How big is a photon?

f95toli

But it DOES matter. There are lots of different single photon sources available, so we now know how to generate snlge photons at many different wavelenghs.
However, all of these sources are fundamentally "quantum mechanical" in that they are able to generate a single excitation, they are very different from a flashlight.
If you start with a thermal source you can of course attenuate it so that it looks like it on average emitts say a single photon per second when you measure the energy it outputs; but it won't be a true single photon source since a thermal field (as it is known) does not contain a fixed number of photons. The emitted radiation simply does not HAVE a property "X number of photons". A source that can generate single photons emitts radiation that is in what is known as a number (or Fock) state, and then this property exists (but the price you pay is that now the phase is undetermined).

Naty1

From the OP:

That's a good description, I think....

Here are a few from others:

Carlo Rovelli:presented in Weinberg's "The quantum theory of fields" vol.1.....

and from a prior discussion in these forums:

Marcus quoting a prior post:

Marcus :

Marcus' comment:

And you will note, of course, these are not all in agreement....

cnickford

My understanding is that when we make a measurement we are really poking at the wavefunction, which holds all the measurable information about a particle within it. When the measurement is made it causes any probabilities to collapse and take on a definite value.

godw2718

Let us assume that a photon propagate in z direction.
For the z direction size, we can make it as small as possible (at least theoretically) by superpositioning a various wavelength photon states, which becomes a delta function in position space while it's just a plane wave in momentum space. Moreover, we can also make it small in x and y direction by superpositioning standing waves, thereby make it small.
If you express a photon as a wave packet, we can see that the wavefunction does not spread as time goes on.(for a mass zero particle while for electrons which as a finite mass it spreads out)
I think, if the principle of superposition in quantum mechanics is valid in all circumstances, we can make it localized in position space.

Cthugha

This already implies that this "duration" of the wavepacket defines some photon size. This is incorrect - besides that you do not take into account emission time uncertainties, there is no reasonable and accepted definition of photon size.

Can you? Really? So what defines your "photon size" then. You can get some superposition to get some kind of spatial localization of the real space photon wavefunction. If you then go ahead and calculate the energy density distribution of that photon, you will find that it is not locally connected to your real space wavefunction. It spreads and falls off as r^-7. Even worse you will find a non-zero detection probability away from the position where you "localized" your wavefunction. This probability also falls off as r^-7. So, which of these quantities defines size now?

For details, read the famous quantum optics bible "optical coherence and quantum optics" by Mandel and Wolf. In my edition chapter 12.11 discusses the problems of a meaningful localization of photons.

Naty1

So while we are at it describing the 'size of a fundamental particle' and seeing there is no 'real' answer, at least no simple one, here is perhaps the craziest explanation of all. From Leonard Susskind whose work in black hole complementarity has won him widespread recognition, from THE BLACK HOLE WAR, Chapter 20:

[Susskind is relating here views of quantum field theory and string theory and while he uses 'atom' in the following description, he is could just as well have used 'particle' or 'photon'

[This refers to the fact than a hovering observer and a free falling observer will 'read' very different radiations emanating from a black hole horizon. So the observed 'size' of a particle as well as the very existence of a particle is impacted by the presence of a cosmological horizon.]

He compares such 'particles' to a rotating airplane propeller....where maybe all we can see is the hub, and maybe the inner portion of the blades....but progressively faster high speed photos would reveal additional extended structure....we can see further out on the rotating blades....see further quantum jitters!!

Khashishi

Size isn't something with a precise physical meaning in this context. The size of a photon depends on how you measure it.

0xDEADBEEF

There are basically two (compatible) definitions of a photon. The usual one is that it is an eigenmode of the electromagnetic spectrum. In the strict sense an eigenmode does not change in time. Therefore in a cavity the eigenmode fills the whole cavity, and in free space a photon is an infinite plain wave (with no amplitude... but well...) Inside these eigenmodes energy is stored, and that is the real idea of a photon. The intensity of the eigenmode drops by a quantized amount a multiple of [tex]\hbar \omega[/tex] when the light field interacts with something else.

When particle physics are discussed they are usually discussed in Fourier space. One infinite plain wave of say protons interacts with another infinite plain wave of protons and they exchange an infinite plain wave of photons or other stuff. The reason why one sees the particle traces in collider experiments is that protons in colliders are a short bit of such a plain wave: a wave packet. But the main physics is captured by the plain wave description.

The interaction of a photon with the other elementary particles in the beams is point like, because its interaction does not depend on the momentum of the particles that it interacts width, leading to flat line in Fourier space and thus a delta peak in real space, for particles like neutrons which have an extend the interaction changes with the momentum of the interacting partners.

So in a way photons are point like (in their interaction) but in another way they can be really large, in a mathematical description as large as the universe.

Sorry if this reply is very technical, but we have gone a long way since the corpuscle theory of Newton, and this is all very much wave particle duality stuff.

sophiecentaur

I think it's worth pointing out that 'the photon' cannot be pictured as some sort of 'burst of oscillations' passing through the aether or as a little bullet. These seem to be the most popular visualisations.
Old habits die hard and, before finally biting the bullet and realising that it's much harder than that, people tend to hang on to the idea that QM is, in fact, just like the old mechanical system but with a few inconsequential tweaks. No. It's 180 degrees different and you just have to get over it. "Physical Interpretation"??? Not possible.

Ger

Size defined as the apparture of your measuring device? Being build of the same stuff your detector is made off, it becomes complex to measure size of your own building block. It then is going to depend on how well (a part of) the wave will interact with your detector, transfering just enough energy to make a difference.

sophiecentaur

Sounds ever so much like a diffraction argument is creeping in, in disguise.