PDA

View Full Version : Can light affect Gravity?



DaCaptain
2018-Jun-08, 02:08 PM
Since light is mass less you'd think that gravity wouldn't be able to effect it, but it does. So is the reverse possible?

primummobile
2018-Jun-08, 03:39 PM
Gravity doesn't directly affect light. It curves the path that light travels.

eburacum45
2018-Jun-08, 05:03 PM
Yes, light creates its own gravity, because light is a form of mass-energy. However this effect is tiny, and I don't know which circumstances would cause this phenomenon to have detectable effects..

Ken G
2018-Jun-08, 05:12 PM
Perhaps the most important application of light affecting gravity is in the early universe, when most of the mass-energy in the universe was in the form of light, and the early expansion was being slowed by the gravity from all that light. We now see the residual of that light as the CMB, but expansion has rendered its gravitational influence insignificant now.

George
2018-Jun-08, 07:56 PM
Perhaps the most important application of light affecting gravity is in the early universe, when most of the mass-energy in the universe was in the form of light, and the early expansion was being slowed by the gravity from all that light. We now see the residual of that light as the CMB, but expansion has rendered its gravitational influence insignificant now. I might be the only one interested, due to colorful circumstance, but it wasn't long after the universe shifted from yellow to orange when it all "came together" (Recombination). Of course, I could be wrong...there might be someone else besides me interested. ;)

IR would have had the greatest pulling effect of all the light at Recombo time.

Ken G
2018-Jun-08, 11:15 PM
IR would have had the greatest pulling effect of all the light at Recombo time.
You're right, but even by then the pulling effect was already negligible compared to what the matter was doing. Some time around ultraviolet or even X-ray temps, I believe, is when light's pulling ceased to matter.

Solfe
2018-Jun-09, 03:30 PM
I saw a TV show about a scientist that wanted to bend space-time with a ring of lasers. It was very odd, because he was realistic about it. The result was a computer model showing that you could get enough beams into the right configuration. The effect would be measurable but small. He was very honest about the ability to actual build something like this in space. He quipped that it wouldn't be in his life time or possibly the life time of anyone he knew barring some incredible breakthroughs in dozens of different technologies.

I am 99% certain that Morgan Freeman was on this show, so maybe some salt should be served with that antidote. Personally, I found the candor of the though experiment to be the most convincing part, which speaks more to knowing details than doing science. I love when scientists are shown letting it all hang out and are not especially concerned if they are being silly or wrong. Science is fun. Silly is also fun.

George
2018-Jun-09, 04:54 PM
You're right, but even by then the pulling effect was already negligible compared to what the matter was doing. Some time around ultraviolet or even X-ray temps, I believe, is when light's pulling ceased to matter.
Yep. I wonder what temp. or wavelength peak was the break-even point?

Ken G
2018-Jun-10, 05:26 AM
Yep. I wonder what temp. or wavelength peak was the break-even point?
It's about a hundred thousand K, I forget exactly.

George
2018-Jun-10, 05:05 PM
It's about a hundred thousand K, I forget exactly. Yes, that seems to be in the ballpark. This article (http://www.astro.ucla.edu/~wright/BBhistory.html) states 56,000 years when both densities were equal, and at 9000K, which seems cooler than I would have guessed.

[Added... a 9,000K BB peaks at ~ 410nm (photon flux; 320nm energy flux), so we aren't even out of the visible spectrum (violet and near UV, respectively). It would take 300 million of these peak photons to make one proton, though there is a fair bit more in total photon eV after integrating the distribution. 3E8 is actually less than I would have guessed. There are ~ 1E18 photons/sec that enter the eye if one improperly looks directly at the Sun. This is equivalent to very roughly 100 million protons/sec pounding the retina, if you enjoy both apples and oranges since there are differences in this comparison.]

Ken G
2018-Jun-11, 01:12 AM
Yeah that does seem cool. A good rule of thumb is there are a billion CMB photons for every proton in the universe, and at 9000 K, each photon has about a half an eV of mass-energy. A proton has 1 GeV of mass-energy, so you'd need perhaps two billion times more photons at 9000 K to equal the mass-energy of a proton. That sounds like a factor of 2 shortfall at 9000 K, but it's much worse, because there's about ten times more mass-energy in the dark matter than the protons. So it seems to me you'd need the photons to be at about 20 times 9000 K, so over a hundred thousand K.

George
2018-Jun-11, 02:25 PM
Yeah that does seem cool. A good rule of thumb is there are a billion CMB photons for every proton in the universe, and at 9000 K, each photon has about a half an eV of mass-energy. A proton has 1 GeV of mass-energy, so you'd need perhaps two billion times more photons at 9000 K to equal the mass-energy of a proton. That sounds like a factor of 2 shortfall at 9000 K,...Hmmm, would that be true at say a solar temperature. It seems to me that the average photon for a 9000K BB temp. would be a little over 1 eV, though perhaps my graph is in error:

23396

[The y-scale is if this was a star with an apparent mag. of 1. Don't ask me why? :)]

I picked around 900nm as an average using my kind of integration -- I sliced off everything above the 900nm point and spread this over the rest of it like peanut butter on bread. :) [0.5 eV is near 2500 nm.]


... but it's much worse, because there's about ten times more mass-energy in the dark matter than the protons. So it seems to me you'd need the photons to be at about 20 times 9000 K, so over a hundred thousand K. Could the DE be countering the DM at this point to explain this variance?

Hornblower
2018-Jun-11, 10:02 PM
Since light is mass less you'd think that gravity wouldn't be able to effect it, but it does. So is the reverse possible?

My bold. I would not necessarily think that way. Let us start with Newton's equations.

f =GMm/r^2

If we let m = 0, of course f goes to zero. Now let us solve for the acceleration of particle m.

f = ma, so a = f/m = 0/0. That quotient is indeterminate, not necessarily nonzero. We cannot make a valid prediction from Newton's equations, which break down when m = 0, but we know from observation that photons are affected by gravity. Specifically, they are deflected when going by a massive object, among other things.

DaCaptain
2018-Jun-12, 12:24 AM
My bold. I would not necessarily think that way. Let us start with Newton's equations.

f =GMm/r^2

If we let m = 0, of course f goes to zero. Now let us solve for the acceleration of particle m.

f = ma, so a = f/m = 0/0. That quotient is indeterminate, not necessarily nonzero. We cannot make a valid prediction from Newton's equations, which break down when m = 0, but we know from observation that photons are affected by gravity. Specifically, they are deflected when going by a massive object, among other things.

I see your point. So what’s going on with gravitational lensing?

Hornblower
2018-Jun-12, 12:41 AM
I see your point. So what’s going on with gravitational lensing?

The light is deflected as it goes by a massive object.

DaCaptain
2018-Jun-12, 12:49 AM
The light is deflected as it goes by a massive object.

Isn’t that caused by gravity?

Ken G
2018-Jun-12, 05:05 AM
Hmmm, would that be true at say a solar temperature. It seems to me that the average photon for a 9000K BB temp. would be a little over 1 eV, though perhaps my graph is in error:You could well be right, the average energy does not have to be kT, it could be more than that. So I might be off by a factor of 2-3, but it can't be anywhere near 9000 K, that has to be way too low.

Could the DE be countering the DM at this point to explain this variance?No, the estimate I made is early in the Big Bang, long before dark energy was doing anything important.

Ken G
2018-Jun-12, 05:07 AM
Isn’t that caused by gravity?Yes, he was saying that not only is light affected by gravity, but it is affected in a way that is not so different from the way a particle with almost no mass would be affected. It's not true that the effect should go away as the mass goes to zero, because the mass of a particle has no effect (in Newton's picture) on the motion of a particle under gravity. And when you take the mass to zero in relativity, the motion ends up being only about a factor of 2 different from what you'd expect from Newton's laws. In short, since Newton already regarded light as being made of particles, he would have expected light to be deflected by gravity-- even if someone had told him those particles had zero mass (which probably would have come as a pretty big surprise to Newton). Thus Newton would have expected gravitational lensing of light, if he had the foresight to think about it, and he would have only been wrong by about a factor of 2 after Romer measured the speed of light in 1676. But few thought it would be a measurable effect, even after a few theorists used relativity to calculate it, most astronomers were skeptical it would ever be "a thing." But they did expect the deflection of starlight, that's easier than lensing.

George
2018-Jun-12, 02:07 PM
So I might be off by a factor of 2-3, but it can't be anywhere near 9000 K, that has to be way too low. I found a book excerpt (https://books.google.com/books?id=PVJEAAAAQBAJ&pg=PA538&lpg=PA538&dq=Big+Bang+mass+and+energy+%22density+were+equal% 22&source=bl&ots=E8OqQigpIG&sig=7LZAPGMOdU944ymrPq42Ph13g-s&hl=en&sa=X&ved=0ahUKEwjZ9cLXoM7bAhWE24MKHTVhCr4Q6AEIRDAE#v=on epage&q=Big%20Bang%20mass%20and%20energy%20%22density%20 were%20equal%22&f=false) that supports ~104K. It is surprising.

George
2018-Jun-12, 05:32 PM
My bold. I would not necessarily think that way. Let us start with Newton's equations.

f =GMm/r^2

If we let m = 0, of course f goes to zero. Now let us solve for the acceleration of particle m.

f = ma, so a = f/m = 0/0. That quotient is indeterminate, not necessarily nonzero. We cannot make a valid prediction from Newton's equations, which break down when m = 0, but we know from observation that photons are affected by gravity. Specifically, they are deflected when going by a massive object, among other things.That's helpful. I'd like to extend this a bit by noting that, using these equations, if we used particles with less and less mass, then the pull would be equally less and less since the pull is the product of the two masses. This means that the diverting trajectory acceleration would also be equal with less and less particle mass. So, it does seem to make sense that we should use a lim->0 approach instead of a cold m=0, which would make a huge drop in acceleration that has been flat otherwise.

Ken G
2018-Jun-12, 06:04 PM
I found a book excerpt (https://books.google.com/books?id=PVJEAAAAQBAJ&pg=PA538&lpg=PA538&dq=Big+Bang+mass+and+energy+%22density+were+equal% 22&source=bl&ots=E8OqQigpIG&sig=7LZAPGMOdU944ymrPq42Ph13g-s&hl=en&sa=X&ved=0ahUKEwjZ9cLXoM7bAhWE24MKHTVhCr4Q6AEIRDAE#v=on epage&q=Big%20Bang%20mass%20and%20energy%20%22density%20 were%20equal%22&f=false) that supports ~104K. It is surprising.Yes, the Wiki at https://en.wikipedia.org/wiki/Scale_factor_(cosmology)#Matter-dominated_era says the redshift was 3600 when radiation ceased to dominate, so multiplying that by 2.7 K gives about 10,000 K, so that does seem to be the right answer (though the Wiki says this is 47,000 years of age, not 76,000 years of age, so everything is not fitting together here!). This result is certainly puzzling to me, as I have definitely seen everywhere that there are a billion photons per proton (let's say 2 billion, that may be more accurate), and ten times more dark matter mass than proton mass (let's say 6 times more). So one expects two billion photons per 7 proton masses, so each photon energy should be 7 GeV divided by two billion, or 3.5 eV of energy, when matter mass-energy took over. The average energy is not kT, so you are right about that, but more like 2.8kT, so the T would only be 1.5 eV or something. That would be about 20,000 K, so it's still not 9000 K but it's starting to seem in the ballpark. So some numbers are coming out a bit strange, but 104 K does seem to be right no matter how you do it.

Canadia
2018-Jun-12, 08:24 PM
Since light is mass less you'd think that gravity wouldn't be able to effect it, but it does.

If we just keep things simple and live in a Newtownian universe, large and small objects alike are accelerated the same amount by a gravitational source. For smaller objects, the gravitational force is smaller, but so is the force needed to produce a given acceleration.

George
2018-Jun-12, 09:33 PM
Yes, the Wiki at https://en.wikipedia.org/wiki/Scale_factor_(cosmology)#Matter-dominated_era says the redshift was 3600 when radiation ceased to dominate, so multiplying that by 2.7 K gives about 10,000 K, so that does seem to be the right answer (though the Wiki says this is 47,000 years of age, not 76,000 years of age, so everything is not fitting together here!). The 50,000 year ages may be more accurate. Ned Wright's calculator agrees as well.


This result is certainly puzzling to me, as I have definitely seen everywhere that there are a billion photons per proton... That seems very low since a proton is about 1E9 MeV (1E15eV), so, using a billion to one (which coincidentally matches my odds of getting this level of physics correct) as the ratio, each photon would be 3E6eV (for 7 equivalent proton masses), or did I miss something?


... The average energy is not kT, so you are right about that, but more like 2.8kT, so the T would only be 1.5 eV or something. That would be about 20,000 K, so it's still not 9000 K ... Wouldn't the integration for a 10,000K BB give you that 1.5eV? Here's a 20,000K for comparison:

23400


but it's starting to seem in the ballpark. So some numbers are coming out a bit strange, but 104 K does seem to be right no matter how you do it.Yes, but why does it feel strange to say 10,000K doesn't feel hot enough? ;)

Ken G
2018-Jun-13, 05:30 AM
That seems very low since a proton is about 1E9 MeV (1E15eV), so, using a billion to one (which coincidentally matches my odds of getting this level of physics correct) as the ratio, each photon would be 3E6eV (for 7 equivalent proton masses), or did I miss something?A proton is 1000 MeV.

Wouldn't the integration for a 10,000K BB give you that 1.5eV? Here's a 20,000K for comparison:It seems the average energy is about 2.8 kT, which at 10,000 K is 2 eV. So 3.5 eV seems a bit higher T.


Yes, but why does it feel strange to say 10,000K doesn't feel hot enough? ;)It's pretty hot, that's true!

George
2018-Jun-13, 01:58 PM
A proton is 1000 MeV.It seems the average energy is about 2.8 kT, which at 10,000 K is 2 eV. So 3.5 eV seems a bit higher T.
It's pretty hot, that's true! Yep, that helps. :wall: I should have kept track of where I get my data to see if I can blame someone else. :)