Cosmological Solutions to Einstein's Equations

part of the spacelike submanifold of an AdS3

A metric for Anti-deSitter (AdS) Spacetime in n > 2 dimensions is [Hawking]

ds2 = dr2 + sinh(r)2 n-22 - cosh(r)2 dt2
There is a degeneracy singularity at r = 0. The nonzero Christoffel Symbols for n = 3 and 4 are (with cr = cosh(r))

Γ rr rcr / (1 - cr2) cr / (1 - cr2)
Γ rc c cr (1 - cr2) / (1 - c2)
Γ rφ φcr (1 - cr2) (1 - c2) cr (1 - cr2)
Γ rt t- cr (1 - cr2) - cr (1 - cr2)
Γ cr c - cr / (1 - cr2)
Γ cc c c / (1 - c2)
Γ cφ φ c (1 - c2)
Γ φr φ- cr / (1 - cr2) - cr / (1 - cr2)
Γ φc φ- c / (1 - c2)
Γ tr t1 / cr1 / cr

The r direction is geodesic, and the r-t and spacelike hypersurfaces are geodesic. The latter indicates that the spacetime is globally hyperbolic, and can be foliated into timelike hypersurfaces. Note also that the φ direction is geodesic when c is +-1.

The nonzero components of the Riemann Tensor, which are all negative definite, for n = 3 and 4 are

R r c r c - 1 / (1 - c2)
R r φ r φ- 1 - (1 - c2)
R r t r t- cr2 / (1 - cr2)- cr2 / (1 - cr2)
R c φ c φ - (1 - cr2)2
R c t c t- cr2 (1 - cr2) / (1 - c2)
R φ t φ t- cr2 (1 - cr2)- (1 - c2) cr2 (1 - cr2)

and the Ricci Tensor is

Ra b = - (n - 1) ga b
which implies that AdSn is an Einstein Manifold with negative cosmological constant
Λ = - (n - 1) (n / 2 - 1)
The invariants we have chosen to examine are

R- n (n - 1)
Ra b Ra bn (n - 1)2
Ra b c d Ra b c d2 n (n - 1)
Ra b Rca Rb c- n (n - 1)3
Ra b c d Ra c Rb d- n (n - 1)3
Ra b c d Rea Rb c d e- n (n - 1)2
Ra b c d Re fa b Rc d e f- 4 n (n - 1)
Ra b c d Reafc Rb e d f- n (n - 1) (n - 2)
Ra b c d Reafc Rb f d e- n (n - 1) (n - 3)
Ra b; c Ra b; c0
Ra b; c Ra c; b0
Ra b; a Rcb; c0
Ra b c d; e Ra b c d; e0
Ra b c d; a Reb c d; e0
Euler class - (-1)n / 2 2 / (Area (Sn))
εa b c i ... εe f gi ... Rb c eh Rf g a h- n (2 n - 4) (2 n - 2)

Except for the last, these are all exactly the negative of those for the sphere with unit radius. Hence we recognize AdSn as a spacetime of constant negative curvature. The last invariant is equal to that of the unit sphere.

A metric for deSitter (dS) Spacetime in n > 2 dimensions is [Hawking]

ds2 = a2 cosh(t / a)2 n-12 - dt2
Hypersurfaces of constant time in the deSitter spacetime are spheres. The nonzero Christoffel Symbols for n = 3 and 4 are (with ct = cosh(t / a))

Γ11 1c1 / (1 - c12) c1 / (1 - c12)
Γ11 t1 / ct1 / ct
Γ12 2 c1 (1 - c12) / (1 - c22)
Γ1φ φc1 (1 - c12) c1 (1 - c12) (1 - c22)
Γ21 2 - c1 / (1 - c12)
Γ22 2c2 / (1 - c22)
Γ22 t1 / ct
Γ2φ φc2 (1 - c22)
Γ φ1 φ- c1 / (1 - c12) - c1 / (1 - c12)
Γ φ2 φ- c2 / (1 - c22)
Γ φφ t1 / ct1 / ct
Γ t1 1- ct (1 - ct2) / (1 - c12)- ct (1 - ct2) / (1 - c12)
Γ t2 2- (1 - c12) ct (1 - ct2) / (1 - c22)
Γ tφ φ- (1 - c12) ct (1 - ct2) - (1 - c12) (1 - c22) ct (1 - ct2)
Γ tt tct / (1 - ct2) ct / (1 - ct2)

The timelike direction is geodesic, and the timelike hypersurfaces orthogonal to the φ direction are geodesic. Also, the c1 direction is geodesic when ct is 1, corresponding to the minimum at t = 0 when the space is a sphere of radius a. The c2 direction is geodesic when c1 is +-1, the φ direction is geodesic when cn-2 is +-1, etc.

The nonzero components of the Riemann Tensor, which are all positive definite, for n = 3 and 4 are

R1 2 1 2 a2 ct4 / (1 - c22)
R1 φ 1 φ a2 ct4 a2 (1 - c22) ct4
R1 t 1 ta2 ct2 / ((1 - c12) (1 - ct2))a2 ct2 / ((1 - c12) (1 - ct2))
R2 φ 2 φ a2 (1 - c12)2 ct4
R2 t 2 ta2 (1 - c12) ct2 / (1 - ct2)a2 (1 - c12) ct2 / ((1 - c22) (1 - ct2))
R φ t φ ta2 (1 - c12) (1 - c22) ct2 / (1 - ct2)

The Ricci tensor and all computed curvature invariants are the same as for Sn with r -> a.


FRW Cosmologies

A general metric for the Friedmann-Robertson-Walker cosmologies is [Hawking]
ds2 = S2(t) (- dt2 + dr2 + f2(r) 2)
The spacelike hypersurface for this metric is a space of constant curvature, characterized by a constant K, which determines the functions S(t) and f(r) (and the notation we will use in the following):
K =10-1
S(t) =- E (1 - cos(t)) / 3t2 E (cosh(t) - 1) / 3
f(r) =sin (r)rsinh (r)
cr =cos (r)cosh (r)
ct =cos (t)cosh (t)

where E is the constant energy (kinetic + potential) of the homogeneous dust filling the universe.

These metrics all have a degeneracy singularity at t=0 besides the obvious one at r=0. We will take the coordinate chart for t to be the interval (0, π) for the K=1 case, and (0, infinity) for K=0 or -1. We can use the following Mathematica code to generate the metrics:
ftab = {Sin[r], r, Sinh[r]};

Stab = {(-e / 3) (1 - Cos[t]), t^2, (e / 3) (Cosh[t] - 1)};

rtab = {cr, r, cr};

ttab = {ct, t, ct};

crtab = {ArcCos[cr], r, ArcCosh[cr]};

cttab = {ArcCos[ct], t, ArcCosh[ct]};

xp = Flatten[ {r, Table[ Symbol["t" <> ToString[i]], {i,dim - 3}], ph, t}];

gp = IdentityMatrix[dim];

gp[[dim, dim]] = -1;

fn = ftab[[K]];

Sfn = Stab[[K]];

Do[

gp[[j, j]] = fn^2 Product[ Sin[xp[[i]]]^2, {i, 2, j - 1}],

{j, 2, dim - 1}];

Do[
gp[[i, i]] = Sfn^2 * gp[[i, i]],

{i, dim}];

gp = simpler[gp];

x = Flatten[ {rtab[[K]], Table[ Symbol["c" <> ToString[i]], {i, dim - 3}], ph, ttab[[K]]}];

g = simpler[coordxform[gp, xp, Flatten[ {crtab[[K]], Table[ ArcCos[ Symbol["c" <> ToString[i]]],

{i, dim - 3}], ph, cttab[[K]]}], x]];
In 4 dimensions (henceforth notated D = 4) this yields the following metrics:

K =10-1
g r r(1 - ct)2 E2 / (9 (1 - cr2)) t2 (ct - 1)2 E2 / (9 (cr2 - 1))
g c c(1 - cr2) (1 - ct)2 E2 / (9 (1 - c2)) r2 t2 / (1 - c2) (cr2 - 1) (ct - 1)2 E2 / (9 (1 - c2))
g φ φ(1 - c2) (1 - cr2) (1 - ct)2 E2 / 9 r2 (1 - c2) t2 (1 - c2) (cr2 - 1) (ct - 1)2 E2 / 9
g t t- (1 - ct) E2 / (9 (1 + ct)) - t2 - (ct - 1) E2 / (9 (1 + ct))

The nonzero Christoffel Symbols in D = 4 are

K =10-1
Γ rr rcr / (1 - cr2) - cr / (cr2 - 1)
Γ rr t- 1 / (1 - ct) 2 / t 1 / (ct - 1)
Γ rc ccr (1 - cr2) / (1 - c2) - r / (1 - c2) - cr (cr2 - 1) / (1 - c2)
Γ rφ φ(1 - c2) cr (1 - cr2) - (1 - c2) r - (1 - c2) cr (cr2 - 1)
Γ cr c- cr / (1 - cr2) 1 / r cr / (cr2 - 1)
Γ cc cc / (1 - c2) c / (1 - c2) c / (1 - c2)
Γ cc t- 1 / (1 - ct) 2 / t 1 / (ct - 1)
Γ cφ φc (1 - c2) c (1 - c2) c (1 - c2)
Γ φr φ- cr / (1 - cr2) 1 / r cr / (cr2 - 1)
Γ φc φ- c / (1 - c2) - c / (1 - c2) - c / (1 - c2)
Γ φφ t- 1 / (1 - ct) 2 / t 1 / (ct - 1)
Γ tr r- (1 + ct) / (1 - cr2) 2 / t (1 + ct) / (cr2 - 1)
Γ tc c- (1 - cr2) (1 + ct) / (1 - c2) 2 r2 / ((1 - c2) t) (cr2 - 1) (1 + ct) / (1 - c2)
Γ tφ φ- (1 - c2) (1 - cr2) (1 + ct) 2 (1 - c2) r2 / t (1 - c2) (cr2 - 1) (1 + ct)
Γ tt t- 1 / (1 - ct2) 2 / t 1 / (ct2 - 1)

The timelike direction is geodesic, and the r-t hypersurface and the timelike hypersurface orthogonal to φ are both geodesic.

The nonzero components of the Riemann Tensor are all positive definite:

K =10-1
R r c r c2 (1 - ct) E2 / (9 (1 - c2)) 4 r2 t2 / (1 - c2) 2 (ct - 1) E2 / (9 (1 - c2))
R r φ r φ2 (1 - c2) (1 - ct) E2 / 9 4 (1 - c2) r2 t2 2 (1 - c2) (ct - 1) E2 / 9
R r t r tE2 / (9 (1 - cr2) (1 + ct)) 2 t2 E2 / (9 (cr2 - 1) (1 + ct))
R c φ c φ2 (1 - cr2)2 (1 - ct) E2 / 9 4 r4 t2 2 (cr2 - 1)2 (ct - 1) E2 / 9
R c t c t(1 - cr2) E2 / (9 (1 - c2) (1 + ct)) 2 r2 t2 / (1 - c2) (cr2 - 1) E2 / (9 (1 - c2) (1 + ct))
R φ t φ t(1 - c2) (1 - cr2) E2 / (9 (1 + ct)) 2 (1 - c2) r2 t2 (1 - c2) (cr2 - 1) E2 / (9 (1 + ct))

as are the nonzero components of the Ricci Tensor:

K =10-1
R r r3 / ((1 - cr2) (1 - ct)) 6 / t2 3 / ((cr2 - 1) (ct - 1))
R c c3 (1 - cr2) / ((1 - c2) (1 - ct)) 6 r2 / ((1 - c2) t2) 3 (cr2 - 1) / ((1 - c2) (ct - 1))
R φ φ3 (1 - c2) (1 - cr2) / (1 - ct) 6 (1 - c2) r2 / t2 3 (1 - c2) (cr2 - 1) / (ct - 1)
R t t3 / ((1 - ct)2 (1 + ct)) 6 / t2 3 / ((ct - 1)2 (1 + ct))

The invariants we have chosen to examine are

K =10-1
R54 / ((1 - ct)3 E2) 12 / t6 54 / ((ct - 1)3 E2)
Ra b Ra b2916 / ((1 - ct)6 E4) 144 / t12 2916 / ((ct - 1)6 E4)
Ra b c d Ra b c d4860 / ((1 - ct)6 E4) 240 / t12 4860 / ((ct - 1)6 E4)
Ra b Rca Rb c39366 / ((1 - ct)9 E6) 432 / t18 39366 / ((ct - 1)9 E6)
Ra b c d Ra c Rb d118098 / ((1 - ct)9 E6) 1296 / t18 118098 / ((ct - 1)9 E6)
Ra b c d Rea Rb c d e52488 / ((1 - ct)9 E6) 576 / t18 52488 / ((ct - 1)9 E6)
Ra b c d Re fa b Rc d e f122472 / ((1 - ct)9 E6) 1344 / t18 122472 / ((ct - 1)9 E6)
Ra b c d Reafc Rb e d f61236 / ((1 - ct)9 E6) 672 / t18 61236 / ((ct - 1)9 E6)
Ra b c d Reafc Rb f d e30618 / ((1 - ct)9 E6) 336 / t18 30618 / ((ct - 1)9 E6)
Ra b; c Ra b; c- 393660 (1 + ct) / ((1 - ct)9 E6) - 8640 / t18 - 393660 (1 + ct) / ((ct - 1)9 E6)
Ra b; c Ra c; b- 373977 (1 + ct) / ((1 - ct)9 E6) - 8208 / t18 - 373977 (1 + ct) / ((ct - 1)9 E6)
Ra b; a Rcb; c- 59049 (1 + ct) / ((1 - ct)9 E6) - 1296 / t18 - 59049 (1 + ct) / ((ct - 1)9 E6)
Ra b c d; e Ra b c d; e- 708588 (1 + ct) / ((1 - ct)9 E6) - 15552 / t18 - 708588 (1 + ct) / ((ct - 1)9 E6)
Ra b c d; a Reb c d; e- 39366 (1 + ct) / ((1 - ct)9 E6) - 864 / t18 - 39366 (1 + ct) / ((ct - 1)9 E6)
Euler class243 / (2 π2(1 - ct)6 E4) 6 / (π2t12) 243 / (2 π2(ct - 1)6 E4)
εa b c i ... εe f gi ... Rb c eh Rf g a h- 1944 / ((1 - ct)6 E4) - 96 / t12 - 1944 / ((ct - 1)6 E4)

We see that for constant t, these spaces are indeed spaces of constant curvature, and that the singularity at t=0 is in fact an essential singularity. Note that the order 0 invariants are all positive, and that once again the last invariant is negative. Also note that that even though the spacelike hypersurfaces are of constant curvature, the order 1 invariants are nonzero: the invariants measure the curvature of the whole spacetime.

In general these invariants contain very little independent information, which as we have seen is indicative of the geometrical simplicity of these spacetimes. The coefficients of the invariants have the following patterns, with those for K = -1 the same as those for K = +1:

K =10
R18 (D - 1) (D - 3) 4 (D - 1) (D - 3)
Ra b c d Ra b c d324 (D - 1) (2 D - 3) 16 (D - 1) (2 D - 3)
Ra b c d Re fa b Rc d e f 64 (D - 1) (4 D - 9)
Ra b c d Reafc Rb e d f 16 (D - 1) (D - 2) (4 D - 9)
Ra b c d Reafc Rb f d e1458 (D - 1) (D - 3) (4 D - 9) 16 (D - 1) (D - 3) (4 D - 9)
Ra b; a Rcb; c- 6561 (D - 1)2 (D - 3)2 - 144 (D - 1)2 (D - 3)2
Ra b c d; e Ra b c d; e- 26244 (D - 1) (4 D - 7) - 576 (D - 1) (4 D - 7)
Ra b c d; a Reb c d; e- 6561 (D - 3)2 (2 D - 2) - 288 (D - 1) (D - 3)2
εa b c i ... εe f gi ... Rb c eh Rf g a h- 324 (D - 1) (D - 2) (4 D - 15) - 32 (D - 1) (D - 2) (4 D - 15)

The next section explores black hole solutions to Einstein's Equations.



©2005, Kenneth R. Koehler. All Rights Reserved. This document may be freely reproduced provided that this copyright notice is included.

Please send comments or suggestions to the author.