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

Why should the expansion of the universe be slowing down?

  1. Apr 15, 2009 #1
    Each galaxy (in avarage) doesnt "feel" gravity towards any specific direction so why should the expansion be slowing down due to gravity? even if the expansion is some sort of force, i dont see why it needs to use energy to overcome gravity, which is at balance.

    i am not sure i used the correct formulization so be crative.
  2. jcsd
  3. Apr 16, 2009 #2


    User Avatar
    Science Advisor
    Gold Member

    As you probably know Mass (more precisely the Stress-Energy Tensor) bends space time. According to the Einstein Field Equations mass tends to curve space time in such a way that the expansion of the universe should be slowing down. The cosmological constant is the term that tends to speed up the expansion of the universe.

    I'm sorry, but I can't give any more detailed answer than this. I'm not so good with this. :P
  4. Apr 16, 2009 #3
    In the newtonian framework. infinite space causes paradox!I think.your argument is absolutely
    right!but,If you change your view a little, still in newtonian framework,you will get an contracting universe!just imaging a sphere in space ,and the outside mass can not affect the physics inside.and you can figure out that the length is contracting! no matter how big your imaginary "sphere" is!this is a paradox!this means newtonian view point should have some thing wrong.
    Einstein gives the answer,it is the spacetime itself expanding!So it is ok to have both no force acting on a galaxy for example ,as you pointed out,and the space between galaxies contracting,like the secondary view of newtonian approach.
  5. Apr 16, 2009 #4
    i dont quite understand the paradox you were talking about. why is it contracting?
  6. Apr 16, 2009 #5
    I'd suggest you do some reading on Edwin Hubble..... and his experimentally observed acceleration of the expansion of the universe....we are destined for a clod, dead, empty end not a big crunch....
  7. Apr 16, 2009 #6
    The best way to understand this is to imagine an empty sphere. Einstein says, if you insert mass, the space is bent and contracted. This means that the sphere will shrink in size. The space is being pulled in towards the object. If you keep inserting mass into the sphere, then the sphere will keep shrinking. The cosmological constant just represents the ability of the sphere to counter this energy and expand outwards.

    Galaxies are so far apart that they don't necessarily attract one another. This is just the inverse square law taking effect.

    If you say that the expansion should be slowing down, then you would have to asssume an initial energy, like a bang from a bullet, but no constant acceleration. If you think of the universe this way, then the energy would dissipate due to the stress caused by the mass in the sphere. This is not what we observe. We observe a constant acceleration, which is growing ever more dominate with distance. This is a constant acceleration, hence the cosmological constant.
  8. Apr 16, 2009 #7
    We have not determined the ultimate fate of the universe yet, we only have possible ends, and the Big Crunch has not been ruled out yet.

    The answer to the end of the universe lies in the value of [tex]\Omega[/tex].

    If it is [tex]\Omega[/tex] is greater than 1 then there will either be a the Big Crunch or space and time will continue to expand if there is enought dark energy to cancel out gravity.

    If [tex]\Omega[/tex] is less than 1 then Heat Death or the Big Rip is possible.
    Last edited: Apr 16, 2009
  9. Apr 16, 2009 #8
    It is however, most unlikely. If the expansion is speeding up, then it is apparent that the dark energy is overpowering the gravitational pull of all of the objects, and it will only continue to surpass the gravitational pull as the distances increase.
  10. Apr 16, 2009 #9
    What about gravitational backreaction to speed up the expansion?
    if dark energy is cosmological constant ,and GR is always right,Ok, ultimate de sitter stage is doomed.
    but the two is not exactly confirmed. if dark energy is dynamical ,there is possibility that index of equation of state of dark energy going back to normal in the future. And I don't quite like talking about phantom in FRW cosmology ,I don't think it is right to apply phantom and GR(for freidmann equation) at the same time. If phantom is confirmed ,then GR must be modified.however,I don't know if current observation confirmed the violation of null energy condition.Data fitting toady from supernova or sth just tell the story that the fitter want it to be,in some subtle extend,personally,I think.
    Last edited: Apr 16, 2009
  11. Apr 16, 2009 #10
    you'd better refer to a text book for newtonian viewpoint 2.
    your point,which i call it newtonian viewpoint 1, can not be found in any book since it is a little heresy , but I agree with you.
    you can refer to this one for newtonian contracting in infinite flat space,which is newtonian viewpoint2 I mentioned.
    Liddle A - Introduction To Modern Cosmology.
  12. Apr 17, 2009 #11
    I have to say, there is much of this I am not familiar with.
    "the two is not exactly confirmed" Dark energy and General Relativity are pretty darn accurate. Are you suggesting that Dark energy could be something besides a constant? I don't know exactly what you mean here.

    I am pretty confident that the theory of gravitational backreaction and dark energy being dynamical is just a way to save string theory, i.e. in compactified D brane cases. If not, I don't know exactly what you're talking about.
  13. Apr 17, 2009 #12
    None of this makes sense to me. First, you never state if there is mass in your imaginary sphere, although I assume there is. Also, what mass is outside of this sphere?

    In Newtonian gravitation, nowhere can you ever figure out why two objects attract each other due to their mass. The force is assumed to be the hand of God. So, how does a sphere shrink? We know it does in Einstein's Relativity, because the space is drawn inward, but in Newtonian Gravitation, there is no explanation to justify that. You are just given that two masses will be attracted. This doesn't, in Newtonian Gravity, have any effect on the space itself, and therefore, the size of the sphere. The sphere would be unchanged in length, but the masses inside would attract.
  14. Apr 17, 2009 #13
    My take on slowing has changed over the years - I am now of the persuasion that gravity causes matter to clump - but it doesn't directly affect the rate at which global space is changing - prior to the confirmation of expansion, Friedmann reasoned along the lines that gravity would act upon the expansion rate the same way it acts upon particles - but there was never decent data to empirically support the proposition. It seems that the universe may have critical density, but perhaps this is consequent to something not yet understood
  15. Apr 17, 2009 #14
    sorry for my ambigious statement.
    GR is tested only in our solar system,and it's validity all over universe is not sure. Dark energy is confirmed,but Lamda CDM model is not!
    that's the "two".
    Yes, exactly,dark energy could be many many many things beside a cosmological constant.

    yes,you don't know exactly what dynamical dark energy and gravitational backreaction is.
    for the first, refer to "dynamics of dark energy" ,EJ Copeland, M Sami, S Tsujikawa - Arxiv preprint hep-th/0603057, 2006 ,
    for the latter, search for brandenberger's papers on it.
  16. Apr 17, 2009 #15
    sorry I'm non english native speaker.May be may presentation makes you confuse.
    But what I'm talking about(the sphere thing) is typical text book stuff.I would not explain text books here.
  17. Apr 17, 2009 #16
    just want to be sure, when you say acceleration you mean velocity right? i mean, the farther something is the faster it is (if at rest)?
  18. Apr 17, 2009 #17


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Be careful. When people talk about accelerated expansion they aren't talking about anything moving thru space.

    It refers to the scalefactor a(t). Look up "Friedman equations" in Wiki. The scalefactor is a timevarying parameter that goes into computing spatial distances and it increases with time---governed by the Fr. eqns.

    It doesn't say anything is moving. a(t) increasing simply means distances are increasing. That's to be expected---geometry is dynamic.
    The current rate of increase is a'(t) and the rate of increase of the rate of increase is a"(t). You OK with Freshman calculus notation for first and second derivatives?
    The Fr.eqns. show you how the first derivative and the second derivative of a(t) are determined. They are derived from the main Einstein eqn of GR, but represent a big simplification.

    All that people mean by saying accelerated expansion is that the second deriv. of the scalefactor, namely a"(t), is positive. They're not talking about the Hubble law thing that larger distances increase at a higher rate (that's just a simple proportionality, not what is meant by acceleration.)

    The reason why the rate of expansion a'(t) would be decreasing (except for the dark energy term Lambda) is obvious from looking at the first Friedman equation, the one that determines a"(t). It is clear that a" would be negative except for the Lambda term. Look at the equation in Wiki.

    It is generally a bad idea to try to think about this in Newtonian terms----don't picture galaxies moving, having momentum, being slowed down by some force. What you said about balance is exactly right. To a good approximation all the galaxies are at rest with respect to CMB, and no galaxy feels a force in any direction. The pull in all directions is approximately balanced.

    So Newtonian (motion-type) mechanics is a bad approach. What you need is Friedman mechanics :biggrin: Look at those equations, determining the evolution of a(t) the scalefactor. Instead of trying to think in terms of Newton's equations.

    There is a phony derivation of Friedman model from Newtonian, but it is misleading: gives a false sense of comprehension. And makes people have trouble with the coordinate system that mainstream cosmology uses, and with the fact that distances typically increase faster than c, and (like you said) the approximate balance of gravity pull, and the idea of CMB rest---the Newton pseudo-explanation is rife with potential contradictions so it's better avoided.

    This is good thinking!
    Look at the second Fr. equation here
    The one that determines a"(t)/a(t), do you see it?
    Do you see that it shows that a" must be negative unless there is a positive Lambda term?
    Thats why (without Lambda) expansion rate would have to be declining---the answer to your thread title question.

    And why are the equations trusted? Because they are derived from the Einstein eqn of GR, in the way the Wiki article says. And the GR equation is our theory of how geometry evolves. That is, our theory of how geometry/gravity behaves. It has been repeatedly tested and it is exquisitely precise. There isn't anything more basic than geometry, so there isn't anything more basic than GR.

    Like, GR is the reason that the angles of a triangle add up to nearly 180 degrees (except for very small gravity effects that Euclid didn't know about.)
    It is the reason that geometry is very nearly like what you learned in High School. GR is the reason why parallel lines almost never meet, and why Pythagoras. It is the reason for (approximate) flatness. And why the earth presses on the soles of your feet.

    It has passed all the experimental/observational tests so far and as soon as they find some discrepancy they will refine it and get GR version 1.1. But that hasn't happened yet. So for the time being we take it as right. (And Newton as not right).
    Last edited: Apr 17, 2009
  19. Apr 17, 2009 #18
    thank you very much!
    it took me a while to realize that P isnt momentum :S but i think i managed to understand some of it :P (although i still dont see where it shows that a" must be negative unless there is a positive Lambda term)

    i still dont get why (the favorite question) the gravity affects a".i mean, i can see how it is so by the equations but i dont understand the physics behind it. (like, you can explain why a ball jumps back from the floor by calculating momentum but to realy understand you need to look at the forces and stuff :P)
  20. Apr 17, 2009 #19


    User Avatar
    Science Advisor

    I hope I don't too much disturb marcus's educational efforts, but "understanding physics" quite often means "reducing to known concepts". In this case, I disagree with marcus about the usefulness of Newtonian mechanics: as long as you restrict yourself to a small patch of space (some hundred millions of lightyears, actually), the effect is quite accurately described by the gravitational influence of a homogeneous distribution of matter, where the density of DE is multiplied by -2, twice as repellant as ordinary matter would be attacting. Add "repellant matter" to your mental picture, use Newtonian mechanics, and you do fine in your neighbourhood.
    But keep in mind that it looks all the same in the neighbourhood of every point in space. GR has the complete picture, but locally, Newton is ok.
  21. Apr 17, 2009 #20


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    Yes in that version of the equations, p is pressure associated with matter (not dark energy, as long as the Lambda term is there), and rho is the matter density. Matter pressure and density must be positive.

    so if you put a minus sign in front of the rho and p term, it must be negative.

    So the way they are writing it at first----the first two equations after giving the formula for the metric, the ratio a"/a is equal to a term which must be negative, plus the Lambda term.

    That is why a" must be negative unless you have positive Lamda. Please ask again if you don't see this. I have explained it only very quickly and it may not be clear.

    Let's look again at the second Fr. eqn. sometimes called the Friedman acceleration equation. Let's assume Lambda is zero. We have to show that a" is negative. (expansion rate is declining).

    Just as a matter of taste I like to bring the c^2 out into the coefficient and interpret rho as an energy density. Remember now Lambda = 0.

    a"/a = -(4piG/3c^2) (rho + 3p)

    Since energy density and pressure have the same units, I like to write it this way because then
    rho + 3p makes a nice package expressing the source of gravity (did you know that the pressure at the core of the sun contributes to its gravity?)

    Anyway isn't it clear from this that a"/a is negative? If there is any matter in the universe at all!
    Last edited: Apr 17, 2009
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook