- #1

- 528

- 15

Thanks

You are using an out of date browser. It may not display this or other websites correctly.

You should upgrade or use an alternative browser.

You should upgrade or use an alternative browser.

- Thread starter jaydnul
- Start date

- #1

- 528

- 15

Thanks

- #2

sophiecentaur

Science Advisor

Gold Member

- 27,008

- 5,792

How could one even define the frequency of the photon whilst it is on its way, bearing in mind that it can have been produced by a different system thane the one that absorbs it? You have, in effect, a transmitter with some finite bandwidth and a receiver of finite bandwidth so all you know is that the energy was at some frequency in the 'overlap' between the two bandwidths with the same nominal frequency but with a finite difference. It cannot be looked upon as a little wiggly thing that zaps from place to place, any more than it can be looked at as a little bullet.

The problem you describe in reconciling the idea of waves and fields shows that the simple model of what goes on in between source and absorber is not good enough. There is a finite time involved with the atomic energy transitions and that corresponds to a spread in frequency (like the Q of a resonant circuit) the Q of the transmitting system could be different from the Q of the absorbing system yet the photon would have to be the same; it wouldn't know what sort of atom / system it was going to interact with it whilst it was on its journey.

- #3

Khashishi

Science Advisor

- 2,815

- 493

- #4

sophiecentaur

Science Advisor

Gold Member

- 27,008

- 5,792

Yes. And that wavelength can only really be assigned to the average effect of a large number of photons - large enough to produce a 'good' distribution (good enough to show a pattern). Each photon is, of course,interacting with a different detector / atom / molecule at different places on the projection screen (or whatever).

- #5

f95toli

Science Advisor

Gold Member

- 3,278

- 773

Yes. And that wavelength can only really be assigned to the average effect of a large number of photons - large enough to produce a 'good' distribution (good enough to show a pattern). Each photon is, of course,interacting with a different detector / atom / molecule at different places on the projection screen (or whatever).

I am not sure I agree with this. There are certainly quite a number of examples of systems where we are dealing with single photons, and where the "wavelenght" matters.

A typical example would be cavity-QED experiments, where we can trap a single photon in a cavity with a lenght lambda/2.

Hence, although the concept of wavelenght becomes a bit fuzzy when dealing with single photons, it is not meaningless.

If you go through the math for a cavity-QED setup using a lambda/2 resonator you will also find that where the node would be for a classical EM wave is exactly where the coupling strength to the photon in the cavity is at maximum.

- #6

sophiecentaur

Science Advisor

Gold Member

- 27,008

- 5,792

My reason for bringing this up is to point out that the actual existence of a photon is only certain at either end of any process. Postulating its behaviour 'in between' as particular is only using an alternative model to describe the energy trapped in the cavity. (Is that being too controversial? )

- #7

f95toli

Science Advisor

Gold Member

- 3,278

- 773

This can be used to create single photon sources.

Also, if a non-linear cavity/system is used there are ways of detecting photons non-destructively; meaning you can tell if there is a photon in the cavity without actually destroying it (a "simple" way of doing this involves using the non-linear Kerr effect to detect the number of photons in one mode of the cavity by monitoring another mode).

Haroche won his share of the Nobel prize last year for doing experiments in this field.

- #8

sophiecentaur

Science Advisor

Gold Member

- 27,008

- 5,792

- #9

f95toli

Science Advisor

Gold Member

- 3,278

- 773

Also, the photon does not have wavefunction, so it does not have normal probability distribution (you can plot the Wigner function and other distributions; but they can be negative and so has no "classical" interpretation)

- #10

sophiecentaur

Science Advisor

Gold Member

- 27,008

- 5,792

I still have a bit of a problem with the concept of an 'exact' wavelength, though, in as far as the frequency of the source and the accuracy of the measurement must surely come into it. It's fair enough to have a mathematical model for what's going on in the cavity but it worries me that a popular view of this could not coincide with the actual meaning contained in the QED description. Any given photon would need to be able to interact with a range of atoms, in a spread of actual energy states or you couldn't be sure of ever getting an exact transition to occur. Do you get my point about this? It seems that a source and detector could only 'talk to each other' under exact conditions and this is never actually the case because there are always more Quantum Numbers involved than those which are talked about at a simple level. (Pauli forbids this, doesn't he?)

When you say that the photon does not have a wave function, is there something about the field pattern in a cavity with a single photon in a cavity that makes things different from what you get in a normal standing wave? Where does that Foch state come into it, if not to describe the pattern?

Share: