2

This is about photons of very large wavelength, say radio frequencies on the lower end of the electromagnetic spectrum.

Given a wavelength of several hundreds of meters, it seems to me the carrying photon could easily meet a moving obstacle right in the middle of it.

In my picture, I assume the photon to have a size, which may be incorrect in the first place if it is actually a point with no size, thus making the problem disappear.

But if it has a size, can it be broken in two pieces, one keeping on its route and the other reflected or absorbed?

If, on the other hand, a photon is an indivisible object, then will the issue only be a matter of probability, either the photon passing or not passing?

I could determine what happens with waves, e.g. sound pressure waves in water or air, but photons are different.

Winston
  • 3,226
  • What is your mental picture of a photon? Are you thinking of them as wavepackets? – Superfast Jellyfish Sep 16 '20 at 08:05
  • Photon can self-interfere going around obstacle(s) or interfere with past or future photons. Check double-slit experiment, which is a very similar situation in principle to what you have raised here. – Agnius Vasiliauskas Sep 16 '20 at 08:50
  • I know the double slit experiment but I do not see the connection here. – Winston Sep 16 '20 at 08:57
  • The connection is that when the 2-slit experiment is done one photon at a time, the opaque section between the slits is an obstacle that "cuts" the photon. The same thing can be done with any other obstacle, too, with the same result: each photon not absorbed by the obstacle makes one tiny spot on the photographic plate. Doing this for lots of photons, one at a time, gives an interference pattern (distribution of spots) that matches the pattern we'd get from a classical wave that diffracted around the same obstacle. That's what happens empirically, but do you want a theoretical answer instead? – Chiral Anomaly Sep 17 '20 at 02:01

2 Answers2

2

The word photon is used under several different contexts by physicists. And that leads to a huge confusion when one begins to take it seriously in those contexts. Here is Lamb’s famous rant, Anti-photon.

But there is an actual unambiguous and consistently described entity that is named photon, the elementary excitation of the electromagnetic field. For a better understanding of the same, this is an excellent place to start.

You hit up on the key mystery when you say

In my picture, I assume the photon to have a size, which may be incorrect in the first place if it is actually a point with no size, thus making the problem disappear.

That in essence, is the so called wave-particle duality. Because the delocalised entity called photon, is more or less detected as a localised entity always of fixed energy!

So what’s the right interpretation? Like you guessed,

If, on the other hand, a photon is an indivisible object, then will the issue only be a matter of probability, either the photon passing or not passing?

It’s a matter of probability. Or more precisely, the probability amplitude. After each obstacle, there is an associated amplitude to be reflected, transmitted, or absorbed.

For a concrete example, consider a photon impinging on a perfect beam splitter. After passing through, the photon is in an equal superposition of having been reflected and transmitted. enter image description here But if we place a detector in each of the outgoing branches, we will always detect it in one of the detectors. Never simultaneously in both of them. Furthermore if you recombine the two outgoing branches, and place the detector at the place of recombination, you’ll still detect it in one place. But if you repeat the experiment several times, a fringe pattern emerges in the (integrated) detection! This imposes that we can’t imagine the photon as having travelled through one of the paths over the other!

I could determine what happens with waves, e.g. sound pressure waves in water or air, but photons are different.

When light of usual intensity is used, the description of it as an electromagnetic wave can be applied. But we can always prepare the system to contain very few photons, and this is one of the key differences between the waves you have mentioned and light. There is also another great mystery which is entanglement but that’s a story for another day.

Superfast Jellyfish
  • 9,348
  • Thanks for your answer but you do not seem to address my question directly. My question is about imposing an obstacle at a time when waves would be split in the middle of their cycle, but having photons instead. – Winston Sep 16 '20 at 05:28
  • Each photon has an associated plane wave profile. In fact that is its defining feature. The advantage of treating them as particles is in the fact that describing detection becomes natural. – Superfast Jellyfish Sep 16 '20 at 05:31
  • Not so natural here anyway because I still cannot get a proper answer. When emitting radio waves, we know which phase of the wave is being transmitted. So it is not like it is random. Now having this information, I am asking what happens in the photon picture if they are interrupted at a point where the wave cycle is not complete. – Winston Sep 16 '20 at 05:37
  • 1
    To detect the classical electric field, you need to integrate over multiple detections. The number of detections at a point in time and space tend to the field intensity at that point with appropriate scaling. – Superfast Jellyfish Sep 16 '20 at 05:56
  • I give up, I do not understand your answers. – Winston Sep 16 '20 at 07:02
  • I will try to be rephrase my answer after my classes today. Can you help me out by expressing precisely what you are trying to understand? – Superfast Jellyfish Sep 16 '20 at 07:57
-2

A photon is a point particle. An electromagnetic wave describes these in a probabilistic sense. If you alter the wave by introducing an obstacle you alter the probability distribution of photons by reflection and diffraction of the wave.

Superfast Jellyfish
  • 9,348
my2cts
  • 24,097