Photon Experiment: What Will I See on Wall B?

[student85]

Ok I have no deep knowledge in QM but I have this thought experiment... which I believe is a quite known one... I don't know. But well here goes: Imagine I have a photon gun that can fire one photon at a time. I put this photon gun about half a meter away from a wall(wall A) with two small holes on it. The photon gun is pointing to the place between the two holes (which are quite close together). Now, the photon gun is not very accurate so it can shoot the photon directly toward a hole or even farther away from the middle of the two holes. At the other side of wall A I place a second wall (wall B) to record where the photon hits. So, I shoot ONLY one photon and look at wall B. What do I see? A spot?
If something was not clear please let me know, I am SO interested in this. Thanks.

 

[DieCommie]

Yea, you see a spot. But that is not where the magic of the double slit experiment comes in. :)

 

[student85]

If I see a spot then the photon is behaving entirely as a particle. Where does the magic come in?

 

[hiyok]

The magic comes as you fire a number of photons. Just as you do the same with electrons. Actually, the magic is not that a photon is like a particle, but that you can not predict where it shall arrive.

 

[madmike159]

Well firse youngs experiment proved that light was a wave, because the light was able to cancel out (superposition).
I don't think I can explain it very well so look here. Any thing you don't understand I'm suer some one will explain.
http://en.wikipedia.org/wiki/Double-slit_experiment

 

[hiyok]

You may fire a number of photons one by one. The result shall be the same

 

[student85]

Thank you all for your responses. So if we fire say, a thousand photons we would find a superposition pattern right? Can this be explained in a classical way? Can we ignore QM? (I'm still kind of new to QM so I have to admit I'm a bit skeptical... I believe there are better ways to explain some phenomenon that QM does)

 

[madmike159]

Thank you all for your responses. So if we fire say, a thousand photons we would find a superposition pattern right? Can this be explained in a classical way? Can we ignore QM? (I'm still kind of new to QM so I have to admit I'm a bit skeptical... I believe there are better ways to explain some phenomenon that QM does)

No QM was used to explain things classical physics can't. Basicaly theroys go like this. Some thing is observed, a theory is put together, tested, confirmed/dismissed. If confirmed they can still be dismissed or altered when new technology causes new effects.

 

[hiyok]

Can this be explained in a classical way? Can we ignore QM? QUOTE]
I think what you mean by this is whether one could explain the interferrence using the wave solutions of Maxwell equations. Well, the answer may be either yes or no, for there are two ways of interpreting the wave solutions: the classical way and the QM way. In the former, Maxwell eqs. describe waves that carry energy and momentum, just like water waves. These waves are the outright entities of the theory. However, in the latter, these waves don't have any direct physical properties (such as energy) at all. They are construed as probability waves. The real objects are something else, namely, photons. However, if the number of photons is infinitely large, these two interpretations can be interchanged in the sense that predictions made according to them are the same. But in the case of only a few photons, the classical use of Maxwell eqs. can't do right. Why? Because interference pattern (which is the prediction of the classical Maxwell theory) could be established only when many many photons are participating. Otherwise, what ends up is no more than an irregular 'star map' .

 

[lightarrow]

Ok I have no deep knowledge in QM but I have this thought experiment... which I believe is a quite known one... I don't know. But well here goes: Imagine I have a photon gun that can fire one photon at a time. I put this photon gun about half a meter away from a wall(wall A) with two small holes on it. The photon gun is pointing to the place between the two holes (which are quite close together). Now, the photon gun is not very accurate so it can shoot the photon directly toward a hole or even farther away from the middle of the two holes. At the other side of wall A I place a second wall (wall B) to record where the photon hits. So, I shoot ONLY one photon and look at wall B. What do I see? A spot?
If something was not clear please let me know, I am SO interested in this. Thanks.

Yes, you see a spot, but here we must make an important precisation:
In those conditions you actually won't see any interference pattern, after having shot many photons (or electrons, it's the same). If you have a source of photons so precise that you can be sure you are illuminating a single hole when you point at it, and only the other hole when you point it, and you point the source alternatively to one or the other of the holes (even if you do it casually) then the resulting pattern will be the sum of the 2 singular pattern made by one only hole open, that is a sort of 2gaussians curves close to each other, with almost no interference bands at all (there would be because of diffraction but you can set the experiment to reduce them a lot, with respect to the case of real interference).

I know that you said:

"Now, the photon gun is not very accurate so it can shoot the photon directly toward a hole or even farther away from the middle of the two holes"

but in this case it would actually be too accurate.

To have interference you must be sure the source is not even that collimated, but that it illuminates both holes simultaneously!

Anyway, if that is what you already intended, then forgive me; I just wanted to clarify which is the exact experimental setting here.

 

[lightarrow]

Thank you all for your responses. So if we fire say, a thousand photons we would find a superposition pattern right? Can this be explained in a classical way? Can we ignore QM? (I'm still kind of new to QM so I have to admit I'm a bit skeptical... I believe there are better ways to explain some phenomenon that QM does)

Everything goes as follow: before detection on the screem, the photon or the electron behaves as a wave, because of the interference effect. When detected on the screen, since we see a single little spot, it behaves as a corpuscle.
From my personal point of view, everything would be explained wondefully if we could someway show that a wave can interact that way with a screen of detectors. That's the real mystery, in my personal view.

 

[Marty]

From my personal point of view, everything would be explained wondefully if we could someway show that a wave can interact that way with a screen of detectors. That's the real mystery, in my personal view.

I wonder if this is what really bothers people. Because in that case wouldn't the whole question of interference patterns be a red herring? Shine a very weak light source at a photographic plate, so the photons arrive one at a time. Is it the very appearance of dots, not matter how weak the light source, which is really the essential mystery here?

 

[lightarrow]

Lightarrow: From my personal point of view, everything would be explained wondefully if we could someway show that a wave can interact that way with a screen of detectors. That's the real mystery, in my personal view.

I wonder if this is what really bothers people. Because in that case wouldn't the whole question of interference patterns be a red herring? Shine a very weak light source at a photographic plate, so the photons arrive one at a time.

This is undemostrable. You can only say that you see single pointed flashes of light on the screen, you can't say it was because of a tiny corpuscle which has flown from the source and that have hit the screen (if this is what you intended). You can give this as interpretation, for example within the De Broglie-Bohm interpretation, but you can't prove it (at least, not within the current knowledge of physics).

Is it the very appearance of dots, not matter how weak the light source, which is really the essential mystery here?

In my opinion, yes.

This is how I personally imagine it: The electromagnetic wave is present simultaneously on all the screen, but its intensity is so low that, statistically, only one detector on the screen clicks in a specific interval of time. The details of this interaction between the em wave and the detector are still not totally understood, there probably is a non-linear interaction (see also "decoherence"). We should also remember that when we say that "x" energy has been sent from the source and exactly the same amount "x" of energy has arrived on the detector, this is only true on average; it's the average value of the energy which is conserved, in QM, not its instant value.

I still want to point out that all this it's just my opinion; there are tens of physicists on this forum which assume that all that I wrote is wrong, and I don't really know who's right.

 

[lightarrow]

I didn't mention in my previous post because I assumed it as trivial: of course the intensity of the wave on the screen is given by the result of interfering waves from the two slits, so the detectors on the screen where the interference is constructive have a proportional greater probability to "click" and it's for this reason that, after many "clicks" you start seeing the interference pattern on the sceen.

 

[Marty]

This is undemostrable. You can only say that you see single pointed flashes of light on the screen, you can't say it was because of a tiny corpuscle which has flown from the source and that have hit the screen (if this is what you intended).

No, that's not what I intended.

Is it the very appearance of dots, not matter how weak the light source, which is really the essential mystery here?

In my opinion, yes.

This is how I personally imagine it...(skip ahead)...when we say that "x" energy has been sent from the source and exactly the same amount "x" of energy has arrived on the detector, this is only true on average...

I talked about a photographic plate as my detector. Does anybody know exactly how much energy it takes to induce a transition in the silver halide molecule? Is it necessarily equal to the energy of a photon? If we don't know the answer to this, why do we say that the presence of a dot on the screen is an indication of an amount 'x' of energy being absorbed?

 

[lightarrow]

I talked about a photographic plate as my detector. Does anybody know exactly how much energy it takes to induce a transition in the silver halide molecule? Is it necessarily equal to the energy of a photon? If we don't know the answer to this, why do we say that the presence of a dot on the screen is an indication of an amount 'x' of energy being absorbed?

About a photographic plate I'm extremely doubtful that the efficiency could be 100%. You need at least a bunch of photons (10? 5?) to generate a single spot on the tiniest grain of light sensitive crystal.
Note that I didn't talk about a single AgBr molecule, because a single crystal is necessary to have an irreversible effect.
CCD devices should be more sensitive, however.

 

[Cthugha]

I wonder if this is what really bothers people. Because in that case wouldn't the whole question of interference patterns be a red herring? Shine a very weak light source at a photographic plate, so the photons arrive one at a time.

This is undemostrable. You can only say that you see single pointed flashes of light on the screen, you can't say it was because of a tiny corpuscle which has flown from the source and that have hit the screen (if this is what you intended). You can give this as interpretation, for example within the De Broglie-Bohm interpretation, but you can't prove it (at least, not within the current knowledge of physics).In my opinion, yes.

Well, one has to distinguish between low intensity coherent light and real single photons. When using coherent light the photon number is always Poisson distributed and you will never have single photons for sure, but on the other hand there are single photon sources, which do not show this problem.

So how would you explain photon antibunching effects in the emission of single photon sources like single quantum dots or single atoms with a purely wave-like theory?

 

[Marty]

That's not the question. When "the world at large" talks about the mysteries of quantum mechanics, they're not talking about photon antibunching. They talk about things like the double slit experiment. All I'm asking is where exactly in the double slit experiment do we find any great mysteries?

 

[Astronuc]

Does anybody know exactly how much energy it takes to induce a transition in the silver halide molecule? Is it necessarily equal to the energy of a photon?

Yes. If one has worked in a darkroom, one knows to use red light to see while working with film. Exposure of the print film is done with yellow or white light.

See - http://en.wikipedia.org/wiki/Silver_bromide#Photosensitivity

Photons have a spectrum of energies.

 

[Cthugha]

That's not the question. When "the world at large" talks about the mysteries of quantum mechanics, they're not talking about photon antibunching. They talk about things like the double slit experiment. All I'm asking is where exactly in the double slit experiment do we find any great mysteries?

Oh, I did not intend to answer your question beforehand. I was just commenting Lightarrow's answer.

However, regarding the great mysteries of the double slit, I would say, that it is one the easiest experimental settings, which introduce complementarity. The insight, that the inability to have which-way information and an interference pattern at the same time is fundamental and not only a problem of measurement, leads directly to some of the fundamental questions in QM.

 

[lightarrow]

Well, one has to distinguish between low intensity coherent light and real single photons. When using coherent light the photon number is always Poisson distributed and you will never have single photons for sure, but on the other hand there are single photon sources, which do not show this problem.

So how would you explain photon antibunching effects in the emission of single photon sources like single quantum dots or single atoms with a purely wave-like theory?

Ok, how do you establish that a single photon has been sent from the source, before having detected it on the screen?

 

[peter0302]

I would imagine you know that a single photon has been sent because you know how much energy you put into the machine - hv - to create the photon.

 

[lightarrow]

I would imagine you know that a single photon has been sent because you know how much energy you put into the machine - hv - to create the photon.

This is true on average; actually you cannot say: "I have lunched a single photon. Now let's wait. Ok, now, after 1 second, we have detected a single spot on the screen". The photon antibunching effect is established on the analysis of coincidence counting on two independent spatially separated detectors.

 

[Marty]

Oh, I did not intend to answer your question beforehand. I was just commenting Lightarrow's answer.

However, regarding the great mysteries of the double slit, I would say, that it is one the easiest experimental settings, which introduce complementarity. The insight, that the inability to have which-way information and an interference pattern at the same time is fundamental and not only a problem of measurement, leads directly to some of the fundamental questions in QM.

I don't think it was seen that way historically. For example, as far as I know, Einstein never invoked the double-slit experiment as a supporting argument for his particle theory of light.

The idea of which-way information destroying the interference pattern was popularized by Feynmann's famous thought experiment. But that was with electrons, not photons. That's a different experiment.

 

[Count Iblis]

In principle one should be able to detect the photon coherently. Instead of the photographic plate. One can imagine a screen on which there is an array consisting of trillions of nano-antennas. If the photon interacts with some antenna, the antenna goes into an excited state. These antennas can be considered as qubits which can transfer their state to a quantum computer via the CNOT gate.

When the quantum computer reads out the information from the nano-antennas, it can process the information and tell if both holes were open or if only one of the two holes were open (and which one). So, only a single photon is needed to detect the interference pattern.

 

[Cthugha]

Ok, how do you establish that a single photon has been sent from the source, before having detected it on the screen?

Why? Before doing a measurement, the state of the photon is usually ill defined anyway. One could do QND measurements on the photon number, but I do not think this is, what you are after.

Of course you cannot imagine photons as tiny balls flying through space, but this picture is not necessary to have single photons arrive somewhere one at a time. You just need a quantized em-field and a suitable photon source.

I don't think it was seen that way historically. For example, as far as I know, Einstein never invoked the double-slit experiment as a supporting argument for his particle theory of light.

The idea of which-way information destroying the interference pattern was popularized by Feynmann's famous thought experiment. But that was with electrons, not photons. That's a different experiment.

I disagree. In Bohr's View complementarity was at the very heart of QM and his discussions with Einstein about indeterminacy in QM also included the double slit. At the 5th conferences at Solvay Einstein even came along with his own double slit scenario, the recoiling double slit, which was intended to show, that the uncertainty relation was not unavoidable. I am afraid most of the important papers of this time are written in German, but if you look up the Bohr-Einstein debates on Wikipedia (which is usually not a good source in physics), you will at least get a impression of the historical context of the double slit experiment in QM.

 

[Marty]

I disagree. In Bohr's View complementarity was at the very heart of QM and his discussions with Einstein about indeterminacy in QM also included the double slit. At the 5th conferences at Solvay Einstein even came along with his own double slit scenario, the recoiling double slit, which was intended to show, that the uncertainty relation was not unavoidable. I am afraid most of the important papers of this time are written in German, but if you look up the Bohr-Einstein debates on Wikipedia (which is usually not a good source in physics), you will at least get a impression of the historical context of the double slit experiment in QM.

OK, you're right. I got the impression from reading the Feynman Lectures that it was his own arguments about how you couldn't measure the position without screwing up the momentum, but if we believe Wikipedia then he's really recapitulating the historical debate between Einstein and Bohr. Except Feynman does it in terms of electrons and Einstein argued on the basis of light.

 

[lightarrow]

Why? Before doing a measurement, the state of the photon is usually ill defined anyway. One could do QND measurements on the photon number, but I do not think this is, what you are after.

Of course you cannot imagine photons as tiny balls flying through space,

Ok, it's already good to me that we agree about it.

but this picture is not necessary to have single photons arrive somewhere one at a time. You just need a quantized em-field and a suitable photon source.

Here you should explain exactly what you mean with:

"single photons arrive somewhere"

especially the term "arrive".

What I'm trying to say is that it's not completely evident to me that the anti-bunching effect is due to the "arriving" of single corpuscles on the screen and not to a special way in which the detectors have been previously prepared to respond at the incident field. (Hope the term "incident" is appropriate, I mean the field which is present on the screen, generated from the source).

 

[Cthugha]

Ok, it's already good to me that we agree about it.Here you should explain exactly what you mean with:

"single photons arrive somewhere"

especially the term "arrive".

What I'm trying to say is that it's not completely evident to me that the anti-bunching effect is due to the "arriving" of single corpuscles on the screen and not to a special way in which the detectors have been previously prepared to respond at the incident field. (Hope the term "incident" is appropriate, I mean the field which is present on the screen, generated from the source).

Ok, let me at first state, that I consider a single photon to be a single excitation of the quantized em-field and that I consider the discreteness of energy to be the defining property of the term "single". So, considering usual measurements, which involve detection of single photons, arriving is just defined by the absorption of this discrete amount of energy. From this point of view the term "single" comes from the usual results in time resolved second order correlation measurements, which show, that the probability to detect two photons from such a single emitter simultaneously or with just a short time delay is suppressed far below the shot noise level. This suppression at least shows, that there is no positive definite probability distribution for the photon number, which I would expect, if there was somme hidden underlying statistical process of interaction between the field and the detectors.

Let me stress again, that I do not consider single photons to be heavily localized. Although one can define (with some approximations) a photon number operator of some definite volume, this leads to nontrivial results like a non-local relation between the photon probability distribution and the photon energy distribution function (and therefore also the detection probability) in polychromatic fields, which is for example demonstrated in chapter 12.11 of "Optical coherence and quantum optics" by Mandel and Wolf.

 

[lightarrow]

Ok, let me at first state, that I consider a single photon to be a single excitation of the quantized em-field and that I consider the discreteness of energy to be the defining property of the term "single". So, considering usual measurements, which involve detection of single photons, arriving is just defined by the absorption of this discrete amount of energy. From this point of view the term "single" comes from the usual results in time resolved second order correlation measurements, which show, that the probability to detect two photons from such a single emitter simultaneously or with just a short time delay is suppressed far below the shot noise level. This suppression at least shows, that there is no positive definite probability distribution for the photon number, which I would expect, if there was somme hidden underlying statistical process of interaction between the field and the detectors.

Let me stress again, that I do not consider single photons to be heavily localized. Although one can define (with some approximations) a photon number operator of some definite volume, this leads to nontrivial results like a non-local relation between the photon probability distribution and the photon energy distribution function (and therefore also the detection probability) in polychromatic fields, which is for example demonstrated in chapter 12.11 of "Optical coherence and quantum optics" by Mandel and Wolf.

Ok, you have explained it quite well, it's more clear to me now. Thank you very much.
lightarrow.