|
The cosmic balloon analogy is normally used to illustrate a cosmos with a perfectly homogeneous mass distribution. In such a case the balloon is a perfect sphere. It is possible to loosely illustrate inhomogeneities by means of indentations on the balloon. For a given 'time slice' it is however possible to rigorously show the shape of the balloon for a single static black hole embedded into an otherwise homogeneous mass distribution (evenly spread 'cosmic dust').[1]
A Monster
 In order to picture the embedding on a cosmic scale, it is necessary to use a monstrously large black hole. Fig. 1 (right) shows the profile of a cosmic balloon, 'dented' by a single black hole with a mass of 0.01% of the mass of the otherwise homogeneously spread matter, the total mass of the hyperspherical cosmos. (The 'squiggles' on the curves are not inhomogeneities, but just artifacts of the image conversion processes. )
The black circle represents the original homogeneous cosmic balloon with hyperradius R = 100 Gly. The blue curves show the 'dent' caused by this monster black hole. The gravitational effect of such a huge mass concentration will essentially reach around the entire hyperspherical universe.[2]
At the same time, the black hole's central singularity may reach all the way down to the original (hypothetical) singularity of the big bang. The 'may' and 'hypothetical' qualifiers are used because we do not know what happens at any singularity - just that the math of general relativity breaks down there. We require a quantum theory of gravity, which we do not (yet) have. It is nevertheless interesting to speculate that all black holes may be connected to the original big bang singularity...
Be that as it may, the hyperspace vectors of dust particles outside the singularity are a little bit easier to treat - we temporarily 'unfreeze' the situation now, so that the balloon can expand. A homogeneous expanding balloon would have driven free (static) particles outward on radial paths, i.e., with hyperspace vectors normal to the unperturbed (black) surface. It is reasonable to argue that spots on the dented (blue) surface of the expanding balloon will be driven outward with hyperspace vectors that are slanted towards the singularity. This means that free (static) particles at those spots will move closer to the singularity. Gravity 'explained'!  Well, not quite. This is not the full picture of gravity - for that we must also consider the time dimension, which is outside of the scope of this article.[3]
Event Horizons
Singularities do not exist without event horizons, those 'one-way valves' that allow things in but not out of black holes. The event horizon radius of this huge black hole would have been at least 300 million light years, which is almost 1% of the proper radius of our observable universe. This is comparable to the size of superclusters and quite unrealistic, but good for visualization.
 Fig. 2 (right) shows the (almost) vertical 'throat' of the black hole's event horizon. 'Vertical' here actually means going inward radially, because this is still a segment of a hypersphere. The radials converge at the origin (0,0), way down below the segment shown here.
In Cartesian space, the event horizon radius is given by rH = 2GM/c2, where M is the conventional mass of the black hole. In spherical coordinates, the event horizon spans an angle θH = rH/R, where rH is measured along the unperturbed hypersurface and R is the radius of the unperturbed hypersphere (100 Gly for this time slice).[4]
Does this mean that the event horizon radius of a black hole increases due to cosmic expansion? No, not unless matter is 'swallowed' by the hole. For constant black hole mass, rH remains constant, while R increases. This means that θH = rH/R deceases over time. Assuming a perpetually increasing expansion rate, the gravitational effect of a black hole on the global scheme of things will dissipate.
One final, intriguing thought: jumping into a black hole may take you 'back' to the original big bang singularity. Whether that also means being transported back in time is not quite clear. From Einstein's equations, it looks more like you will experience 'imaginary time' (whatever that may mean) until you are first stretched and then crushed...
Jorrie
Notes
[1] The 'given time slice' or 'snapshot' means that for simplicity, we ignore the complication that some of the 'homogeneous cosmic dust' will swirl into the black hole and so create a large void around the hole. We simply place the black hole and then 'freeze' the situation, with the homogeneous conditions still intact around the hole.
[2] If the minimum size of a hyperspherical universe is worked out with Rmin = 100 Gly, the total mass of baryonic plus dark matter comes to M ≥ 1025 Solar masses (Sols). This is just a minimum mass - we do not know how large the 'real cosmos' is. It means that the (0.01% of total mass) black hole must have a mass of at least 1021 Sols - improbable, but still...
[3] I have written a number of Blog articles on black holes, starting at this Blog entry. The effect of time is shown graphically there. More depth can be found in Chapters 4, 5, 6 and 7 of Relativity 4 Engineers. The present Blog article deals only with black holes in relation to expanding space.
[Update: I wrote above: "It is reasonable to argue that spots on the dented (blue) surface of the expanding balloon will be driven outward with hyperspace vectors that are slanted towards the singularity." This does not mean that any 'spot' on the balloon, outside the event horizon, moves closer to the singularity in proper space; it is only the angle θ (see [4] below) that becomes smaller, while spots may remain at the same proper distance, or move away from the BH singularity.]
[4] Some equations used in this article:
The radius of the 'dented' hypersphere (at angle θ from the singularity) and outside the event horizon, can be reworked from 'Gravitation' by Misner, Thorne and Wheeler (MTW), eq. 23.34b, where they give the 'embedding lift' for Schwarzschild space:
z(r) = [8M (r - 2M)]½ + constant ---------- (a)
where MTW's M is a normalized mass, equating to the conventional (SI) mass by GM/c2 (they use geometric units where c=G=1). If we convert this equation to polar coordinates and our usual units, we get:
R' = [8GM/c2 (Rθ - 2GM/c2)]½ + constant ---------- (b)
where R is the 'undented' hyper-radius and the constant is chosen to give R' = R when θ = Π, where ΠR gives the proper spatial radius of the present universe. From present observations, we can deduce that the present R ≥ 100 Gly, but we do not know the actual value.
It is clear from (b) that when Rθ < 2GM/c2, the local hyper-radius R' goes imaginary, giving the event horizon angle in hyperspherical coordinates:
θH = 2GM / Rc2 ---------- (c)
This event horizon angle is obviously relative to the center of the hypersphere, which seems to support the idea that, at least for a hyperspherical universe, the black hole's central singularity resides at the original BB singularity. In a 'flat' or 'open' cosmos, the original singularity was 'everywhere' (as infinite density) and the event horizon of black holes do not point to any specific hyperspace spot, but this is outside of the scope of the cosmic balloon analogy.
 Erratum
While writing this article, I noticed that Figure 1.5 (page 28) of Relativity 4 Engineers contains a typo, as indicated on the right. That r inside the square root of z(r) must be above the line, i.e.:
z(r) = √[8 gtt Mbar r]
where gtt = 1-2Mbar/r and Mbar = GM/c2 in units meters. It is fairly obvious that with r below the line, the curve cannot have a positive slope.
For those readers who have the eBook, you can download an annotated page to patch in, using Adobe Acrobat Professional 7.0 or later (or equivalent). The page contains notes to 'correct' the error (it is a little painful to regenerate the single page from scratch). Otherwise, drop me a CR4 email and I'll send you a link for re-downloading the complete annotated eBook (including the commented page and some extra notes on other pages, clarifying a few things a little better).
-J
| 
Users who posted comments:
G.K. (1), Jorrie (6), Roger Pink (6), StandardsGuy (1)