can either occur in two closely spaced equal doses, or can occur all at once. The electron that results could arguably be associated with the first half of the divided pulses, or the second half. But if the dose is concentrated in one shot there is no ambiguity. The electrons that result from two half-doses arrive at the left screen, and the electrons that result from a single full dose arrive at the right screen. The energies measured on the right screen all clump together in the middle. The energies measured on the left screen come in "fringes" of different intensities because of interference between "the two different paths" (as most people seen to express it) or between the part of the wave-function that comes associated with the first half-pulse and the part of the wave-function that is associated with the second half-pulse. If it were possible to tell "by which path the electron arrived" then there could not be interference. (That's what is happening on the right screen, anyway). If the wave function(s) that arrive at the left screen were separated in time then they presumably could not interfere with each other. So unless time is irrelevant there have to be two components there at the same time. If they are sort of like spheres then one sphere could be part of the way through the detector screen and the other sphere could be just entering, and so there would be an overlap. For some brief time they would share the detection screen, they would interfere, and an electron would be detected wherever their wave-function collapsed.

If you look at the quantum eraser experiments, they all have some way of doing something to whatever goes through one path to make it "distinguishable" from whatever goes through the other path. Sometimes (as in the Scientific American "Make your own quantum eraser" experiment) they cross the polarities of the two paths. Sometimes they change the phase relationships between the two paths. And "erasing" the distinctions always amounts to doing something to fix the polarities, fix the phases, or fix whatever else they have done to "label" the path followers. The simplest way is just to direct the two paths into skewed paths so that no matter how far they went the two paths would never intersect. The "antidote" is to redirect the two paths so that they will intersect. And that means getting them to the detector at the same time and just slightly out of alignment with each other so that the maxima do not all coincide, or the maxima do not all coincide with the minima. The Young experiment fortuitously did all the arrangements more-or-less automatically. (Sometimes lenses are used to make things line up better.)

If it could be demonstrated that the leading half of whatever comes out of the argon gas gets to the detection screen and either manifests as an electron or goes "on" before the trailing half ever gets there, then that would really throw a wrench into our ideas of time sequence, action at a distance, causation, "locality," etc., and things like Cramer's Retrocausal quantum eraser experiment would have some competition for weirdness. Since nobody has been talking about the weirdness of interference fringes due to events that happen at different times, I think it is probably safe to say that everybody believes that whatever comes by way of whichever path has to be at the same place at the same time to interfere.

I hope the above diagram is reasonably able to communicate the idea of the experiment. I labeled it a "schematic" in the SVG image, but for some reason Wikipedia's mechanism for turning SVG images into PNG images "on the fly" seems unable to do the vertical script. Anyway, don't take it without a gram of salt. P0M (talk) 05:08, 14 April 2008 (UTC)