[mattnumbers]

When we discovered that neutrinos oscillate, we knew that they had mass because we were able to observe a change from within our frame of reference, implying a finite time dilation. I thought that light was unable to change its physical properties in time (as observed from our frame of reference) due to infinite relativistic time dilation. How is it possible that we can observe torque, a change in angular momentum, for a particle moving at the speed of light?

[abdullahkhalids]

All special relativity says is that if an object is moving at the speed of light, its motion will appear to be the same in every reference frame. That is all. It says nothing on what can happen to photons when they are interacted with other photons or or with other media. Photons have several internal properties and all of these can be manipulated, either in the lab or naturally. The polarization of photons can be changed by Faraday rotation for example. In cosmology, you might be aware of Hubble's law, which involves the changing of the frequency of light.

I haven't read this paper or intend to, but here is a key quote

> We demonstrate that the self-torque arises as a necessary consequence of angular momentum conservation during the extremely non-linear, non-perturbative optical process of high-order harmonic generation (HHG). In HHG, the interaction of an intense field with an atom or molecule leads to the ionization of an electronic wavepacket, which acquires energy from the laser field before being driven back to its parent ion, and emitting a high-frequency photon upon recollision.

What they are saying essentially is that by interacting the light with matter they are able to generate the said torque.

[bladedtoys]

I think the question comes from the following understanding:

- change requires the passage of time.

- particles moving at the speed of light relative to us have infinite time dilation and so no time passes from our point of view.

- the article is talking about particles moving at the speed of light which under go change.

I have the same (mis?)understanding and am curious how this can work as well.

[abdullahkhalids]

Your second point is where all the trouble is arising. Special relativity is talking about one thing and one thing only - how an object moves from one point to another in space, and how fast this process appears. The position (and speed and acceleration) are external properties of an object, they depend on who is doing the describing of these properties. Special relativity is concerned mainly with these things.

Objects also have other internal (non-space/position) related properties. For example, light has polarization. These properties can and do change with the passage of time. For example, you can change the polarization of light with a strong magnet. Relativity can and does place some restrictions on how this evolution happens, in so far as something is moving in space, but that is a more involved story.

[carlob]

And the third point is where that trouble can be resolved: light doesn't move through matter at c.

[mygo]

> Objects also have other internal (non-space/position) related properties. For example, light has polarization.

So the angular momentum of light would be put under the same category as polarization, which is a property that a single photon can have?

Or is it a property like wavelength, momentum, frequency, intensity, etc -- a property of light that can be observed only when the light is a system of lots of photons, instead of just an individual photon?

Also, sorry if my categories are wrong. If so, I hope it doesn't take away from my main question of whether it is a property of a single photon or a property of a system of photons

[abdullahkhalids]

Every category you have named, wavelength/frequency, intensity (function of frequency), linear momentum (function of frequency), polarization are properties that individual photons have. It is possible to construct photon detectors that are able to resolve these properties for a single photon thrown at them.

Orbital angular momentum is also a property of single photons, related to the shape of their wave function. You can read about it and see some example shapes on the wikipedia page https://en.wikipedia.org/wiki/Orbital_angular_momentum_of_li...

I can't think of a property of light that is only in the bulk and not at the single photon level (somebody correct me if I am wrong). But I haven't read this paper in any detail to find if the "torque" is a bulk property or a single photon property.

[nine_k]

Wavelength = 1 / frequently and is directly related to photon's energy. You only need one photon to observe these.

Intensity is number of photons per unit area per unit time, so it takes many photons normally, but a single photon produces a non-zero intensity.

[ridewinter]

...so no time passes from the photon’s point of view.

Infinite time dilation means that a photon from a galaxy 3 billion light years away arrives here instantaneously from the photon’s perspective. Which means there’s actually no distance from the photon’s perspective - the entire universe is a singularity. Pretty mind-blowing IMO.

[Retra]

That would be true if you can find a frame of reference in which the photon's velocity is 0 (its rest frame -- a pre-requisite for associating that frame of reference with the photon in the first place). But special relativity does not allow photon velocity to be zero in any frame of reference, so there is no frame of reference associated with a photon.

[ridewinter]

Is that like saying that time & distance have no meaning for a photon?

[jfengel]

More or less.

It goes on to mean that nothing really "has a meaning" for a photon, so it's hard to really talk about a photon's perspective one way or the other. We generally just say that the photon doesn't have a perspective, i.e. that any discussion of the photon's rest frame is meaningless.

So to the degree that your question has an answer, the answer is "yes". But a better answer is that it's misleading to think that the question even has a yes-or-no answer.

[pontifier]

Take it a step farther and the photon isn't even truly nessicary. The recoil experience when the photon was emitted, and the push felt when it is absorbed should match exactly, and essentially correspond to a single, equal, collision event to the right observer.

[chongli]

Photons only travel at the speed of light in a vacuum, when they're not interacting with anything. Throw interactions into the mix, and they go much slower.

The light produced by fusion in the sun's core can take up to 170,000 years to escape! Plenty of time for the angular momentum (or any other property) of the light to change.

[mattnumbers]

Thanks for the explanation. I completely misunderstood the effect. Now, I misunderstand it a little bit less. Progress!

[pdonis]

> I thought that light was unable to change its physical properties in time (as observed from our frame of reference) due to infinite relativistic time dilation.

This is not correct. What is correct is that the concept of a "rest frame" for a light ray doesn't make sense. But our rest frame, as observers watching the light, is perfectly well defined, and in our frame it's perfectly possible to see light changing with our time.