Vacuum has friction after all - some physics questions

[Askance]

These questions are provoked by this New Scientist article Vacuum has friction after all. It essentially says that an object spinning in pure vacuum will eventually slow down due to friction with the virtual photons that continually pop in and out of existence even in a vacuum.

1/ This is said to happen because

[QUOTE=NS]
a virtual photon hitting an object in the direction opposite to its spin collides with greater force than if it hits in the same direction.

So over time, a spinning object will gradually slow down, even if equal numbers of virtual photons bombard it from all sides. The rotational energy it loses is then emitted as real, detectable photons
[/QUOTE]

… so imparting a greater force against the object’s spin than with that spin, thus slowing it down. But my understanding was that the quality called “spin” in atomic physics was just a label for a property that had no real analog in the world of classical physics. The relevant Wikipedia article seems to say this is not so, and elementary particles’ spin is real spin - kinda. That is, it has some of the properties of classical-world spin including angular momentum (or something that acts like it, anyway):

[QUOTE=Wikipedia]
Spin is a type of angular momentum, where angular momentum is defined in the modern way (as the “generator of rotations”, see Noether’s theorem). This modern definition of angular momentum is not the same as the historical classical mechanics definition, L=r×p. (The latter definition, which does not include spin, is more specifically called “orbital angular momentum”.
[/QUOTE]

Can anyone clarify this for me; in what sense is a photon’s “spin” actually spin? How does that allow it to transfer angular momentum to a large scale object?
2/ As I understand it these virtual photons occur in pairs, that mutually annihilate after a brief life. But what happens when one of the pair has interacted with something (in this case, lost some of its “spin”) and the other has not? I know this is the basis of Hawking Radiation, but does it take a black hole to cause the pair to behave differently than each other, and interfere with the particles disappearing again?
3a/ The article also says:

The rate of deceleration also depends on temperature, since the hotter it is the more virtual photons pop in and out of existence, producing the friction.

But in the situation under consideration we are in a vacuum, unpopulated by any particles other than the virtual ones. Isn’t it so that temperature is solely a measure of particle energy; if so how can there be a higher or lower temperature in a vacuum? On the other hand if we are not in a perfect vacuum it’s unremarkable that the object would experience friction and slow down.

3b/ This thought also triggers another: does the soup of virtual photons have a temperature, absent any real particles?

[Chronos]

Quantum-mechanical spin is definitely a kind of angular momentum, just a kind different from what we’re familiar with in the macroscopic world (all of which is ultimately what’s called “orbital angular momentum”). Neither orbital nor spin angular momentum is conserved; it’s only the total angular momentum, which is the sum of the two, is conserved. So there’s no physics being violated by a spinning object slowing down by radiating photons.

[Sailboat]

Is this the point at which we owe an apology to the people who came up with the theory of aether? Or do we have a bit more research to go, first?

[Asympotically_fat]

Some of this is a WAG, be warned:

1. as Chronos pointed out spin is a type of angualr momentum, another namefor it is intrinsic angular momentum infact. However it can be expalined in terms of the rotation of the particle that’s what makes it different, but it still contributes to the total angualr momentum.

2. I think it’s best to approach any explaination that invokes virtual particles with caution. This where my WAG comes in, I think they are talking about an effect simlair to the Unruh effect. In the rotating frame of the body in question particles appear which do not appear in the inertial body-centred frames. In the rotating frame these particles are just as real as any other particles and it’s these particles that slow down the body’s rotation (thoguh like I say this is a bit of a WAG, but I’d be willing to bet money that this is just a result of a variation of the Unruh effect).

3)a) very good point. It may be that they’re referring to the body itself more likely they’re just referring to ambient black body raditation (which itself would take the form of particles such as photons).

3)b) Unruh radiation is takes the form of a ‘thermal bath’.

[Asympotically_fat]

@Sailboat

Is this the point at which we owe an apology to the people who came up with the theory of aether? Or do we have a bit more research to go, first?

No because we still have Lorentz-symmetry. Aether theories are not invariant under a change in velocity, the theories involved here are.

[panamajack]

It seems to me that the effect being described has nothing to do with the QM usage of “spin” at any rate. It’s talking about any object spinning in a classical fashion. If you consider the force from the virtual photons to be just a Newtonian force, it’ll tend to reduce the object’s rotation.

Consider a similar situation : You have a large wooden paddle that is spinning in one direction. There are two ping-pong ball cannons mounted away from the paddle and opposite each other. They are timed in such a way that whichever one fires, it will hit the paddle exactly when it is perpendicular to the cannon. So one of the cannons will hit the paddle in a “head-on” collision, directly as the paddle swings toward the ball, and the other will “read-end” the paddle as it moves away from the shot.

Hopefully you can see that in this situation, the “head-on” hits will tend to slow the paddle down more than the hits from behind will speed it up. If you now send an equal number of shots from both sides in random order, the paddle will end up spinning more slowly than before.

The actual mechanism involved is more complicated than that, but unless I’m missing the point of the article, that seems to be a correct analog.

[Ludovic]

@These are my own pants

However it can be expalined in terms of the rotation of the particle

That’s the SI unit of spin, right?

[Asympotically_fat]

@panamajack

It seems to me that the effect being described has nothing to do with the QM usage of “spin” at any rate. It’s talking about any object spinning in a classical fashion. If you consider the force from the virtual photons to be just a Newtonian force, it’ll tend to reduce the object’s rotation.

Consider a similar situation : You have a large wooden paddle that is spinning in one direction. There are two ping-pong ball cannons mounted away from the paddle and opposite each other. They are timed in such a way that whichever one fires, it will hit the paddle exactly when it is perpendicular to the cannon. So one of the cannons will hit the paddle in a “head-on” collision, directly as the paddle swings toward the ball, and the other will “read-end” the paddle as it moves away from the shot.

Hopefully you can see that in this situation, the “head-on” hits will tend to slow the paddle down more than the hits from behind will speed it up. If you now send an equal number of shots from both sides in random order, the paddle will end up spinning more slowly than before.

The actual mechanism involved is more complicated than that, but unless I’m missing the point of the article, that seems to be a correct analog.

Yes, that’s correct, there’s no real need to consider intrinsic angualr momentum. Though I think you’d probably find that the angualr momentum of the spinning body ends up as the intrinsic angualr momentum of the particles it emits.

[Asympotically_fat]

@Ludovic

That’s the SI unit of spin, right?

It’s a typo as I’m sure was self-evident.

Usually I would say pointing out trivial typos is absolutely pointless, but in this case it’s a double typo because it should be: “However it can’t be explained in terms of the rotation of the particle”. Which is obviosuly quite an important difference.

[Askance]

@Chronos

Quantum-mechanical spin is definitely a kind of angular momentum, just a kind different from what we’re familiar with in the macroscopic world (all of which is ultimately what’s called “orbital angular momentum”). Neither orbital nor spin angular momentum is conserved; it’s only the total angular momentum, which is the sum of the two, is conserved. So there’s no physics being violated by a spinning object slowing down by radiating photons.

I hope I didn’t imply it was. My point was that I could see how the explanation given in the article (if a colliding virtual photon’s spin directionally matches the object’s spin then that slows the object’s spin, whereas if their spins were opposite then there’s no, or a lesser, speeding-up effect) works if the virtual photons’ spin is real classical spin, rather than a QM analog of spin.

So it seems from the answers so far that the QM-spin-analog is just enough like the spin of the “real world” so as to have this effect, in that it conserves total angular momentum.

[Askance]

@These are my own pants

It’s a typo as I’m sure was self-evident.

Usually I would say pointing out trivial typos is absolutely pointless, but in this case it’s a double typo because it should be: “However it can’t be explained in terms of the rotation of the particle”. Which is obviosuly quite an important difference.

Actually it was quite a clever play on the exact nature of your typo:

However it can be expalined in terms of the rotation of the particle

That’s the SI unit of spin, right?

[Askance]

@panamajack

It seems to me that the effect being described has nothing to do with the QM usage of “spin” at any rate. It’s talking about any object spinning in a classical fashion. If you consider the force from the virtual photons to be just a Newtonian force, it’ll tend to reduce the object’s rotation.

Consider a similar situation : You have a large wooden paddle that is spinning in one direction. There are two ping-pong ball cannons mounted away from the paddle and opposite each other. They are timed in such a way that whichever one fires, it will hit the paddle exactly when it is perpendicular to the cannon. So one of the cannons will hit the paddle in a “head-on” collision, directly as the paddle swings toward the ball, and the other will “read-end” the paddle as it moves away from the shot.

Hopefully you can see that in this situation, the “head-on” hits will tend to slow the paddle down more than the hits from behind will speed it up. If you now send an equal number of shots from both sides in random order, the paddle will end up spinning more slowly than before.

The actual mechanism involved is more complicated than that, but unless I’m missing the point of the article, that seems to be a correct analog.

Ah, I see what you’re getting at now! Indeed the article doesn’t explicitly say it’s the spin of the virtual photons that slow the spinning object, just the fact of their collisions with it regardless of their spin. Got it.

[Asympotically_fat]

@Askance

Actually it was quite a clever play on the exact nature of your typo:

That one went right over my head.

[John_DiFool]

Related question: is it possible for an object (assuming macro-size here) to form in a spinless state, and just kind of float there without rotating?

[Chronos]

Ah, I see what you’re getting at now! Indeed the article doesn’t explicitly say it’s the spin of the virtual photons that slow the spinning object, just the fact of their collisions with it regardless of their spin. Got it.

Hm, true. The object could also emit photons with orbital angular momentum.

I wonder how this mechanism would compare to rotation-slowing by gravitational radiation, which also works in a vacuum? It’d probably have to depend on the geometry of the object, but if we take some generic shape, would it be in proportion to the charge-to-mass ratio?

[Askance]

@Chronos

Hm, true. The object could also emit photons with orbital angular momentum.

Good question. When a spinning object emits photons due to collisions with virtual photons, do those emitted photons have any angular momentum due stemming from that of their parent object?

[jackdavinci]

@panamajack

Hopefully you can see that in this situation, the “head-on” hits will tend to slow the paddle down more than the hits from behind will speed it up. If you now send an equal number of shots from both sides in random order, the paddle will end up spinning more slowly than before.

Assuming the paddle spins faster than the ball speed yes. But, in that particular situation, it would seem that on average, hits will tend to eventually change the absolute value of angular velocity to eventually be the same as the velocity of the ball. So it’s not ‘friction’ per se (attraction to zero), but a positive attraction to some larger than zero velocity. So then the question becomes, how fast are virtual particles. Or, I guess, more importantly, what is the momentum of virtual particles?