Does light have a gravitational effect?

[LunarJim]

I asked this in "lighting dark matter" but think I over-complicated the question, so here it is bite-sized...

Stars have mass, and have gravitational effect in spacetime. When their mass in converted to energy (and rightly or wrongly I consider mass just to be extremely concentrated energy), does the electromagnetic radiation then exhibit no gravitational pull, albeit tiny per unit energy?

If you consider all the - let's call it light, but I unclude other forms of EM - which has been produced from stellar mass over billions of years, there would be enormous amounts of energy ambient in the Universe. I.e. where has all the mass of all those stars over all that time gone?

If the gravitational pull of mass is lost when it's transferred to energy in stellar reactions, then would the Universe have less net gravitational force over time? (there is far less mass in stars now than there would have been 11 billion years ago).

If the gravitational pull is not lost but made much much harder to measure, wouldn't there then be a significant "background" gravitation attributable to all that lost stellar mass which is now ambient as EM radiation?

[mugaliens]

Yes. Unbelievably and inconsequentially small, but yes.

[LunarJim]

So... given that the amount of energy per unit mass is extremely large, as a multiple of c, would the gravitational pull of the EM radiation (and any other particles) produced through fusion EQUAL the amount of gravitational pull from the source fuel mass?

For example, if this were the case, our Sun burns up around 1 Earth Mass of fuel every 70,000 years or so. It is 4.6 billion years old. Although I'm sure it hasn't burned fuel at the same rate for all this time, that would mean roughly 66,000 Earth Masses of EM radiation out there, the vast majority of which will be radiant in space given the relatively tiny space that absoprtive objects such as planets and dust take of its radiative sphere.

Multiply that process up by all the billions of stars (and other sources of EM and other radiation), and there is surely a tremendous amount of radiative energy travelling through space. Multiply that up to include the first phase stars in the early Universe... would even a miniscule per-photon gravitational effect be extremely significant on this scale?

And would this not give an additional Universal gravitational effect which is relatively uniform?

[IsaacKuo]

So... given that the amount of energy per unit mass is extremely large, as a multiple of c, would the gravitational pull of the EM radiation (and any other particles) produced through fusion EQUAL the amount of gravitational pull from the source fuel mass?

No. In order to do that, stellar fusion would need to consume 50% of the rest mass of the initial fusion fuel. It is not.

[LunarJim]

So some fuel is left behind in the form of other massive particles, and a proportion of mass (which has gravitational effect) is converted into EM and other radiation?

Is the amount of total mass loss to EM radiation still not significant over long periods of time? By significant, I mean there would still be the equivalent of many solar masses radiating across spacetime?

[IsaacKuo]

One crude way to look at stellar fusion is that it takes some protons and electrons and converts them into neutrons. To balance things out, photons and neutrinos are emitted, but these are much less massive than protons/neutrons.

[LunarJim]

They are much less massive per unit yes, but emitted in large quantities. The power output of the sun in clearly substantial, and we only receive a tiny slice of it's spherical output. The sun is also clearly transferring significant mass into energy... my point is that if this energy retains similar overall gravitational effect to the mass which created it, the gravitational effect which was localised at the star is instead dissipated into space.

I have no idea whether this effect is taken into account when people are working out where all this "Missing matter" is etc. but I've never seen it referred to.

All the EM radiation which has been emitted since the BB and is in transit in space, must surely have a wide area, almost "background" gravitational effect because there's so damn much of it after 11+ billion years?

[Kwalish Kid]

All the EM radiation which has been emitted since the BB and is in transit in space, must surely have a wide area, almost "background" gravitational effect because there's so damn much of it after 11+ billion years?

Because of the redshift introduced into light, the energy density of light falls off faster than matter in an expanding universe. Both matter and light get "thinner" in space over cosmological time because the same amount of each occupies a bigger volume at later times. But since redshift decreases light energy as well, it's gravitational impact is much less today relative to matter than it was 13 billion years ago. Back then, gravitational energy from photons dominated, but today it hardly matters.

[IsaacKuo]

The power output of the sun in clearly substantial,

The mass of the sun is much more substantial.

my point is that if this energy retains similar overall gravitational effect to the mass which created it, the gravitational effect which was localised at the star is instead dissipated into space.

Sure, but the amount of energy converted is only a small fraction of the stellar mass which created it.

I have no idea whether this effect is taken into account when people are working out where all this "Missing matter" is etc. but I've never seen it referred to.

It is taken into account.

All the EM radiation which has been emitted since the BB and is in transit in space, must surely have a wide area, almost "background" gravitational effect because there's so damn much of it after 11+ billion years?

There isn't much of it, compared to the mass of the normal visible stars and gas.

You seem to be stuck on the idea that light has been shining for 11 billion years. You're not seeing that there's a hard upper limit on how much energy is available from fusion power. Think of the total energy available from fusion power as water in a large water tank. This water tank has a small spigot at the bottom, releasing some water. Depending on the size of this spigot, the water might be released over a few minutes, or a few hours, or a billion years, or a trillion years...but regardless of how quickly or slowly it's released there's only so much total water to be released.

That's how it is with fusion energy. The universe starts off with some amount of hydrogen (and other atoms). As stellar fusion converts this hydrogen into helium and other elements, fusion energy is released. This total amount is only a fraction of the mass of the original hydrogen, so the EM radiation is NEVER going to have similar mass to the remaining atomic matter. As it is, the universe is young enough that most of the original hydrogen is still around--this vast potential of energy has barely even been tapped yet.

[IsaacKuo]

Because of the redshift introduced into light, the energy density of light falls off faster than matter in an expanding universe. Both matter and light get "thinner" in space over cosmological time because the same amount of each occupies a bigger volume at later times. But since redshift decreases light energy as well, it's gravitational impact is much less today relative to matter than it was 13 billion years ago. Back then, gravitational energy from photons dominated, but today it hardly matters.

Note that these photons were not emitted from fusion reactions.

[LunarJim]

Isaac thanks for the explanations - I understand that the rate of release of EM radiation from fusion reactions is limited. My point is that, if you take our Sun, how much mass has it transferred to energy over its lifetime so far? Even if the majority of mass is converted to other elements rather than released as EM radiation, it would surely be significant? All the EM energy that has ever been released by our Sun (or by Quasars, Supernovae etc), minus the energy which is absorbed by interacting with other atomic matter in its transit, is still out in spacetime.

I know the relative amounts of gravitational force per unit are tiny, but I guess at every place in the observable Universe, EM radiation is present. You can still see stars. And the Universe is clearly BIG.

At every point in normal space (aside from black holes etc.) you are bombarded with radiation from a gazillion stars and other energetic events.

However, if as you say the full effect of this gravitational effect of current and historical EM radiation is already taken into account, I guess I haven't changed the world (again)!

[grant hutchison]

Maybe one way to get a handle on this is to consider the following:
1) Stars are responsible for most of the conversion of mass to energy in the Universe. (I'm slightly nervous about that categorical declaration, but I think I'm correct, so far as we know at present.)
2) Most of the energy from the stellar fusion chain is released at the first step, from hydrogen to helium.
3) The H->He step converts 4.032 atomic mass units to 4.003 atomic mass units, producing a mass deficit of just 0.7%.
4) There is a great amount of hydrogen out there that has never participates in this reaction, because it's either not consumed during a star's lifetime or has never been part of a star.

So the photons that are produced in this way necessarily constitute a small fraction of the mass-energy present in the original gas clouds and stars.

Grant Hutchison

[LunarJim]

Thanks, that helps put things into perspective much better.

What about Supernovae, given that we've had maybe one or two "rounds" of them? I'm not referring just to main burn cycles, but ALL EM radiation traversing the Universe. Do gamma rays etc. also have the same (tiny per unit) gravitational effect as photons?

What percentage of a star's mass would be converted to EM radiation in a supernova event?

Thanks again for explaining.

[grant hutchison]

Thanks, that helps put things into perspective much better.

What about Supernovae, given that we've had maybe one or two "rounds" of them? I'm not referring just to main burn cycles, but ALL EM radiation traversing the Universe. Do gamma rays etc. also have the same (tiny per unit) gravitational effect as photons?

What percentage of a star's mass would be converted to EM radiation in a supernova event?

The energy released by a supernova comes from its gravitational potential energy, rather than the mass of its individual particles. Of course, the gravitational potential energy contributes to the mass-energy of the original star, but it's again a small fraction of the total. For instance, the Sun has a gravitational potential energy on the order of [10⁴¹J. Sounds a lot, but its mass of 2x10³⁰kg translates to 2x10⁴⁷J, a million times greater.
During the supernova collapse, some of the released potential energy goes into nucleosynthesis of elements heavier than iron, some into the velocity of the ejected debris, and a lot into neutrinos. So again I think we can conclude that the supernova photons have a trivial mass-energy compared to the components of the original star.

Grant Hutchison

[LunarJim]

I know this appears I'm labouring the point, but massive early stars, lots of supernovae, billions and billions of stars - even a millionth of the solar masses of all supernovae, still traversing space, would again equal significant amounts of background gravitation wouldn't it? And again what about GMBs; hugely energetic?

I'll settle on the point that there is SOME gravitational effect from EM radiation ambient in the Universe, and it's already catered for in calculations

[Murphy]

Hmm, I'm a little confused. It was my understanding that Photons do not have mass, so how can they produce any gravitational field at all?

[LunarJim]

Murphy if you read further up the post (#2), apparently they do but it's teeny tiny. My premise is that an almost unimaginable amount of teeny tiny still equals quite a bit.

[DrWho]

Hmm, I'm a little confused. It was my understanding that Photons do not have mass, so how can they produce any gravitational field at all?

Photons do not have mass but they do carry energy, in proportion to their frequency, so this gives them momentum. Any energy produces gravitation and since photons have momentum and energy they will have gravitation.

[LunarJim]

Thanks DrWho

[loglo]

Going back to the original OP and this statement:- "there is far less mass in stars now than there would have been 11 billion years ago"

Is this correct? I would have thought the opposite given that the dwarf stars are mostly still around and that there is a plentiful supply of gas. I've googled around "star formation rates" and "stellar mass fraction" but can't say for sure though this paper seems to agree:-

between 50% and 75% of the present-day stellar mass density had formed by z ~ 1, but at z ~ 2.7 we find that only 3% - 14% of today's stars were present.

I know present day star formation rates are fairly low on average, does this translate to a decreased mass in stars in the present era though?