How do we know we have a single-photon source?

[Johan0001]

In the double slit experiments , with 'Single photon emission'.
Taking into account the arguments above referring to one photon,, how do we establish that exactly one photon was emitted by the source.
In the light of the above explanation , is it not possible that they are may photons in a coherent state?

 

[fishin_kitten]

In the double slit experiments , with 'Single photon emission'.
Taking into account the arguments above referring to one photon,, how do we establish that exactly one photon was emitted by the source.
In the light of the above explanation , is it not possible that they are may photons in a coherent state?

All I can say is good question and I've been wondering the same.

 

[DrChinese]

There are a couple of answers, depending on what you are looking for. Check the following reference:

http://sciencedemonstrations.fas.ha...-demonstrations/files/single_photon_paper.pdf

In Section IV.A., they discuss whether there is a single photon or not.

"Although the statistics of the Poisson distribution for a classical light source allows for bunching of photons, the probability of measuring two or more bunched photons is very small for an extremely attenuated light source. For example, Pearson and Jackson18 claim that their reduced-intensity laser beam is comprised predominantly of single-photon states, even though g[2](0)=1. They report that 99.9% of their measurements result in single photons as opposed to coincident photons. This result suggests that reduced-intensity laser light is a good single-photon source, a fact that has been utilized in experiments demonstrating interference.22 Furthermore, if the wave packet is spread out over an array of detectors, such as our 668 496 pixel array, the probability that two photons from that single packet are measured in the same pixel is incredibly small. Thus, to a very good approximation, the single-photon contribution to the interference pattern is measured."

I don't believe they used that particular method in their setup, but it would be an option. Another technique would be to use entangled photon pairs, where one is sent to the double slit (after some manipulation) and the other is used to register a coincidence.

 

[f95toli]

The "gold standard" is to measure the second order correlation function g(2) (g is from Glauber who did the theory for this); sometimes combined with the g(1) as well. It is what is used both in research and for calibration in commercial applications.

This is standard measurement in just about every single quantum optics lab and is fairly straightforward (using off-the-shelf equipment) for optical photons; but can be quite tricky for e.g. very low frequencies.

 

[Johan0001]

My question was based on the famous , which slit did the photon go through problem.
For example if we there is a possibility that the ' single photon' emitted from the source, was possibly a number of coherent 'photons' then some could go through the left slit and others through the right and then recombine to be detected as one click on the detector screen,thus creating interference over time?
But this would imply that coherent photons are acting as one wave packet which we cannot distinguish as separate wave packets when emitted from the source, although they can split and recombine at some detection point?

 

[f95toli]

My question was based on the famous , which slit did the photon go through problem.
For example if we there is a possibility that the ' single photon' emitted from the source, was possibly a number of coherent 'photons' then some could go through the left slit and others through the right and then recombine to be detected as one click on the detector screen,thus creating interference over time?
But this would imply that coherent photons are acting as one wave packet which we cannot distinguish as separate wave packets when emitted from the source, although they can split and recombine at some detection point?

There are a number of issues with this.
One obvious problem is you would have to be very paranoid to believe this explanation. It would mean that a source that has already been verified to be a true single photon source (via g(2) measurement) somehow magically detects the fact that it is now used in double slit experiment and should now stop emitting single photons and instead become a weak laser (which is basically what "a number of photons in a coherent state" is).

(for the record: you can also do the double slit experiment with e.g.. electrons or other particles where the question about what is a "single particle" is much more straightforward; and it still works as expected)

 

[Johan0001]

There are a number of issues with this.
One obvious problem is you would have to be very paranoid to believe this explanation. It would mean that a source that has already been verified to be a true single photon source (via g(2) measurement) somehow magically detects the fact that it is now used in double slit experiment and should now stop emitting single photons and instead become a weak laser (which is basically what "a number of photons in a coherent state" is).

Agreed electrons and carbon atoms show the same interference.

But my confusion then lies in the definition of where "should now stop emitting single photons and instead become a weak laser" the distinction lies.
Can a weak laser not emit a single photon , and what is the physical mechanism or threshold defines it as a single photon, whether it is a weak laser or not?

 

[DrChinese]

But my confusion then lies in the definition of where "should now stop emitting single photons and instead become a weak laser" the distinction lies. Can a weak laser not emit a single photon , and what is the physical mechanism or threshold defines it as a single photon, whether it is a weak laser or not?

This article might help:

http://blogs.dickinson.edu/pearsonb/files/2012/02/AJPv78p4711.pdf

 

[Cthugha]

Can a weak laser not emit a single photon , and what is the physical mechanism or threshold defines it as a single photon, whether it is a weak laser or not?

Many people are not aware that "single photon" has become a technical term, which is often interpreted incorrectly and needs to be contrasted to what weak laser light would give. The difference is that single photon states are considered to have one photon at most, where weak laser light has one photon most of the time.
The difference is crucial if you consider safety-relevant applications, quantum communication and the like.

The difference is that a true single-photon emitter has a non-linearity that really eliminates the possibility to emit two photons. Consider a single atom in the excited state. Once it has emitted a photon, it will be in the ground state and the probability to emit another photon is exactly zero. The atom is blocked for some time until it is brought to the excited state again. A laser reduced in intensity does not have this blockade effect. There is always a finite probability to have more than one photon.

The way to differentiate is the g2-measurement mentioned above. You basically place a beam splitter and two single photon detectors and check how often they show simultaneous detections. For a real single photon state the photon number variance is zero, so you have no coincidence counts at all. For coherent light photons are statistically independent of each other. So if you take pulsed coherent light and reduce the intensity, the coincidence count rate is easy to calculate. If you have on average one photon per pulse and detector, you also expect one coincidence count per pulse. If you have 1/10 of a photon on average, you expect one coincidence count in 100 pulses. If you have 1/100 of a photon per pulse, you expect a coincidence count every 10000 pulses. As you can see, you can get the coincidence count rate arbitrarily low by using filtered coherent states, but the relative ratio of the variance of the photon number to its mean will be constant for coherent light. For single photons it is strictly zero. This is the difference between a single photon and a weak coherent state. Does this difference matter? For quantum communication: yes. For the double slit: not at all.

 

[Johan0001]

Many people are not aware that "single photon" has become a technical term, which is often interpreted incorrectly and needs to be contrasted to what weak laser light would give

Thank you Cthugha , your distinction is much clearer to me now.

 

[Derek P]

This is the difference between a single photon and a weak coherent state.

Brilliant. Thanks.

 

[vanhees71]

My question was based on the famous , which slit did the photon go through problem.
For example if we there is a possibility that the ' single photon' emitted from the source, was possibly a number of coherent 'photons' then some could go through the left slit and others through the right and then recombine to be detected as one click on the detector screen,thus creating interference over time?
But this would imply that coherent photons are acting as one wave packet which we cannot distinguish as separate wave packets when emitted from the source, although they can split and recombine at some detection point?

There is no "which-way-did-the-photon-take problem". There's only the problem of popular science writers (and unfortunately also teachers at all levels of education) providing a wrong intuition about what photons are. Photons are in fact the most complicated beasts directly observable in all of physics, and one in fact cannot answer the question adequately in a B-level thread. The only way to understand photons really is to learn quantum electrodynamics.

On the B level you can only say to not take too seriously the picture any non-mathematical picture can give you. The worst idea about photons is the idea to think of them as if they were little classical particles, i.e., in some way localizable point-like objects. You cannot even formally define what the position of such a particle-like photon should be. It is almost always more safe to think in terms of electromagnetic waves first. They are always extended "objects", and indeed you can make beams of light that are pretty well loalized in transverse direction (like a laser beam of a laser pointer) but they are always finite in extent. If you should with such a laser beam on a double slit and the beam covers more or less both slits you cannot say through which slit the beam goes and that's why you see a pretty double-slit interference pattern. Now you can dim the laser down very much so that it consists almost only of "vacuum" (i.e., no photons at all). Then you'll most probably register only one photon at a time and you can (in principle) watch the interference pattern build up by "points" hitting a fluorescent screen which are with very high probability caused by a single photon. As was emphasized already above, that's strictly speaking not a pure single-photon source, but it turns out that you can have pure single-photon sources today, i.e., you can prepare light sources which really give only one photon at a time, and you still see the interference pattern been built up on the screen when shooting many such single photons through the double slit.

That is because the theory tells us that a photon is NOT a little particle but something that's described by quantized electromagnetic fields, which quite often behaves such that the intensity distribution of classical electrodynamics (i.e., the energy density of the electromagnetic field) gives the probability to register a photon. This is still not really accurate, because as massless spin-1 quanta photons do not allow for a wave function as massive quanta do in the non-relativistic limit, but it's better than thinking of photons as little particle-like bullets.