Solution for Chapter 27

(compiled by Xinkai Wu)

Exercise 27.2 The 3-sphere geometry of a closed universe [Xinkai Wu/02]
(a) We claim that
w = acos, x = asinsincos, y = asinsinsin, z = asincos
As can be veried easily, the above expressions of (w, x, y, z) satisfy w
2
+
x
2
+ y
2
+ z
2
= a
2
, and when we plug them into the 4-d metric ds
2
= dw
2
+
dx
2
+ dy
2
+ dz
2
, the line element reduces to
ds
2
= a
2
[d
2
+ sin
2
(d
2
+ sin
2
d
2
)]
(b) The volume of this 3-sphere is given by
V =
_

0
d
_

0
d
_
2
0
d
_
|det g|
=
_

0
d
_

0
d
_
2
0
d a
3
sin
2
sin
= 2
2
a
3
Exercise 27.3 Energy conservation for a perfect uid [Xinkai Wu/02]
(a) The conservation law of 4-momentum says T

;
= 0. Contracting this
with the 4-velocity of the observer (the uid), we get u

;
= 0.(Its tempting
to put u

inside the covariant derivative and say that we have J

;
= 0 with
the 4-momentum ux measured by the observer being J

= u

. However
this is in general wrong in curved spacetime unless u
;
+ u
;
= 0, namely u
is a Killing vector eld, which is in general not the case.)
(b) Plugging the expression of the uids stress tensor into the conservation
law
0 = u

;
= u

_
( + P)
;
u

+ ( + P)[u

;
+ u

;
] + P
;
g

_
[using u
2
= 1 u

;
= 0]
= u

( + P)
;
( + P)u

;
+ u

P
;
= u

;
( + P)u

;
which is
d
d
= ( + P)

u
1
(c) In the (t, , , ) coordinate system, u = (1, 0, 0, 0). Thus we have
d
d
= u

;
= u
t

t
=

t
=
d
dt
noting that (also a) only depends on t by the homogeneity and isotropy of our
universe. Also we have

u =
1

g
(

g u

)
,
=
1
a
3

2
sin
(a
3

2
sin)
,t
=
1
a
3
da
3
dt
Thus eq. (27.31) becomes
d
dt
= ( + P)
1
a
3
da
3
dt
which can be rewritten as
d(a
3
)
dt
= P
da
3
dt
Exercise 27.5 Einsteins static universe [Xinkai Wu/02]
The Einstein equation with a cosmological constant is
G
ab
= 8T
ab
g
ab
= 8(T
ab

8
g
ab
)
For our homogeneous and isotropic universe, the stress tensor is that of perfect
uid T
ab
= ( + P)u
a
u
b
+ Pg
ab
, and we have
G
ab
= 8
_
( + P)u
a
u
b
+ Pg
ab

8
g
ab
_
= 8
__
+

8
+ P

8
_
u
a
u
b
+
_
P

8
_
g
ab
_
So we can use eq. (27.24) and (27.26) with +

8
and P P

8
.
Eq. (27.24) becomes
d
__
+

8
_
a
3

dt
=
_
P

8
_
da
3
dt

d(a
3
)
dt
= P
da
3
dt
Eq. (27.26) becomes
_
a
a
_
2
+
k
a
2
=
8
3
_
+

8
_
Now we assume the equation of state for the uid P = 0. So eq. (27.24) gives
=
0
_
a0
a
_
3
. Plugging this into eq. (27.26) we get
1
2
a
2
+ V (a) =
k
2
2
with V (a) =
4
3
a
2
_
+

8
_
=
4
3
_
0a
3
0
a
+

8
a
2
_
.
(a) The evolution equation for a is thus
a = V

(a) =
4
3
_

0
a
3
0
a
2
+

4
a
_
Hence to get V

(a
0
) = 0 we must have = 4
0
. And in that case, a
0
= 0. To
get a static universe we also need a
0
= 0, which requires
0 =
k
2
+ V (a
0
) =
k
2

4
3
a
2
0
_

0
+

8
_
=
k
2
2
0
a
2
0
which implies k = +1 and 4
0
a
2
0
= 1.
(b) Today
0
10
31
g/cm
3
, thus = 4
0
1.2 10
30
g/cm
3
, and

=

8
=
1
2

0
0.5 10
31
g/cm
3
.
(c) V

(a
0
) = 4
0
< 0, thus this model is unstable against gravitational
collapse. Also see Fig 1.
0.5 1 1.5 2
-2.75
-2.5
-2.25
-2
-1.75
-1.5
Figure 1: Einsteins static universe: the x-axis is a/a
0
, the y-axis is V (a) (not
normalized)
Exercise 27.6 Cosmological redshift [Xinkai Wu/02]
(a) From the solution to Exercise 24.4(a), we have
dp

d
=

=
1
2
g
,
p

In our case the only nonvanishing components of p are p

t
and p

. So
dp

d
=
1
2
g
,
p

=
1
2
(g
tt,
(p
t
)
2
+ g
,
(p

)
2
) = 0
since g
tt,
= 0 and g
,
= 0. So p

is conserved along the particles worldline.

(b)
m
2
= p
2
= g
tt
(p
t
)
2
+ g

(p

)
2
= (p
t
)
2
+
_
p

a
_
2
3
which gives
p
t
=
_
m
2
+
_
p

a
_
2
also note that E = p

= p
t
=
_
m
2
+
_
p
a
_
2
. Thus we nd
p =
_
E
2
m
2
=
p

a
, p
0
=
_
E
2
0
m
2
=
p

a
0
and thus
p
0
p
=
a
a
0
=
1
1 + z
(c) For a photon we simply set m = 0 and notice that =

p
, then eq.
(27.43) and (27.44) follow from eq. (27.54).
(d) The particles velocity is given by v = p/E. For v << 1, we can approx-
imate E m, thus
v
0
v
=
p
0
p
=
a
a
0
Exercise 27.7 Cosmic microwave radiation in an anisotropic cosmological
model [Xinkai Wu/02]
(a) Take t = const, ds
2
= a
2
(t)dx
2
+ b
2
(t)dy
2
+ c
2
(t)dz
2
. This 3-d met-
ric is positive denite, thus having Euclidean geometry. Since the 3-d met-
ric components a
2
(t), b
2
(t), c
2
(t) are independent of x, y, z, the connection co-
ecients on the const-t slice all vanish, namely its spatially at. Also the
x, y, z-independence of the 3-d metric components means that we have transla-
tion symmetry along the spatial directions, namely its spatially homogeneous.
Consider the observers whose 4-velocities are given by u

= (1, 0, 0, 0) in the
(t, x, y, z) coordinates (i.e. they have constant x, y, z coordinates). Since the
4-d metric is diagonal, u is orthogonal to the const-t slices and thus these slices
are hypersurfaces of simultaneity for these observers. Also u
t
= dt/d = 1 gives
= t.
(b) In the local Lorentz frame of the homogeneous observer
N =
2
h
3
1
e
E/kTe
1
where E is the energy measured by the observer: E = p
t
. Using 0 = p
2
=
g
tt
(p
t
)
2
+ g
xx
(p
x
)
2
+ g
yy
(p
y
)
2
+ g
zz
(p
z
)
2
we nd p
t
and thus E
E =
_
_
p
x
a
e
_
2
+
_
p
y
b
e
_
2
+
_
p
z
c
e
_
2
_
1/2
4
(c) Since the metric has no dependence on either x, y, or z, by Exercise 24.4
(a), p
x
, p
y
, p
z
are all conserved along the photons worldline.
(d) Since we are assuming that theres no collision along the photons tra-
jectory, by Vlasov equation, N is conserved along the worldline of any photon.
Since (27.57) is expressed solely in terms of the conserved p
x
, p
y
, p
z
and T
e
(which is constant by denitioin), N remains precisely the form (27.57) for all
later times.
(e) The basis vetors of the local Lorentz frame of a homogeneous observer
at time t
0
is given by
e

0
= e
0
, e
x
=
1
a
0
e
x
, e
y
=
1
b
0
e
y
, e
z
=
1
c
0
e
z
(f) The photon energy measured by the observer at t
0
is E
0
= p

t
. By
0 = p
2
= (p

t
)
2
+ (p
x
)
2
+ (p
y
)
2
+ (p
z
)
2
, we get
E
0
=
_
(p
x
)
2
+ (p
y
)
2
+ (p
z
)
2
which means
p
x
= E
0
n
x
, p
y
= E
0
n
y
, p
z
= E
0
n
z
Using the result of part (c) this gives
p
x
= E
0
a
0
n
x
, p
y
= E
0
b
0
n
y
, p
z
= E
0
c
0
n
z
Using the above result we see that E/kT
e
in the expression for N can be written
as
E
kT
e
=
E
0
kT
0
with
T
0
= T
e
_
_
a
0
a
e
n
x
_
2
+
_
b
0
b
e
n
y
_
2
+
_
c
0
c
e
n
z
_
2
_
1/2
(g) When a = b = c, eq.(27.58) reduces to (noting that n
2
= 1)
T
0
= T
e
a
e
a
0

T
0
T
e
=
a
e
a
0
=
1
1 + z
(note that in this isotropic case E
0
/E = a
e
/a
0
gives the photons redshift.)
Exercise 27.8 Angular-diameter distance [Xinkai Wu/02]
(a) By denition the angular-diameter distance of an object is given by
r
AD
=
D

5
with D being the objects physical size and being its observed angular diam-
eter. Put the object in the equatorial plane = /2, at coorinate distance ,
then
D =
_
ds =
_

0
a()d = a()
Thus we get
r
AD
=
a()

= a() =
a
0
()
a
0
/a
=
R
1 + z
with R a
0
().
(b) The worldline of a radial photon is given by 0 = dt
2
+ a
2
d
2
d
dt
=
1
a(t)
integrating gives
=
_
t0
te
1
a
dt =
_
1+z
1
a
a
0
dt
da
d
a
0
a
Using equations (27.40) and (27.41) we have
a =
_
k 2V (a)
=

k +
8
3

crit
a
2
0
_

R
u
2
+
M
u +

1
u
2
_
where we have denoted u
a0
a
. The denition of
crit
eq. (27.36) gives
8
3

crit
a
2
0
= ( a
0
)
2
. Eq. (27.60) tells us a
0
=
_
k
1
_
1/2
=
_
1
|1|
_
1/2
(assuming
k = 1), where
M
+
R
+

. Plugging all these into the expression for

, we get
= |1 |
1/2
_
1+z
1
du
_

M
u
3
+
R
u
4
+

+ (1 )u
2
Since
R
is very small, we can neglect it and
M
+

, and we get
= |1
M

|
1/2
_
1+z
1
du
_

M
u
3
+ (1
M

)u
2
+

(c) See Figures 2 through 4. Note that for the at universe case k = 0, = 1,
the expression for is given by
=
1
H
0
a
0
_
1+z
1
du

R
u
4
+
M
u
3
+

1
H
0
a
0
_
1+z
1
du

M
u
3
+

6
-10 -5 5 10
-4
-2
2
4
Figure 2: r
AD
(z): the x-axis is Log(z), the y-axis is Log(r
AD
);
M
= 0.3,

=
0.73, its a closed universe.
-10 -5 5 10
-4
-2
2
4
Figure 3: r
AD
(z): the x-axis is Log(z), the y-axis is Log(r
AD
);
M
= 0.3,

=
0.6, its an open universe.
Exercise 27.9 Magnitude-redshift relation [Xinkai Wu/02]
Assume that the supernova emits monochromatic photons. In time inter-
val dt(as measured in the supernovas local Lorentz frame) the supernova emits
dN =
Ldt

photons. These photons arrive at the earth in a time interval mea-

sured in earths local Lorentz frame dt
0
= dt(/
0
)(namely, because of time-
dilation, the arrival rate is smaller than the emission rate), each photon having
energy
0
as measured by earth bound observer. Also note that the area of
the two-sphere at the earths distance is given by S = 4(a
0
())
2
= 4R
2
.
Thus the energy ux measured at earth today is
F =
dN
0
dt
0
S
=
Ldt

0
dt(/
0
)
1
4R
2
=
L
4R
2
_

_
2
=
L
4R
2
(1 + z)
2
Exercise 27.11 Inationary explanation of the isotropy of the CMB [Xinkai
Wu/02]
A portion of the universe with size L
PW
at the Planck-Wheeler time is in
7
-10 -5 5 10
-2
2
4
6
Figure 4: r
AD
(z): the x-axis is Log(z), the y-axis is Log(r
AD
);
M
= 0.3,

=
0.7, its a at universe.
causal contact with itself. This portion inates, during the inationary era, to
size
a
I
a(t = NL
PW
) = L
PW
exp(N) (1)
After that the expansion is dominated by radiation during which 1/a
4
,
hence
a
I
a
0
=
_

I
_
1/4
where the size of the universe today is a
0
H
1
0
, and we can take
I
, the
radiation energy density right after the ination, to be of order
I
, with
being the the vacuum energy density driving ination. Thus we have
a
I
= H
1
0
_

_
1/4
(2)
Combining eq. (1) and eq. (2), we get
N = ln
_
H
1
0
L
PW
_

_
1/4
_
70
is the minimum number of e-foldings we need to solve the horizon problem.
8