Physics Forums

I would like to open this new topic to continue discussions on single photons which turned up in several other topics, but were a bit out of place there. So here the discussion continues. The first two quotes were basically about coincidence counting, photon statistics and the HBT-experiment. The last post was about polychromaticity of single photons.

sophiecentaur said:

Does that mean that the total number of counts is higher for thermal light in that experiment? Or that, during a random set of counts, some happen to coincide? There are Energy implications here, I think.

sophiecentaur said:

So you are saying (confirming) that photons have only ever been seen coming out of one slit?

I am not quite sure I know what you mean by "coming out of one slit". To get back to the HBT experiment: The coincidence count rates basically give you a method to check the nature of your light field. For example one could have a steady flow of exactly one photon every few ms. That would be a non-classical Fock state. You could have a randomly fluctuating number in the same time interval. That identifies coherent light. Or you might have strongly fluctuating photon numbers where you mostly have no light at all and lots of light at few times. hat corresponds to thermal light. I could back that up with some math, but I do not know whether that makes visualization easier.

sophiecentaur said:

This is beginning to make some sense to me but there are some loose ends. For such a photon to be absorbed (detected) it would need to deliver all its energy into the detecting system. It would need to match the 'transmitter' characteristics and have some sort of 'spectrum' in its own right. What could its energy be? Would it be a version of hf; a kind of
h∫(F(f)df expression, giving a 'massively energetic' photon?
In your description, are you really saying that this sort of photon is in the minority (I guess you are)? These photons would not be like your regular photons which are totally anonymous - able produced by one process, possibly frequency shifted on their journey and then absorbed by a totally different process - as with light produced thermally in a star and then absorbed in a particular gas atom transition. What sort of fractional bandwidth are we talking about?
Is this any different from the familiar photoelectric emission, in which a range of photo-electron energies will result from a broad range of incident frequencies?

Yes, you are right. Upon detection you will of course only have one energy "delivered" to the detector. Collapse to some eigenstate via measurement breaks superpositions. This is true for single photons just like in ordinary qm. It is just like asking for the momentum or emission direction of a single photon emitted by a symmetric emitter. You will detect it at some certain position, but until then it is typically in a superposition state of all possible directions. I am not quite sure how similar this is to photoelectronic emission. This is not really a topic I am familiar with, but I remember that it is more complicated than it looks at first sight, so I do not really dare to give a definitive answer.