[chris (OP)]

In situations such as the double-slit experiment, which shows that light behaves as a particle and a wave, scientists talk about sending single photons through the apparatus yet still achieve an interference pattern. This proves that the photon must be behaving both as a particle and a wave.

But is it really feasible to produce a single photon and, if so, how?

Chris

[Vern]

Yes; we think we can generate single photons on demand. But it takes a little effort.

This paper describes how to do it.

Every time you switch on a light bulb, 10 to the power of 15 (a million times a billion) visible photons, the elementary particles of light, are illuminating the room in every second. If that is too many for you, light a candle. If that is still too many, and say, you just want one and not more than one photon every time you press the button, you will have to work a little harder. A team of physicists in the group of Professor Gerhard Rempe at the Max Planck Institute of Quantum Optics in Garching near Munich, Germany, have now built a single-photon server based on a single trapped neutral atom. The high quality of the single photons and their ready availability are important for future quantum information processing experiments with single photons. In the relatively new field of quantum information processing the goal is to make use of quantum mechanics to compute certain tasks much more efficiently than with a classical computer. (Nature Physics online, March 11th, 2007)

[lightarrow]

In situations such as the double-slit experiment, which shows that light behaves as a particle and a wave, scientists talk about sending single photons through the apparatus yet still achieve an interference pattern. This proves that the photon must be behaving both as a particle and a wave.

But is it really feasible to produce a single photon and, if so, how?

Chris

Reducing the intensity of the light beam: fixed the wave's frequency, the intensity is proportional to the number of photons emitted in the unit time.
http://www.aip.org/pnu/2002/split/591-3.html

[lyner]

In situations such as the double-slit experiment, which shows that light behaves as a particle and a wave, scientists talk about sending single photons through the apparatus yet still achieve an interference pattern. This proves that the photon must be behaving both as a particle and a wave.

Should we be so cavalier in out description of photons?
The only thing we can say, definitely, about photons is that their energy is well defined (they are quanta) and that wave calculations will give us an indication about where their effect will be felt. Evidence about their 'particle' nature is a bit more tenuous, I feel.

A single 'particle' does not exhibit interference- the statistics of a large number of particles can be said to mimic a classical interference pattern. I think it can interfere with a proper understanding.

It concerns me that a picture in our minds (a metaphor) is used to define something which is  unique and not necessarily  as we commonly describe it.

[yor_on]

Hmm, that is probably a trickier question than what it seems?
"is it really feasible to produce a single photon and, if so, how"
I agree it is a remarkable feat to be able to produce one at a time but as Vern points out we seem to be able to do so. As for how

"A light source that emits only one photon at a time would be an invaluable tool for quantum optics and quantum computing. Quasi- zero-dimensional systems such as atoms, nitrogen vacancies in diamond, and quantum dots emit photons one by one when excited by laser light. The reason for this is that when charge carriers are strongly confined, the multiple excited states necessary for emission of two photons do not exist.

In materials that are extended in one or more dimensions, multiple excited states (namely, electron-hole pairs) can coexist and these materials tend to be poor candidates for single-photon emitters. However, in the 27 May 2008 issue of Physical Review Letters, Alexander Högele, Christopher Galland, Martin Winger, and Ataç Imamoğlu report that at low temperatures, semiconducting carbon nanotubes act as single-photon emitters when excited by a laser beam.

The unexpected finding is due to a combination of effects that prohibit double occupancy of excited states: the electron-hole pairs are highly localized and Auger processes (in which electron-hole pairs recombine without photon emission) are strong. Fewer than 1 in 20 events are reported to be multiphoton emissions, making carbon nanotubes promising single-photon sources."

So it may be used in quantum computing, isn't that when you introduce superpositioning as a method for getting a answer? But how do you superposition one photon? One way is to do like this.


"This experiment uses photodetectors, which are sensitive enough to detect even a single light particle (a photon) and beam splitters. Classically, a beam splitter can be seen as an intensity divider, letting half of the light through and reflecting the other half. The intensity of the laser is reduced so that only a single photon is travelling through the setup at any in time. Using the setup it is expected that a photon will have a 50% probability to reflect and an equal probability to transmit. This results in an equal probability of detecting the photon at detector 1 or detector 2.

Now that we know a bit more about quantum mechanics we can also argue that a beam splitter puts a photon into a superposition in which it is half-transmitted and half-reflected. Upon detection by a photon detector, the quantum mechanical state collapses into either the transmitted state or the reflected state, ending up in detector 2 or 1 respectively. Because the photon is in a superposition of an equal amount of transmission and reflection, the same result is expected that the classical view of a beam splitter predicts; detector 1 and 2 detect the same amount of light, half of the total amount "

http://www.electronicsforu.com/electronicsforu/Articles/ad.asp?url=/EFYLinux/efyhome/cover/may2005/Quantum-Computers.pdf&title=Computers%20Taking%20a%20Quantum%20Leap

[yor_on]

There is a third use for photons or possibly two 'third' uses:)
One is encrypting and the one hand in hand with that is decrypting.
http://www.nikon.com/about/feelnikon/light/chap04/sec01.htm

[JP]

In situations such as the double-slit experiment, which shows that light behaves as a particle and a wave, scientists talk about sending single photons through the apparatus yet still achieve an interference pattern. This proves that the photon must be behaving both as a particle and a wave.

Should we be so cavalier in out description of photons?
The only thing we can say, definitely, about photons is that their energy is well defined (they are quanta) and that wave calculations will give us an indication about where their effect will be felt. Evidence about their 'particle' nature is a bit more tenuous, I feel.

A single 'particle' does not exhibit interference- the statistics of a large number of particles can be said to mimic a classical interference pattern. I think it can interfere with a proper understanding.

It concerns me that a picture in our minds (a metaphor) is used to define something which is  unique and not necessarily  as we commonly describe it.

Indeed.  The usual idea that you can tune down a light source such as a laser until you get only one photon at a time is a bit subtle.  You know how many things in QM can't be determined until you try to measure them?  In the case of laser light, you can't determine how many photons are in a state until you measure it.  In other words, the state consists of a wide range of photon numbers until you measure it and force it to choose a number.  If you tune down the intensity of your laser, you make it more and more likely that it will choose a 1-photon state, but you don't guarantee it.  

The single photon states in Vern's article seem to be genuine single-photon states, where they consist purely of 1 photon, more formally called single photon "Fock states."  There's been other research into producing these, including generating entangled pairs of photons via the standard technique of parametric down conversion, and throwing one photon out:
http://en.wikipedia.org/wiki/Spontaneous_parametric_down_conversion

[lyner]

Doesn't a laser produce more than one photon by definition? Photons are emitted by stimulated radiation- one in causes at least one other out.

[Vern]

I think the idea was that when you reduce the transmitter to one atom, the transmitted photon is singular. It represents one electron transition from one state to another. I don't see a problem in that.

[JP]

Doesn't a laser produce more than one photon by definition? Photons are emitted by stimulated radiation- one in causes at least one other out.

It should.  You excite the atoms somehow, then send a photon by and get an identical photon as well.  This is probably why you can't pick a single photon out of a laser, no matter how much you decrease the beam's intensity.  The more interesting point, I think, is that the light waves of Maxwell's theory all have this property of not being in well-defined photon-number states, which is probably a reason why it's so hard to make the photon picture make sense. 

[lyner]

Do you mean Maxwell implies a continuum?

[yor_on]

jpetruccelli you write that "This is probably why you can't pick a single photon out of a laser, no matter how much you decrease the beam's intensity."

Here they seem to state that you can? using superposition.
http://www.sciam.com/article.cfm?id=how-can-a-single-photon-p

[JP]

jpetruccelli you write that "This is probably why you can't pick a single photon out of a laser, no matter how much you decrease the beam's intensity."

Here they seem to state that you can? using superposition.
http://www.sciam.com/article.cfm?id=how-can-a-single-photon-p

Hey!  That's my University!  (And I know those folks.)  Anyway, if you read carefully, "the group prepared weak pulses of light that on average contained less than one photon."  I know they also do work with entangled photons by using a technique called "parametric down conversion" that generates 2 (entangled) photons at a time. 

[JP]

Do you mean Maxwell implies a continuum?

I've been told so.  I've never seen a derivation proving it, but what I've been told is that classical coherent light is generated by quantum "coherent state."  Coherent states are made by a superposition of photon-number states.  I'll have to go read up a bit more to really justify it to myself, and to figure out how it applies to light that isn't fully coherent, but I think it's just an extension of these coherent states. 

[lightarrow]

Hey!  That's my University!  (And I know those folks.)  Anyway, if you read carefully, "the group prepared weak pulses of light that on average contained less than one photon."  I know they also do work with entangled photons by using a technique called "parametric down conversion" that generates 2 (entangled) photons at a time. 

I've read that linked article, but what does it mean that some pulses didn't contain any photon? How could they have established it?

[Bored chemist]

I'm not sure how you would do double slit type experiments with them but gamma ray photons are produced one at a time.

[JP]

Hey!  That's my University!  (And I know those folks.)  Anyway, if you read carefully, "the group prepared weak pulses of light that on average contained less than one photon."  I know they also do work with entangled photons by using a technique called "parametric down conversion" that generates 2 (entangled) photons at a time. 

I've read that linked article, but what does it mean that some pulses didn't contain any photon? How could they have established it?

One way to do it is to try to measure "clicks" on your detector, which correspond to a photon hit.  If you have a state coming in that has some probability of being 1 photon and some probability of being 0 photons, and you know when it should arrive, you can watch for a "click" at that time.  The trick is beating out the quantum noise, which you can probably do by repeating the experiment until you get enough data.

[lightarrow]

Do you mean Maxwell implies a continuum?

I've been told so.  I've never seen a derivation proving it, but what I've been told is that classical coherent light is generated by quantum "coherent state."  Coherent states are made by a superposition of photon-number states.  I'll have to go read up a bit more to really justify it to myself, and to figure out how it applies to light that isn't fully coherent, but I think it's just an extension of these coherent states. 

What I know is that there is a quantum indeterminacy relation between phase and photon number: the more the phase is determined, the less it is the number of photons; so if the wave is totally coherent, the number of photons is completely indetermined and so, in this sense, we could maybe conclude that the "quantumness" is lost in this case.

[lightarrow]

I've read that linked article, but what does it mean that some pulses didn't contain any photon? How could they have established it?

One way to do it is to try to measure "clicks" on your detector, which correspond to a photon hit.  If you have a state coming in that has some probability of being 1 photon and some probability of being 0 photons, and you know when it should arrive, you can watch for a "click" at that time.  The trick is beating out the quantum noise, which you can probably do by repeating the experiment until you get enough data.

Do you mean that what they call "pulse" is just something "expected" from the emission?

[JP]

I've read that linked article, but what does it mean that some pulses didn't contain any photon? How could they have established it?

One way to do it is to try to measure "clicks" on your detector, which correspond to a photon hit.  If you have a state coming in that has some probability of being 1 photon and some probability of being 0 photons, and you know when it should arrive, you can watch for a "click" at that time.  The trick is beating out the quantum noise, which you can probably do by repeating the experiment until you get enough data.

Do you mean that what they call "pulse" is just something "expected" from the emission?

I'm not sure I 100% follow.  I think their "pulse" is a very low-energy packet of light localized in time.  They know when it will arrive and have it set up so it's likely that they see either 0 or 1 photon when they try to observe it.