Why photons don't fall to Earth?

[looka]

Some introductory line about me apologizing for my ignorance and inability to find answers on interwebs.


If photons get affected by gravitational field (or follow the curvature of space-time created by nearby massive objects) and have in essence some non-rest mass how come they don't just fall down due gravity? Shouldn't all things fall at same rate regardless of their mass, no matter how small?

I understand that photons are just too fast so it's not really observable, but what about when traveling much slower through a specific medium? What about when reflected back and forth or in circle for longer time period? Shouldn't it start to fall? It is still going through curved space-time and being affected by gravitational field, right?

Thanks in advance.

 

[phyzguy]

Let me answer your question with a question. Since the moon is attracted to the Earth by the Earth's gravity, why doesn't the moon fall to the Earth?

 

[DrDu]

Did they stop to fall to the earth? That would explain why it is so cold today.
Seriously: Photons do fall to the earth, their energy increases while doing so and also the photons passing by Earth get slightly deflected by the gravitational field of the earth. Think about it.

 

[looka]

@phyzguy: well of course because it is moving at very specific constant speed and trajectory. But it would fall if we slowed it down, right? Or keep the speed, but we made it bounce back and forth between some huge walls or force it to some other traveling path e.g. arctic circle at same distance or something like that.

@DrDu: So trapped light can be observed to fall? How about when slowed down? (Does that just means absorption and re-emission is slowed, something gravity does not influence?)

 

[DrDu]

When falling of trapped light should be measurable, then this effect should be seen in the gravitational wave detectors.

 

[looka]

So light gets bent but it does not fall? Does it just get absorbed before? Single photon can't be trapped long enough? Thanks!

 

[DrDu]

As phyzguy was already trying to point out to you, there is no difference between bending and falling. A light beam which would just pass by the Earth if there were no gravitation will fall onto Earth when gravity is taken into account.
Let's calculate how far a photon has to travel between two mirrors to fall a distance d :
The acceleration acting on the photon is 2G, where G =10m/s^2 and the factor 2 is due to relativistic effects.
Hence d=1/2 2G t^2 where t is l/c with l being the distance the light has to travel.
Solving for l yields: $$l=c \sqrt{\frac{d}{G}}$$. So to fall 1 mm, the light has to travel 3000 km hence and forth between the two mirrors.
So you see that it is quite a tiny effect although it should be important to compensate for in a gravity detector experiment.

 

[chrispb]

Why do all of the molecules in the atmosphere not fall? In other words, why can we still breathe?

 

[DaveC426913]

Why do all of the molecules in the atmosphere not fall? In other words, why can we still breathe?

This is a different question,. Please start a new thread.

 

[looka]

Thanks DrDu. So 3000km is a problem for light? I mean it would only take a millisecond, but I guess it can't really "survive" let's say 3 million reflections in mirrors 1m apart? (or 30 million for 10cm apart) without getting absorbed.

But light was successfully slowed down to 10km/s, which is 30000 times slower. Distance would then have to be only 100m to see it fall 1mm. That is 100 reflections, that is certainly possible? I think it would be really impressive to have a photo of laser beam falling.

Either way, you did answer my question so thanks again. I just wish I could see it somehow.

 

[ZapperZ]

Please read the FAQ thread in the General Physics forum. An entry on the curvature of the spacetime lines for light might help. You have a misunderstanding on why light is affected by gravity (it is not because it has a non-zero mass).

Zz.

 

[DaveC426913]

I think it would be really impressive to have a photo of laser beam falling.

Read up on Einsteinian rings as a good example of light being curved by gravity.

 

[pellman]

Why don't photons fall to earth?

Because they travel faster than the http://en.wikipedia.org/wiki/Escape_velocity" of the earth.

A black hole, on the other hand, is a body whose escape velocity is greater than light speed, so photons always "fall back in".

How this result is calculated from General Relativity differs from that in Newtonian theory, but the essence of it is the same.

 

[looka]

Thanks all.
So space-time here on Earth's surface is curved a lot, and things fall not due their mass but due space-time curvature (and of course their relative velocity). So it should still be possible then to trap light (for a few milliseconds at least) so it does not escape but to show some gravity effects.

I appreciate Einstein rings (some real nice imagery on Wiki), but I am kinda having in mind some normal size transparent material surrounded by mirrors (or maybe even just looped) where you would fire a beam into a loop at the top of it and then literally see the beam fall down.

As if you would throw a small rubber ball between two walls, no matter how fast you throw it, it would still just fall down (with a lot of bouncing of course), no matter how much bigger it's velocity is than escape velocity. (Or if you throw at the top of the inside cylinder wall).

Seems to me that it is doable nowdays, but yet unseen (by myself anyway)! :)

 

[PAllen]

Thanks all.
So space-time here on Earth's surface is curved a lot, and things fall not due their mass but due space-time curvature (and of course their relative velocity). So it should still be possible then to trap light (for a few milliseconds at least) so it does not escape but to show some gravity effects.

I appreciate Einstein rings (some real nice imagery on Wiki), but I am kinda having in mind some normal size transparent material surrounded by mirrors (or maybe even just looped) where you would fire a beam into a loop at the top of it and then literally see the beam fall down.

As if you would throw a small rubber ball between two walls, no matter how fast you throw it, it would still just fall down (with a lot of bouncing of course), no matter how much bigger it's velocity is than escape velocity. (Or if you throw at the top of the inside cylinder wall).

Seems to me that it is doable nowdays, but yet unseen (by myself anyway)! :)

Good to learn about order of magnitude calculations. Let's say you have two mirrors 10 meters apart, and a beam going back and forth between them. You want to try to detect photon falling. The time for the beam to cross is about .33 * 10-7 seconds. Amount of fall is of order a*t**2. This works out to about 10-14 meters per trip. No mirror is either perfectly flat or perfectly reflective. To reach a fall the width of a virus would take about 10**8 reflections. No mirror is reflective enough to allow that (all light would be scattered or absorbed in that many reflections). Maybe someone has attempted to measure something like this, but it would seem very hard, even 'today'.

However, this experiment has been done many, many ways astronomically, establishing that light is bent by gravity as predicted by GR.

 

[PAllen]

I found a paper discussing the issues in measuring light deflection on earth. They specifically analyze the issues involved in using 'slow light'. They imply the exeriment is possible, but it hasn't been done yet (so far as I've been able to find).

http://arxiv.org/abs/0801.0060

[EDIT] And here is a paper from 2 year later that agrees the effect is potentially measurable, but disputes the approach of the prior paper (which it references), coming to different conclusions as to the experimental requirements and the quantitative prediction.

http://arxiv.org/abs/0810.4849

 

[phyzguy]

Thanks all.
So space-time here on Earth's surface is curved a lot, and things fall not due their mass but due space-time curvature (and of course their relative velocity). So it should still be possible then to trap light (for a few milliseconds at least) so it does not escape but to show some gravity effects.

I appreciate Einstein rings (some real nice imagery on Wiki), but I am kinda having in mind some normal size transparent material surrounded by mirrors (or maybe even just looped) where you would fire a beam into a loop at the top of it and then literally see the beam fall down.

As if you would throw a small rubber ball between two walls, no matter how fast you throw it, it would still just fall down (with a lot of bouncing of course), no matter how much bigger it's velocity is than escape velocity. (Or if you throw at the top of the inside cylinder wall).

Seems to me that it is doable nowdays, but yet unseen (by myself anyway)! :)

A couple of comments:

(1) If you haven't already, you should look up the Pound-Rebka experiment. Instead of sending the light horizontally, they measured the "deceleration" of the light when propagating ~30 m upward. To carry your bouncing ball analogy further, it is like looking at a ball slowing down as it goes upward, only with light the "deceleration" is in the form of a redshift instead of actually slowing down.

(2) The closest thing that I can think of to what you are talking about is at the LIGO gravitational wave observatory. They have a pair of mirrors like you are talking about ~ 1 km apart. You might try contacting them to see if they have seen (or looked for) this effect, although like PAllen pointed out, it would be very difficult to see.

 

[nickthrop101]

they don't fall to earth, as they have no mass! gravity onnly affects objeects that have mass, so they only affect it a very small amount, the speed that they are going and their mass ( or lack of it) mean they are hardly affected by it.
:)

 

[Fredrik]

they don't fall to earth, as they have no mass! gravity onnly affects objeects that have mass, so they only affect it a very small amount, the speed that they are going and their mass ( or lack of it) mean they are hardly affected by it.
:)

This answer is unfortunately incorrect, especially the claim that gravity only affects objects with mass. The theory we should use to answer the question is general relativity, since there is no gravity in special relativity, and GR says that a) the motion of a massless particle is represented by a special kind of curve in spacetime, called a "null geodesic", and b) the distribution of matter in the universe is a part of what determines which curves are geodesics. So photons are definitely affected by large concentrations of mass like planets and stars. (This has been verified by experiments).

 

[looka]

Good to learn about order of magnitude calculations. Let's say you have two mirrors 10 meters apart, and a beam going back and forth between them. You want to try to detect photon falling. The time for the beam to cross is about .33 * 10-7 seconds. Amount of fall is of order a*t**2. This works out to about 10-14 meters per trip. No mirror is either perfectly flat or perfectly reflective. To reach a fall the width of a virus would take about 10**8 reflections.

Hmm...I'm lost...didnt get the

This works out to about 10-14 meters per trip.

part. And I get different numbers. On Earth for anything to fall 1mm from being still you need a hundredth of a second.

1/1000 m = 10 m/s/s * (1/100 s)²

In a hundredth of second light would go through 3 million meters, which is 300000 reflections in our apparatus. So it's 3*10⁵ reflections for whole 1mm (not 10⁸ for width of virus.)

Also "slow light" will help, allegedly up to 30000 times... so it comes to only 10 reflections to see it fall 1mm! Mirrors can certainly do 10 reflections. :) Am I doing something wrong here?

Thanks for Arxiv link (Deflection of Ultra Slow Light by Earth’s Gravity on
Laboratory Length Scale) that seems to be exactly what I had in mind! And If I understood it right they seem to be able to slow light even more (100m/s?! I used 10000m/s from Wiki) and get fall of 0.1mm after .1m! Theoretically. Up to experimenters I guess.

Maybe I should throw a mail to LIGO as phyzguy proposed, but I won't get my hopes up. :)
Thanks all for great thread.

 

[DrDu]

Slow light is something observed in special materials and non-linear optical settings. The light is so slow because the photons most of the time do not prevail but energy is stored in excited atomic states. So as long as you don't drop the whole experimental set up, slow light will not appear to fall more than normal light.

 

[PAllen]

Slow light is something observed in special materials and non-linear optical settings. The light is so slow because the photons most of the time do not prevail but energy is stored in excited atomic states. So as long as you don't drop the whole experimental set up, slow light will not appear to fall more than normal light.

See the arxiv links I provided earlier. The authors propose how slow light can be used to detect 'falling photons' in laboratory scales, though they disagree on the details.

 

[PAllen]

Hmm...I'm lost...didnt get the part. And I get different numbers. On Earth for anything to fall 1mm from being still you need a hundredth of a second.

1/1000 m = 10 m/s/s * (1/100 s)²

In a hundredth of second light would go through 3 million meters, which is 300000 reflections in our apparatus. So it's 3*10⁵ reflections for whole 1mm (not 10⁸ for width of virus.)

I'm computing fall across 10 meters. My figure is approximately correct. I did err in the following sense: the angle of the light is changing on each reflection, so the deviations don't add linearly, thus my total conclusion is too pessimistic. The main point remains: I don't believe it is possible, even today, to do such an experiment. The paper authors also seem to take it as a given that it is not possible, thus their investigation of slow light.

Thanks for Arxiv link (Deflection of Ultra Slow Light by Earth’s Gravity on
Laboratory Length Scale) that seems to be exactly what I had in mind! And If I understood it right they seem to be able to slow light even more (100m/s?! I used 10000m/s from Wiki) and get fall of 0.1mm after .1m! Theoretically. Up to experimenters I guess.

Please also look at the second paper I gave, which references the first, and claims the first one's results are incorrect. It claims the, while slow light is still useful, the result is much smaller, and the experimental set up more complex, than implied by the first paper. I have no expertise to decide who's right, but you shouldn't blithely assume the first paper is right.

 

[pellman]

they don't fall to earth, as they have no mass! gravity onnly affects objeects that have mass, so they only affect it a very small amount, the speed that they are going and their mass ( or lack of it) mean they are hardly affected by it.
:)

This answer is unfortunately incorrect, especially the claim that gravity only affects objects with mass. The theory we should use to answer the question is general relativity, since there is no gravity in special relativity,

It is incorrect even in Newtonian theory, isn't it? The acceleration of a body due to the gravity of other bodies is independent of its own mass. Feathers fall at the same rate as bowling balls and all. Since the body's mass does not appear in the equation of motion, should be the same even if mass=zero.

 

[DrDu]

No, for photons there is a factor two difference to non-relativistic massive bodies Cf. also my post #7. That's an effect of general relativity.