Talk:Double-slit experiment/Archive 3

Archive 3

Special Request
This is a request for YOU: Please read #29 "Questions about the double slit experiement", and, if you are able to do so, provide a commentary. God bless you. Honeychurch 09:38, 7 November 2007 (UTC)

See above. Aquinas said abut God that it was impossible to use finite human cogitation to form accurate ideas of the infinite. He was aware of human limitations (and he was one of the least limited of humans). It turns out that finite human conceptual processes are not even very adequate for dealing with the very small (but still of some finite size) and the very large (but still in this finite Universe). So one works with models, or what somebody called "useful fictions" much as Aquinas said we get at some idea of God by denying of him all human imperfections -- which eliminates an awful lot. P0M 08:30, 12 November 2007 (UTC)

What does this mean
"However these experiments performed with detectors are dubious because we do not find, know design and working of these detectors." What does the italicized portion mean?

"There is no reliable source to verify that these detectors do not mess up path and properties of electrons and photons involved in experiments." Is this encyclopedic language? 212.179.210.204 04:12, 18 October 2007 (UTC)


 * I agree--these lines are vague and not encyclopedic-sounding. What the hell does it mean to say that we don't know the design and working of the detectors?  Obviously SOMEONE does, because someone built them. I'm assuming they weren't gifts given to us by little green men. --Lode Runner 23:36, 23 October 2007 (UTC)

Doubting Rothman
The article cites as fact Rothman's doubt that Young actually carried out his experiment, spending a considerable amount of time on that issue. It would be helpful to know who else doubts it besides those who accept Rothman on blind faith. Repeating isolated theories as fact runs the risk of filling Wikipedia up with half-truths shouted at full volume. One could then sympathize with the man who wanted to charge $10 per printed Wikipedia article, who when confronted with the statement in Wikipedia that it is free replied, "You believe everything you read in Wikipedia?"

If Rothman is right about the experiment not having been performed, we would have the interesting situation of someone who, in a culture totally wedded to the corpuscular theory of light, guesses its wavelength and then describes a gedankenexperiment that if carried out would demonstrate quantitatively the wave nature of light. To accomplish this without actually doing the slit experiment, he or an accomplice would either have had to carry out some other experiment yielding that information involving for example Newton's rings (in which case why would he not report that experiment instead?), or be visited by a time traveler from the future. --Vaughan Pratt 23:48, 22 July 2007 (UTC)


 * Yeah, I'm not really sure why it's such a big deal for a scientist to reuse explanatory diagrams. That seems to be the foundation of the entire suspicion.  &mdash; Laura Scudder &#9742; 14:04, 23 July 2007 (UTC)
 * It seems that this problem was disposed of long ago. P0M (talk) 07:29, 23 November 2007 (UTC)

Improper use of copenhagen interpretation?
I'm surprised that no one has mentioned this earlier, but the following statement in the article I would claim is not true:

''The Copenhagen interpretation posits the existence of probability waves which describe the likelihood of finding the particle at a given location. Until the particle is detected at any location along this probability wave, it effectively exists at every point.''

Given that the Copenhagen interpretation is the primary explanation given in most physics courses, I have never encountered any explanation regarding the existence of a particle at every point due to the wave-function changing in time. Based on the copenhagen interpretation, we could claim that the probability distribution changes in time giving us the knowledge of where the particle is most likely at, but I do not think we can claim that the particle exists at all points at once... this would contradict the idea of the probability distribution. Any thoughts on this? —The preceding unsigned comment was added by 71.117.4.108 (talk) 22:50, 16 April 2007 (UTC).


 * I think I've fixed this problem. Have a look. P0M (talk) 07:27, 23 November 2007 (UTC)

fractal spacetime
What is the scientific consensus on El Naschie's theory of fractal spacetime? El Naschie says the theory explains the results of the double-slit experiment. --70.130.45.233 09:38, 21 September 2007 (UTC)

Electron Motion sensors to detect 'which-way' info.
Single electron double slit experiment produce interference pattern even though radio waves are present in room. But it is said that it is not possible to detect through which slit electron entered. Why don't they fix simple radiowave motion detector in one slit? The electron will break radiowave signal triggering alarm and we will know through which slit electron entered.

We can fix separate motion sensors in both slits and we can arrange second alarm with different sound. —Preceding unsigned comment added by 6E656F (talk • contribs) 10:12, 6 October 2007 (UTC)


 * I don't think there is any question that electrons could be detected. Whether there is a detector at one slit or at both slits does not really matter. It's a spooky way of talking about things, perhaps, but if something exists that leaves a sign of a particle having passed one way or the other, then interference will not occur. It's not that physicists have contrived some odd way that they think the universe ought to work, but that the universe works in certain ways and those workings drive the formulations that sound so improbable that people try to think of ways around them. If people could detect the presence of an electron in such a way as to indicate that it had come out of one slit or the other and interference still occurred, then that result would be big news. Your idea sounds like it would be easy to put into practice in the real world. Probably people have already done it. The experiments are probably on record somewhere, but the news value would be something like, "Female collie gives birth to five pups." It would be a small town's second newspaper that would bother to publish such news.  P0M (talk) 07:20, 23 November 2007 (UTC)

A rumored variation of this experiment
The last chapter of the Scott Adams book The Dilbert Future mentions (page 231) a variation of the double slit experiment that goes something like this:

Detectors are put in place, with the familiar result. Then, the detectors are programmed to automatically delete the recorded data the moment after it's recorded, so that no one gets a chance to see it--the post-event act of data deletion supposedly altered the pattern the photons left. The self-purging detectors caused the photons to leave the same pattern as if there were no detectors at all.

I remain sceptical, because Adams freely admitted that he was explaining it all from memory and had probably gotten some things wrong, but if anyone can confirm that this variation of the experiment has actually been performed I think it definitely merits mention. If the experiment took place like Adams described, I think it would without a doubt support the (currently pretty shaky and pseudoscientific) consciousness causes collapse concept. If, on the other hand, Adams was mistaken and such an experiment never took place, then we can mention that in the book's article. --Lode Runner 09:47, 23 October 2007 (UTC)

Update: The video here http://www.youtube.com/watch?v=_OWQildwjKQ describes the experiment as Adams described it. I do not know whom the speaker is. --Lode Runner 10:43, 23 October 2007 (UTC)


 * creepy —Preceding unsigned comment added by 68.106.248.211 (talk) 10:31, 1 November 2007 (UTC)


 * This is "quantum eraser" stuff. It is very interesting. It reminds me of the experiments that show that when a time frame in chosen in which there are two maxima and one minima, and an interference effect can occur wheneven the maxima is there -- but you can't tell whether it is the earlier or the later maximum that is involved in the real-world phenomenon under observation. It's hard for me to digest this stuff because I almost have to write out a script with notes to myself about what we measure and what we know in all aspects of the experiment and at all times in one run of the experiment.  It is very easy for me to let in some assumption from the macro world.  I'm looking for clear accounts, accounts that will help me avoid stupid mistakes. Of course if anyone wants to write such an clear account and supply citations then I won't need to do the work, and there is less chance that I will make a mistake. I'm starting to wish for half-silvered mirrors, etc. P0M (talk) 07:44, 23 November 2007 (UTC)

A note on the templates
Just a quick note as to why I've tagged this article for cleanup and reference improvement. Large sections of this article are written less like an encyclopedic entry and more like a textbook--inline questions, enthusiastic descriptions of the experiments, and so on. Tidying up these sections should take care of the style issues.

The references tag should be pretty self-explanatory; six references in an article (with three whole sections having no references at all) of this importance is simply not enough. Given the relatively wide appeal of this experiment, the number of times it has been repeated, and the speculation it invites, I would imagine that a veritable cornucopia of references could be found.

I'll do my part once I get a shot, but I just wanted to put down the specific issues I saw to help ensure these improvements get made. --jonny-mt(t)(c) 09:21, 31 October 2007 (UTC)


 * I've asked this editor to give indications of what things still seem to require citations. So far nothing has been done. Is there still any reason to post a wolf ticket at the top of the article? P0M (talk) 01:31, 21 November 2007 (UTC)

Beginning to respond to the templates and critique
I have just wiped out the nonsense about "a single band" being produced by one slit. Strangely, even Sears made the same mistake in his Optics although his own photograph show otherwise. Anyway, wrong is wrong and lots of places have the right information. I will tuck in some of the references I have found when I get a little more time. I made my own photos of both the single and double slit results from the same basic apparatus, which gives the reader what one really will see when the experiment is done. The photos in Sears are better, but they are copyrighted. My equipment cost a couple dollars. I suspect that his equipment cost a bit more. P0M 04:15, 6 November 2007 (UTC)

I couldn't stop myself. I found some very useful diagrams on the Commons and was able to write out the basics of the experiment in a way that I hope will be correct but also clear to the average well-informed reader.

The rest of the article needs some editing both with regard to fact (one photon experiments were done very early, with the math to prove that the photons had to be arriving one at a time at least most of the time) and with regard to writing suitable for the non-expert. Frankly, there are things that ought not to be incomprehensible that I could not understand on a rapid reading. Since Sears and Heisenberg don't provoke these confusions and uncertainties in my mind I suspect that it is the writing that is at fault. (Sears is a great textbook writer, by the way.) P0M 05:35, 6 November 2007 (UTC)

problematical content
The paragraph that starts "A remarkable result follows from a variation of the double-slit experiment" Are there any citations to back this idea up? If one photon can go through two slits at the same time, what happens if two photons arrive at those two slits at the same time? At the same time? How does anyone know which instances of transit through the two slits occur at exactly the same time? And if one photon goes through two slits anyway, isn't it going to interact with itself regardless of what any other photon does? What would be predicted if two photons were entangled and one or both of them went through a detector that absorbed and re-emitted it? How else would one get two photons known to be temporally matched? P0M 06:35, 6 November 2007 (UTC)

If nobody can provide citations I will cut this part out. P0M (talk) 00:47, 21 November 2007 (UTC)

A sentence that does not make sense.
The article says: "In case two pinholes are used instead of slits, as in the original Young's experiment, hyperbolic fringes are observed. This is because the difference in paths travelled by the light from the two sources is a constant for a fringe which is the property of a hyperbola." Does a hyperpobola have a property? Maybe, although I do not believe that mathematicians ordinarily speak of a definition that corresponds to the word "hyperbole" and maybe the lines that can be drawn on the basis of such a definition. And is the property supposed to be a fringe or a constant? The syntax is totally bonkers. Can somebody rewrite it so that it is a statement that is true or false? That would be a start. P0M 06:16, 12 November 2007 (UTC)

I will delete the sentence if nobody can fix it. -- P0M (talk) 21:41, 16 November 2007 (UTC)

I have deleted the sentence. P0M (talk) 00:46, 21 November 2007 (UTC)

present state of the article
Is anybody else monitoring this article? The request for more citations, etc., was very vague. Perhaps someone would like citations regarding specific issues. Please provide such notices.

Right now the article probably is suffering from its own lack of coherence. It is difficult to change things one by one without creating problems of repetition or even inconsistency. P0M (talk) 06:25, 23 November 2007 (UTC)

The observer role
The article states: "The remarkable consequence discovered by this experiment is that anything that one does to try to locate a photon between the emitter and the detection screen will change the results of the experiment in a way that everyday experience would not lead one to expect. If, for instance, any device is used in any that can determine whether a particle has passed through one slit or the other, the interference pattern formerly produced will then disappear."

I understand that this video: http://www.youtube.com/watch?v=DfPeprQ7oGc ... illustrates this phenomenon.

... but the article doesn't explain why that could be and what it means. (Whereas the full length version of the above video draws a meta-physical conclusion on the subject).


 * It's a little complicated, actually. What you are asking is sort of like asking to prove a negative. If the article has any defect as it is currently written, it is that is makes it sound as if you can absolutely prove a positive general proposition (e.g., "All swans are white."). What people know so far is that lots of things were done to try to detect whether a photon passed by way of one slit or the other, and that when those methods were tried the interference pattern disappeared. So let's say that 100 ways of detecting the passage of a photon have been tried and I say, "I guess that shows you!" Then somebody comes along with formula 101 and detects the photons somehow without ruining the interference pattern. Bingo.


 * One of the "metaphysical" questions is whether it is "human observation," or just doing something that will leave a mark that somebody a million years from now (or at any point in time) could choose to look at. Heisenberg thought that the essential thing was that when the dog went through the dog door the dog door slapped the dog's tail. Whether anybody heard or recorded the sound of the slap was irrelevant. Even a deaf dog would still be impelled to move his tail forward a little faster. (Sorry for the lame analogy.)


 * There are several attempts in motion to detect the slit that different photons pass through and then "erase" that knowledge so that the interference pattern will still appear. One such experiment is demonstrated by the following diagram:




 * Instead of dividing a stream of photons into two groups by putting a double slit in the middle of things and letting half the photons go through each slit, this experiment uses a mirror the is designed to reflect half of the incident and allow half to pass through. Then the two beams of light are led off in different directions, made to turn corners by fully reflective first surface mirrors, and finally their beams cross (which doesn't do anything to either) and the beams each end up throwing a single spot on the approriate detection screen. So from the time the light hits the first beam splitter (half reflective mirror), two groups of photons are going through entirely different paths. And when they get to the end of the free flight trajectories and hit a wall, they do just what anybody would expect them to do. They form a spot of light on the wall just as the original light would have formed a spot on the wall if there hadn't been any mirrors in its way. However, if a second beam splitter is put at the other end where the two beams split, the two beams get mixed and it is no longer possible to tell which beam a given photon originally belonged to. At this point it becomes possible to get an interference pattern.


 * To me what this experimental result shows it that to get an interference pattern you have to give photons two (or more) paths to follow and then combine those paths in a way that will let them interfere. But once you let them combine you inevitably lose track of which photon is which. Feynman argues that it is wrong to talk about photon one going by the left path and photon two going by the right path; instead, each photon goes by both paths -- and if you don't grok that it's your problem. He's not sympathetic because he doesn't understand it either.


 * There is another experiment that is probably lots harder to set up, but it is much more threatening to our everyday ideas. It even seems to upset our ideas of time flow and causality. If the double-slit experiment is done with electrons then a light shining on the other side of the slit will bounce off individual electrons and in so doing will be diffused. Because the light interacts with the electrons it changes them individually enough that they do not interfere with themselves and the expected interference pattern does not emerge. However, if a lens is used to collect the bounced light and direct it into a camera, the interference pattern returns. The most disturbing part of the experiment is that the optical path to the camera can be extended far enough to make the electrons reach the detection screen before the light reaches the camera. Why should this experiment work? Some people think that something goes backwards in time "after" the light reaches the camera. The camera appears to function as some kind of terminal point from which a time-inverse reflection propagages.  I'm just going on memory, but you can look it up. (It's somewhere in the links to this article.)


 * The salient thing seems more and more to be that if one could have knowledge of something then it will affect the experiment, and if possibility of gaining knowledge of that thing is destroyed then the interference patterns come back. It sounds like some variant of solipsism, but there are ideas of multiple universes that don't make me any happier. (Conservation of matter and energy, well yes, but another entire universe appears every time a geiger counter clicks.)

If possible, I'd like someone qualified to complete this article and maybe give us plebeians further information on the generally accepted meaning of these results. Thanks! :D


 * The vanilla cases are pretty clear. To get interference patterns you have to put a divider of some kind in front of an emitter of light such as a laser and then you have to take whatever it is you have divided (and it is very much in question or even unknowable what you have actually divided) and let it come back together again in such a way that the two streams are not in phase. Then you will see an interference pattern. The only way you can have an interference pattern is is you have some kind of "peaks and troughs" deal so that the peaks can sometimes fill in the troughs, and sometimes peaks can stand on top of other peaks.


 * At least some of the ways to mess up the interference phenomenon are known. Anything that you put in one path or the other that interacts physically with what is passing by that point will let you say that "a photon was detected here" will destroy the interference pattern. In this case it is pretty clear what has happened because the vanilla case of a photon betraying its presence is that it is absorbed by an atom. It then is no more, but an electron in that atom rises to a higher level. Day-glo paint would be an example. Day-glo works because infra-red light is absorbed, and energy of a higher frequency is subsequently emitted. I am not sure, but it looks like what happens is that more than one infra-red photons are absorbed, boosting an electron up some for each infrared photon that is absorbed and then the electron falls down near where it started, but in one jump.  Anyway, you could perhaps put a piece of doped glass in one or both slits and then observe the emission of the visible frequency light. But the photon that leaves that point was not the photon that arrived there. So it can't interfere with "its other half" because even if its other half went through a similar change it wasn't the same change.


 * As I understand it, the next attempt to mark photons somehow is to do something that will identify them without changing significant features of their identity that would prevent them from interfering with themselves. One idea is just to rotate photons with a polarizer, so you know if the photon went through slit A it was rotated and if it went through slit B it was not rotated. Then later on you could put in another polarizer and re-rotate the ones that were earlier changed. To me this seems rather lame because the next thing you are going to have to do is to merge the restored paths, and at that point there will be no "knowledge" of which one is which.


 * It is easy to reduce this picture to very simple terms, which makes it seem trivial.




 * If you make a wall that extends the area between the two slits so that the two resulting beams can never get together, then you can associate a photon detected on the screen with the slit on that side of the wall. You will know (or know with very high probabilistic certainty) that it didn't go through the other slit and then tunnel through the median wall. But you will not get interference. Tear down the wall and you will get interference, but you will be dealing with a mixed path situation again.P0M (talk) 05:01, 9 December 2007 (UTC)

EDIT: in particular the article does not mention if the measuring equipment can actually alter the movement of the photon, which would explain why the experiment results are different.


 * If you "measure" something you have to interact with it, and if you interact with it you change it. It's not clear that you alter the "photon's" "movement" because between the time you flick the switch on the laser and the time you see a flicker of light on the detection screen you have only inference and guesswork about what is going on. You know there is a time factor involved, you know there is a frequency factor involved, and you know that what happens depends on the widths of the slits and on their separation. But those are the only things you can actually observe. The fact is that lots of things that can be done to measure photons actually end up destroying photons. It may be that some things can be done that change part of the psi function of a photon. Maybe an example of that would be polarizing the photon. The hope of experimenters appears to be that they can identify, somehow, which slit a photon has gone through and do so without destroying its ability to interfere "with itself."


 * I think I agree with Feynman that it is presumptuous to say that there has to be "a particle" that either "goes through the left slit" or "goes through the right slit." That's the way baseballs and bullets work in the macro world, and we are more comfortable with thinking in such familiar terms. On the other hand I would be uncomfortable about thinking about a baseball that went through one window in my living room and that had to be accompanied by a "ghost" baseball that went through the other window. But at this point all I have is a point of view. P0M (talk) 05:01, 9 December 2007 (UTC)

xxxx

"Everything propogates like a wave, and exhanges energy like a particle." Tipler, ISBN: 0-87901-432-6, but pretty much anywhere you look. So when you obseve the position of the electron, you collapse the wavefunction describing the probability of it's position; i.e., because you have measured it to be in one physical location, the probability of it being anywhere else becomes zero.


 * Personally I agree with this point of view. But encyclopedia articles are not supposed to take sides. P0M (talk) 05:01, 9 December 2007 (UTC)

To detect something you must interact with it, and to interact you must exchange energy; therefore you collapse the wavefunction, the position of the electron is determined, and the interference pattern is removed.


 * Again, I agree, but it appears that physicists are trying to find ways of "marking" particles without changing irrevocably the aspects of them that would permit them to interfere. P0M (talk) 05:01, 9 December 2007 (UTC)

Now, the above is a little controversial. Mainly because I've described the collapse as a real change in the system. The majority would agree, but not all, and there is a discussion around 'real'. Secondly, some believe that it is observation by conscious beings that is important rather than interaction. This is a view that is in NS this month, as is also in the article in this commment:


 * "NS"? P0M (talk) 05:01, 9 December 2007 (UTC)

"Even less in line with the expectations of human scale interactions with nature, if the information about which slit a given particle came through is "erased" before a photon has time to interact with the detector screen interference will be restored."

The change to include this is recent, as far as I can tell. If this is true, then at the very least it needs references. This is what has prompted me to make this, my first contribution. Apologies for any Wikipetiquette broken. Davini994 21:12, 30 November 2007 (UTC)


 * No problem with etiquette at all. Good comments. I've been wondering how to work in the newer stuff about erasure. If you will take a look at the Wheeler's delayed choice experiment you will see an article that was apparently conceived and then abandoned by someone. I've been trying to get it straightened out, but I haven't had time to get Wheeler's book yet.  What I found worthy of note about that article was how the mere assertion that somehow looking at two slits of the double-slit apparatus with "telescopes" directed at the slits could prevent the formation of an interference pattern could apparently be totally convincing to some people -- all without a shred of experimental evidence and obviously without carefully following through the individual steps of the thought experiment.  Wheeler was an eminent physicist. I doubt that he could be a hazy thinker, but some of the assertions originally in the article were without any basis I could find. What he did say, from what I can make out from secondary sources, was quite deep and has to do with issues of the fundamental structure in space and time of the universe. Anyway, my point is that when writing an article on a question of fundamental importance like this one we need to be sure to hold ourselves to what the fathers of the field said and trace out the development of these ideas. Everything, ideally, should have a good clear citation to back it up.  P0M (talk) 05:01, 9 December 2007 (UTC)

'''People, this article is on edge of dumbest false speculations when it comes to "observer role" or "quantum erasing". Why not ask Markus Arndt or  Anton Zeilinger from Univercity of Vienna? They both are on the bleeding edge of quantum wave experiments, and their home pages and emails are well known by Google.'''


 * Please sign your postings.


 * As long as the article does not go over any "edges" but instead explores those edges, it will probably be o.k. I personally agree that Heisenberg was right in cautioning about any "metaphysical observer with sentient intelligence, blah, blah, blah."  I don't think, however, that the article does anything more than report on how various people who are important in the field have expressed themselves.


 * I am in doubt as to what "quantum erasing" would actually prove, anyway. The salient feature of these experiments, which everyone seems to risk overlooking by worrying about "which slit the photon really went through," is that something is important about the second unobstructed slit. I personally am deeply agnostic about whether a "particle" goes through either slit. The operational definition of a particle has to do with sticking some kind of detector up and seeing whether it takes a hit or not. Absent that, there is no particle.  But people with any sense will prefer to know what Feynman, Heisenberg, et al., have to say about it.


 * That being said, I still have to do a lot of study to get up to speed on the "erasures" business. Some of them involve entangled photons, measuring one entangled photon and drawing conclusions on that basis about the other one, etc.  It will be difficult to think carefully through all of these things, and I think the reader will be better served if we put the history of conclusions about the double-slit experiment up in this article and let links to other articles lead readers to the newer and perhaps more unsettled stuff. P0M (talk) 05:01, 9 December 2007 (UTC)


 * I just had a look at the article on quantum erasure, which is pretty poor. I also had a look at the external link, which was to a school project somewhere. That site had a quite clear description of the physical set-up. If you don't start out by believing in the power of the human will to determine the universe, the importance of having a sentient observer to make quantum probabilities come to rest in the real world (which conscious thought creates I guess), then it does not look so wonderful at all. It's rather like a magic trick.


 * By using entangled photons, it is possible to change the polarization of one stream of photons by polarizing the other stream. That's disturbing because of non-locality. But the rest of the setup in almost transparent.  The second stream of photons is divided and sent through two polarizing crystals. After that operation is performed, the two streams will not interfere because they are crosswise to each other. But now a second polarizing crystal is put into the apparatus. It is placed so as to make half of the incoming photons on the other path vibrate in the x direction ("horizontally") and half vibrate in the y direction ("vertically"). But doing that has the same effect on the entangled photons in the first path, so now rather than going either vertically or horizontally into the polarizers that turn vertical vibrations into one kind of circular polarization and horizontal vibrations into the other kind of circular polarization, the second stream goes in being half vertical and half horizontal, so the polarizing crystals on that path now each produce an equal number of circular "S" mode and circular "Z" mode polarized photons, so there are enough circular "S" mode coming through slit A to meet their "other halves" coming as circular "S" mode through slit B, and they interfere nicely. And there are enough circular "Z" mode coming through slit A to meet the appropriate circular "Z" mode ones coming through slit B, and they interfere nicely too.


 * The essential question in regard to interference and non-interference is whether photons (or whatever we really should call whatever "ghostly" thing it is that goes between both slits or through all possible paths (Feynman) are afterwards permitted to coincide where they can betray the fact that they are out of phase or whether they are not permitted to get together with themselves and are fenced apart one way or another.


 * It's probably actually more interesting that the "segments" of the original disturbance that is propagated through the apparatus seem to stick together. One of the major early figures in the study of this problem said that photons only interfere with themselves. (You get big problems, contradictions, if you assume they can interfere with others.) So the two "segments" found after going through two slits and encountering polarizers will coordinate their polarizations. That has to happen, else there could not be a good balance at the end of the erasure experiment when interference is reestablished. In the Wheeler experiment (the single and dual half-wave devices), if it were not this way then the two segments of one original disturbance propagating through the apparatus could end up each on a different one of the two detection screens at the end of the apparatus.  P0M (talk) 08:17, 9 December 2007 (UTC)

reference to figures and graphs
Is it possible to reference figures exactly rather than by "in the graph below", "The second drawing", "third drawing" or "the second photograph above)"? Of course, someone would have to label the figures as 1, 2, 3 first.  CowardX10 (talk) 00:32, 30 October 2008 (UTC)

Another Interpretation?
By way of background I have a PhD in quantum theory so I am not just a crank. I wonder if another possible interpretation of the experiment can be drawn by considering the mechanics of the interaction between the particle and the slits. That interaction is electromagnetic in nature, and is mediated by photons, which have a wavelike nature. Consider the case in which a single electron is fired towards a screen with two slits. There will be an electromagnetic interaction between the electron and the screen, mediated by the exchange of photons between the electron and the atoms that form the screen. The pattern of photons emited by a screen with two slits will be different from the pattern emited by a screen with a single slit, so in principle it is possible for an electron to be influenced by the existence of additional slits even if it passes only through one of them. Does this make sense? —Preceding unsigned comment added by 86.147.150.132 (talk) 22:51, 23 January 2009 (UTC)
 * If you are a physics Ph.D. then you may be qualified to design an experiment that would test your ideas.
 * One of the underlying epistemological questions lies in your assumption that it makes sense to ask whether an electron passes 'only through one of them." There is no evidence from which one might draw that conclusion.
 * Another problem would be to calculate the characteristics of the influence of the "extra" slit on the electron -- to supply an explanation for how the photons from slit A would influence something in the vicinity of slit B, and by this influence would create the appearance of an interference pattern consistent with waves having gone through both slits.
 * A major problem in inter-human communications is involved in all cases where there are "which path" questions. It is tempting to say that the photon, electron, or whatever is going by one path or the other. Much interest has been generated by the question whether it would be possible to detect which path was "actually" used. But if one tries to understand the experiment, explain the experiment, by saying that a wavicle goes through one way, then one has to speak of a ghostly non-wavicle that goes through the other path and interferes with the wavicle. What to call this traveler who wasn't there because the wavicle went the other way is a useful irritant because it reminds us that the whole conceptualization keeps getting pulled back to common language and the assumptions about objects in the macro world.
 * If one gives up the idea of there being a thing that goes from point to point in space and so leaves from a cathode and follows a straight-line path to a point on a detection screen, then the experiment looks something more like the following:
 * There is a cathode (or a laser) at one end of a big box and a detection screen at the other end, and the box is divided in the middle by a partition. If there are no holes or slits in the partition, wavicles are usually noticed on the partition or appear to have been reflected from the partition and get detected on one of the other five walls of that part of the apparatus.
 * If there is one hole or slit in the partition, markedly fewer wavicles are detected in the first compartment, and many wavicles are detected at the far detection screen. If bullets had been shot through a hole in a thin partition, the pattern formed would be almost entirely due to the configuration of the cone from the center of the gun barrel to the edges of the hole. But wavicles pass through a single slit and produce a diffraction pattern. There are simple models that can predict the diffraction pattern without any reference to quantum mechanical theory, and there are quantum mechanical explanations that clarify and improve the model and its predictions.
 * If there is an explanation for how atomic-scale interactions are mediated by photons that pertains to a single slit experiment, then the conditions described for a single slit must also apply for each slit in a double slit experiment.
 * If there are two holes or slits in the partition, then many wavicles are detected at the far detection screen but their pattern is not the pattern that would be explained by the analogy of a .22 rifle of low accuracy spraying a cone of bullets some of which go through each of two holes. Instead, the pattern of wavicles observed is consistent with a wave that is propagated through two slits. The atomic scale reactions that could influence an electron would have to explain or be consistent with the diffraction through each slit, and, if it is not assumed that the wavicle has a position in space as it transits the partition, any atomic scale reactions would have to be equally applicable equally to the "whatchamacallit" that goes through both slits (or whose arrival at the distant detection screen is influenced by the presence of the two slits). Looking at things this way, if one could account for an influence from slit B that affects interactions in the region of slit A,then there would equally be an influence from slit A that affects interactions in the region of slit B.
 * As far as model making goes, it seems preferable to keep any model as simple as possible while still explaining/predicting experimental results.
 * It is characteristic of all of the double-slit, double path, experiments that they are highly symmetrical. As soon as symmetry is broken, the interference pattern disappears.
 * So to get a model where an effect is produced in slit B and felt in the region of slit A, one will ideally show how this level of complexity is necessary to predict/explain something that is not already taken care of. P0M (talk) 09:58, 25 January 2009 (UTC)

Detection in the Double-slit Experiment
2/28/09

One of the central mysteries in this experiment is why the "mere" process of detecting which slit a single electron passes through collapses the wavefront of the particle. Some have suggested that the detector causes a polarization effect which prevents a wavefront from forming; others that the process of detection slows down the wavefront -- relative to the one emerging from the other slit -- enough to prevent the interference pattern on the screen.

Here is my question: Must not the detector itself be changed in some way by the photon, i.e., gain energy from it?Which, in turn, would lose energy? This should shift the photon's wavelength towards the red. Would not a tiny spectroscope built into the detector (maybe this is just not possible) reveal this red shift and explain the puzzle?

12.213.224.56 (talk) 21:37, 28 February 2009 (UTC)


 * You ask why the "mere" process of detecting which slit a single electron passes through collapses the wavefront of the particle. The answer is that there is no such collapse. The wavefunction does not describe an individual particle but it describes the whole interference pattern. Wave function collapse is a relic from the founding period of quantum mechanics. Most realistic measurements can be demonstrated to behave quite differently.WMdeMuynck (talk) 22:14, 28 February 2009 (UTC)


 * For WMdeMuynck:Is there an interference pattern before the wall with the double slits is reached?


 * Quantum mechanics does not predict any interference in front of the wall with the double slits (caused by these slits) unless there would be backscattering of particles from these slits.WMdeMuynck (talk) 20:43, 1 March 2009 (UTC)


 * For WMdeMuynck:If there is only one slit in the wall, is there in that case an interference pattern? P0M (talk) 00:16, 1 March 2009 (UTC)


 * If there is only one slit, there may be refraction, resulting in certain fringes. These might be explained by interference between different parts of the slit, considered as different (interfering) particle sources. Usually this is not referred to as interference, however. This is all analogous to what happens in optical phenomena.WMdeMuynck (talk) 20:43, 1 March 2009 (UTC)


 * 12.213.224.56, one of the problems with discussing attempts to determine which slit the electron goes through is that, to give an accurate picture, the measuring equipment (that determines which slit) itself must be described as a wave function.  This is beyond my physics ability.   That said, I think your suggestion is correct - the measuring equipment, to locate it, has to exchange energy with the electron, which (randomly) changes the phase of its wavefunction, which destroys the interference pattern.  Another way I've heard it described is by the uncertainty principle.  Any equipment that localizes the position of the electron in the axis of the slits (say the y axis) must, from the uncertainty principle, increase the uncertainty of it's y-momentum.  The momentum is described by a wavefunction just as the position is, and narrowing the y-position wave widens the y-momentum wave, in effect giving the electrons leaving the slits a wider spread of random transverse velocities.  This 'smears out' the diffraction pattern at the screen.  In terms of particles, you can say any measuring equipment that localizes the electrons scatters them randomly in the y direction, destroying the pattern, though of course the precise description is in terms of waves. -- Chetvorno TALK 08:56, 1 March 2009 (UTC)


 * It would really be helpful to have a list somewhere of all of the ways that people have used to detect "which slit was used."
 * It actually prejudges the whole thing to assert that a wavicle/disturbance propagates through one or the other slit. That's the macro world prejudice that we inevitably carry into the micro world.
 * If the experiment produces an interference pattern, that means that, whatever it did, the wavicle/disturbance got away clean when it went through the apparatus, and that there were two open slits (at a minimum). If something is done that succeeds in creating some kind of detectable change at or near one of the slits, that will localize the passage of the wavicle/disturbance to one slit or the other, and the result will be that there will be no interference pattern. A spectroscope measurement at one slit or the other would localize the wavicle, the disturbance, to that one slit.
 * But your question suggests another interesting experiment that could clarify what happens to a photon passing through a slit. You could use one spectroscope and one open slit, or you could use one spectroscope and position it where one of the bright bands should develop when two slits are involved. If the apparatus was generating one photon at a time you might have to send several photons through before getting one to go through the spectroscope. If you were using, say, an ordinary laser pointer, then you could put the spectroscope at the position of the brightest interference brand. Now measure the frequency of the photons that come through. Record that data and refine it in all ways possible. Then pull the screen with the slits in it and measure the frequency of light produced by the same photon generator. If going through a slit does anything to the photons that go through it, then the two frequencies should be different.
 * The only process I know of that could change the frequency of the self-same photon would be the kind of red shift that is a kind of Doppler effect. But the parts of the apparatus are not moving at all with respect to each other. They certainly are not moving with relation to each other at a significant fraction of c. So any observed frequency change would necessarily involve swapping one photon for another and losing some energy somewhere in the middle.
 * Frequency changes were noticed by people in Heisenberg's circle before he developed his quantum mechanics. They were trying to account for the dispersion of light by some dispersive medium. They learned that their math had to model a physical situation in which an incoming photon disappeared in the change of state of an electron in the dispersive medium, and then the electron returned to its equilibrium state. When it returned to its equilibrium state it could do so in a one-stage operation that would generate a photon of the same frequency as the incoming photon, or, it could undergo a multiple-stage series of changes according to the rules laid out in the Ritz combination principle, and emit a number of photons the sum of whose energies equaled the energy of the incoming photon. In such a case, the photon that leaves the laser is not the photon that reaches the detection screen. P0M (talk) 11:23, 1 March 2009 (UTC)

Back to the question from 12.213.224.56:

I am going to discuss the issue in terms of an apparatus in which photons enter one at a time.

The questioner had a real question, but perhaps it was not phrased properly. The questioner wanted to know what it is about passage through a slit with a detector that spoils interference. His own question suggests a couple of kinds of detectors, each of which would defeat interference in a different way.

To paraphrase his question, if a polarizer is inserted into one path and a polarizer of different orientation is inserted into the other path, would the frequency of photons emerging from these polarizers have a lowered frequency? And if an optical device is inserted into one path such that the phase of the photon emerging from that device is different from the phase it entered with, would the emerging photon have a different frequency than the photon that entered? For polarizers, the answer is no. Polarizers absorb some photons and heat up as a result, but they typically do not change the frequency of photons that pass through them. And the interference phenomenon is spoiled because of the crossed polarization anyway. The same thing applies to phase changes. The frequencies are not changed, but the phase relationships are changed. Changing the phase relationships is enough to prevent interference.

I suspect that by the words "collapses the wavefront of the particle" the questioner was referring to a kind of detection device that would create a physical record connected to a process that occurs in or just outside a slit. The implicit assumption is that any physical process by which a photon leaves a mark of its passage would demand that one photon be absorbed (collapsing a wavefunction in the process), an electron experience a change of state ("orbit" gets boosted up), and then (if a photon is to be detected at the far detection screen) a second photon must be emitted. A re-emission process in which part of the energy of the incident photon resulted in an enduring change in the detector (a chemical change, for instance), would mean that only a photon of lower energy could reach the remote detection screen. A spectroscope could detect these reduced frequency photons. The important point, however, is that they are new photons, and that they have a straight shot at the remote detection screen. See below:

What would be the situation if the detector at the slit site sometimes returns all of the energy to a new photon as the electron returns to its equilibrium state in a single transition? It is essential to describe carefully the event without the detector and the event with the detector is place:

In the first case, the event has a beginning when the laser (or whatever source of single photons is involved) emits a photon, and an ending when a photon arrives at the remote detection screen. In the middle there are two slits and all we really know about what happens between the beginning of the event and the end of the event is that it matters how many slits are involved, and their widths and separations as functions of the frequency of the photon in flight.

In the second case, one event has a beginning at A when a photon is emitted, and an ending at B when it is detected at one or the other slit. If a photon goes on from there and arrives at the far detection screen, that is a different photon and one that is the "actor" in a different event. That second photon is one that goes from point B to point C with no intervening slits at all. So the conditions needed for interference will not exist. P0M (talk) 05:57, 2 March 2009 (UTC)

It's better to describe the experiment without mixing in explanations like the concept of a "probability front"
A passage in the Overview section reads:

''Normally, when only one slit is open, the pattern on the screen is a diffraction pattern, a fairly narrow central band with dimmer bands parallel to it on each side. When both slits are open, the pattern displayed becomes very much more detailed and at least four times as wide. When two slits are open, probability wave fronts[7] emerge simultaneously from each slit and radiate in concentric circles. When the detector screen is reached, the sum of the two probability wave fronts at each point determines the probability that a photon will be observed at that point. The end result when many photons are directed at the screen is a series of bands or "fringes." The interference of probability wave fronts is shown in the graph below.''

Each person reading this article should have the opportunity to understand what the setup is, and what the raw observations are, in the experiment -- without having to simultaneously read about concepts like "probability fronts".

Explanation(s) and proposed conceptual frameworks for the results of the experiment should come later and separately, and should be labeled as such. But the experiment per se should be unadulterated by someone's explanation of it, no matter how widely accepted that explanation may be.Daqu (talk) 03:08, 1 March 2009 (UTC)


 * I agree. It is difficult to maintain scrutiny of an article such as this one. When I examined the top of the article I discovered on blatant mistake (one that conforms well to "what everybody know" I guess), and the inclusion of two paragraphs that are highly specialized. I deleted both. One was repeated at a more appropriate place midway down in the article. I'll try to find a new spot for the other one. However, it is the kind of thing that requires lots of explanation and probably ought to go into its own article.
 * Anyway, I have tried to clean some of the theory out and to keep things to a phenomenological description of what the experimenter sees. Explaining how people have tried to deal with the paradoxical facts can wait until later. P0M (talk) 07:21, 2 March 2009 (UTC)

misunderstanding of facts
in text---"Restriction to the two experiments in which either both slits are open or one slit is closed has given rise to the idea of wave-particle complementarity (to be distinguished from wave-particle duality) according to which a microscopic object (photon, electron, etc.)"

but infact the wave-particle complementarity page redirects to wave-particle duality.