Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Bending Light= Gravity

  1. Oct 10, 2005 #1
    Ok We All Know That A Gravitational Feild Can Bend A Beam Of Light Now Can The Reverese Of This Be True?

    Can A Bent Beam Of Light Produce A Gravitational Field ?
  2. jcsd
  3. Oct 10, 2005 #2
    Since light is a form of energy, and since energy is equivalent to mass and since mass produces a gravitational field, I would say that all light, bent or otherwise, produces a gravitational field.
  4. Oct 10, 2005 #3
    Does the average beam of light have mass? It has momentum...

    I'd imagine that there would be very strange gravitational phenomena if it is so.
  5. Oct 10, 2005 #4
    On average, no. As I understand it, the standard deviation is 0.
  6. Oct 10, 2005 #5


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    The light has momentum because of its frequency p=hf. In relativity that doesn't imply it has mass.
  7. Oct 10, 2005 #6


    User Avatar
    Staff Emeritus
    Science Advisor

    Light doesn't have mass, but it doesn't need to have mass to contribute to the stress-energy tensor.

    In Newtonian theory, mass causes gravity. In General Relativity, it is not mass, but the stress-energy tensor which causes gravity.

    Because light contributes to the stress-energy tensor, it causes gravity. The issue of whether or not it has mass is irrelevant to this question.

    The contribution of "light' to the gravity of the universe is well accepted in cosmology. In cosmology, we talk of "matter dominated" universes, where most of the gravity of the universe comes from matter, and "radiation dominated universes", where mot of the gravity of the universe comes from radiation, i.e. light.

    Our universe is presently "matter-dominated", but it is expected that at earlier times this was not the case, and that the universe was "energy dominated".

    While it is clear that in general radiation does contributes to the stress-energy tensor, and hence the gravitational field, the exact details of the beam solution aren't known to me. I believe that an approximate solution has been published in a textbook by Tolman, but I haven't seen the text to be able to comment on the approximations that were undoubtedly used. There's more at


    I say undoubtedly because writing down the exact equations for a "beam" of light turns out to be very complicated, even if one does NOT include gravity. (more at the below link).

  8. Oct 10, 2005 #7
    The light doesn't have to be bent to produce a gravitational field. But yes. I can create a field of light.


    My brain just started hemorrhaging. :biggrin: pervect means that a photon (or a beam of light with all its energy flowing in the same direction) has zero of the so-called proper (aka zero rest mass). When others say that light has mass they mean that a photon or beam with all its energy traveling in the same direction has non-zero relativistic mass.

    It is note worthy to let you know that a beam of light can have a zero total momentum and yet produce a gravitational field. In this case the proper mass of the beam will be non-zero.

    See - http://www.geocities.com/physics_world/gr/grav_light.htm

    Last edited: Oct 10, 2005
  9. Oct 13, 2005 #8


    User Avatar
    Science Advisor

    Then consider a potentially dumb question, all the radiation produced by stars since the big bang which has not been "absorved" by any matter, and is supposedly still travelling away in space would account for an enormous amount of relativistic mass. I'm supposing that if were to travel to the edge of the "material universe" after which no more matter could be encountered, and if i were to look back on all the matter i left behind i would see the radiation emitted by stars that has avoided "absorption" meaning that radiation is crossing this edge, never to be absorved at all. How much relativistic matter then is out there still travelling outwards, and how much gravity does it generate?
  10. Oct 13, 2005 #9


    User Avatar
    Science Advisor
    Homework Helper


    At best (or worst, depending on your perspective) the amount of mass would be the cumulative contribution of the mass defect of fusion over the lives of stars, i.e. very small -- and it would be spread over a very large volume!
  11. Oct 13, 2005 #10


    User Avatar
    Staff Emeritus
    Science Advisor

    According to

    http://relativity.livingreviews.org/open?pubNo=lrr-2001-1&page=node6.html [Broken]

    the energy density in radiation is about 1/2,000,000 the energy density in matter, and is dominated by the cosmic microwave bacground radiation.

    Note that in the center of a galaxy, you'll see a lot more visible light than the average observer will, who will be somewhere out in intergalactic space. There's more figures at

    http://arxiv.org/PS_cache/astro-ph/pdf/9912/9912038.pdf [Broken]

    which is referenced by the first link above.

    You can probably actually come up with a number by estimating the volume of the universe (from the Hubble radius), looking at the critical density for flatness, and multiplying volume * density * 1/2,000,000. (Or just compute the volue times the energy density of the CMB directly). Besides being a messy calculation, though, this sort of number doesn't mean much, mainly because calculating the volume of the universe from the Hubble radius isn't quite kosher.
    Last edited by a moderator: May 2, 2017
  12. Oct 13, 2005 #11
    He needs no qualification. The beam of light and the photon have zero mass. The beam gravitates by its stress energy in accordance with general relativity. The Newtonian gravitation by mass equation isn't correct. They deflect in following geodesics in relativity. The Newtonian forced mass dynamics isn't correct.
    Why? That one's yours. It isn't correct.
    Last edited: Oct 13, 2005
  13. Oct 13, 2005 #12


    User Avatar
    Science Advisor

    The light at the edge of the universe which is travelling outwards, is it expanding/stretching space as it goes? In other words, was the big bang like an explosion inside a balloon, the force of which is causing the balloon to stretch? How much energy or matter is necessary to stretch space in this way? Suppose a single photon is farther from the "center" of the universe than anything else, is this single photon, by itself, stretching space as it goes? Are the physics at the edge of the universe the same as elsewhere?
  14. Oct 13, 2005 #13
    Yup. This sure is davy waite that's for sure. Same old ignorant claims and same old nasty temperment and arrogance.

    Go away waite. We don't want you here.
  15. Oct 21, 2005 #14
    I just remembered somesing.

    Have we taken into account ze fact zat any celestial body wiz a gaseous atmoshphere will more zan likely bend light beams by way of optical refraction, as well?

    Just sinking.
  16. Oct 22, 2005 #15


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    And you are going to sink even faster (probably off this board entirely) if you don't stop trying to be "cute".

    To answer your question, optical refraction is subject to chromatic aberration as different light frequencies are refracted at different angles. (this is what allows prisms to break white light up into the spectrum.) The bending of light due to gravity does not suffer from this and all the frequencies of light bend at the same angle.
  17. Oct 23, 2005 #16


    User Avatar
    Science Advisor

    I have also a question, which is quite appropriate in this topic, I think.

    If I have an elektromagnetic field, I have the elektromagnetic field tensor etc, and I can construct with this the elektromagnetic energy-momentum tensor. Now I want to calculate the curvature of space time due to this elektromagnetic field. Can I put the Einstein-tensor now equal to k (that constant, something like G/c4 ) times the enery momentum tensor of the elektromagnetic field, just like you do with ordinary tensors like that of perfect fluids etc? Somehow I have ethical objections to this, I think it's my classical thinking that only mass "produces gravity". I hope someone can answer this question, or confirm my suspicion :) Many thanks in forward.
  18. Oct 23, 2005 #17


    User Avatar
    Staff Emeritus
    Science Advisor

    Yes - that's a perfectly fair application of Einstein's equations. The philosophical point is that it's energy that causes gravity, not mass.
  19. Oct 25, 2005 #18
    True enough. But, at smaller angles of incidence and refraction, a spherical atmosphere would act more like a converging lens than a prism. This means that, in the ‘lens’ region near the optical axis, the light that passes through would not be subject to the dispersion to which you refer. Instead, it creates the exact visual effect that has been attributed to gravity. Of course, focal length comes into play. But if you don’t believe me, look at something through a magnifying glass. And, that effect has nothing to do with either light dispersion, or any chromatic aberration. Simply put, you may not disregard this consideration if you hope to maintain legitimate credibility, in a purely scientific sense.
  20. Oct 25, 2005 #19


    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    Any optical refraction is subject to chromatic aberration it doesn't have to be a prism. If you look through a cheap simple magnifying glass you will see a slight "rainbow" fringe around objects that is the result of chromatic aberration. The reason you do not see this (or at least it isn't as noticeable) with better made lenses is that they are specially made by layering materials of specifically chosen different refractive indices. With the correct choices, the chromatic aberration of one material offsets the aberration of the other. With a spherical atmosphere you are not going to have such specifically chosen materials.
  21. Oct 25, 2005 #20


    User Avatar

    Staff: Mentor

    And people pay lots and lots of money for telescopes that are able to eliminate that aberration.

    AEIII, you may be thinking that since a converging lens makes light rays converge, it can't separate colors. But that isn't what is going on: a converging lens brings parallel rays together and focuses them at a point. Spherical aberration causes light rays of different frequencies to come into focus at different distances from the lens.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Bending Light= Gravity