How can I get only one single photon with a coherent light source?

[alizeid]

Hi!

We will make a lab analogous to the stern gerlach experiments but with polarized light. How can we get only one single photon in experiment with a coherent light source? I'm going to make a lab where we need to get in only one single photo at a time. I have read that you can use some inhomogeneous crystals to get a very small fraction of the photos. But how do you really do ?? What confuses me is that a laser has, for example, 10mW, only here, we have basically 10 ^ 15 photons (in order of magnitude) per second. How do you stop all photos and get through a single photo?

I'm grateful for answers.

 

[berkeman]

Hi!

We will make a lab analogous to the stern gerlach experiments but with polarized light. How can we get only one single photon in experiment with a coherent light source? I'm going to make a lab where we need to get in only one single photo at a time. I have read that you can use some inhomogeneous crystals to get a very small fraction of the photos. But how do you really do ?? What confuses me is that a laser has, for example, 10mW, only here, we have basically 10 ^ 15 photons (in order of magnitude) per second. How do you stop all photos and get through a single photo?

I'm grateful for answers.

How are you planning on detecting the photons? With a PMT?

It seems like you just need to increase the optical attenuation to the point where you get the extremely low flux that you want. How many photons per second is low enough for your experiment? Something like 10-100 as long as they are singles?

 

[berkeman]

https://en.wikipedia.org/wiki/Single-photon_source#Faint_laser

Faint laser
One of the first and easiest sources was created by attenuating a conventional laser beam to reduce its intensity and thereby the mean photon number per pulse[18]. Since the photon statistics follow a Poisson distribution you can achieve sources with a well defined probability ratio for the emission of one versus two or more photons. For example a mean value of μ = 0.1 leads to a probability of 90% for zero photons, 9% for one photon and 1% for more than one photon[19].

Although such a source can be used for certain applications, it has a second-order intensity correlation function equal to one (no antibunching). For many applications however, antibunching is required, for instance in quantum cryptography.

 

[alizeid]

https://en.wikipedia.org/wiki/Single-photon_source#Faint_laser

"Since the photon statistics follow a Poisson distribution you can achieve sources with a well defined probability ratio for the emission of one versus two or more photons"

I do not really understand how the connection between reducing intensity and Poisson distribution. If we have 10 ^ 15 photons, we can, for example, reduce the number of photons through a medium. I decrease the number of photons coming through the medium. What does this have with the probability of emission ?? The number of photons from a blue rays of 10 mW is about 10 ^ 15 photons. These are photons that have already been emitted, so how can I influence the number of emissions when they have already been emitted from the laser?