SPA6311 Physical Cosmology 2015

Lectures 3.1, 3.2 and 3.3

Lecture by K. A. Malik

The Stuff the Universe is Made Of

Non-dark or Luminous Matter

• Baryons: The Constituents of atoms i.e massive and non-relativistic. Note: non-
standard definition of baryons, but usual in Cosmology

• Neutrinos: Weakly interacting particles, very light but not massless. Usually rel-
ativistic

• Radiation: photons, massless, relativistic

Relativistic: The energy of the particles is dominated by the kinetic energy

Non-relativistic particles: energy is dominated by the particles rest mass

Dark Stuff

• Dark Matter: so far hypothetical particles, massive, very likely non-relativistic,

non-interacting or weakly interacting

• Dark Energy: recent addition to the energy balance of the universe. Maybe cos-
mological constant, or maybe something even stranger (e.g. a scalar field)

Note: dark we mean non-interacting with radiation, means interaction is through grav-
ity. Also note, very embarrassingly, dark stuff forms ∼ 95% of the energy density today.

The Expanding Universe

We frequently use comoving coordinates, xi (where the index i runs from 1 to 3), which
are carried along with the expansion of the universe. These are related to physical
coordinates ri by

ri = a(t)xi

where a(t) is the scale factor of the universe, time dependant only (spatial homogeneity
and isotropy). It is often convenient to define a(t0 ) = 1, so that today physical and
comoving coordinates coincide, and a < 1 in the past. Comoving coordinates move with
the expansion, and hence ẋi = 0.

1
Get Hubble’s law from r = ax

ṙi = ȧxi + aẋi

ȧ i
= r
a
v = Hri

where H = ȧa and v i = ṙi . Or H = H0 if H is evaluated today. Hubble’s law is

not exact, (as on small scales the cosmological principle does not hold), galaxies have
peculiar velocities vpec (due to random motions),

v i = Hri + vpec
i

Expansion and Redshift

Recall the Doppler-shift of the frequency

ωobs 1
= v
ωem 1− c
v
= 1+
c
approaching and for receding emitter
ωobs v
= 1−
ωem c
Consider now electromagnetic radiation passing two observers (comoving). They are
separated by dx = cdt. The second observer recedes from the first with

v = Hdx
ȧ
= cdt
a
Observed reduction in frequency is
v
ω(t + dt) − ω(t) = −ω(t)
| {z } |{z} c
ωobs ωem

Use Taylor expansion

dω
ω(t + dt) = ω(t) + dt + ...
dt
combine with above and get
ȧ
ω̇dt = − cdtω
a

2
or
dω da
∝ −
ω a
−1
ω ∝ a

This gives for the wavelength λ

λ∝a

and finally we find

λobs a0
1+z = =
λem aem
true in general though derivation holds only for short distances.

Recap

We introduced comoving coordinates

ri = axi

Redshift

λ∝a

From our previous definition of redshift

a0
1+z =
aem
Consider an electromagnetic wave travelling through space. We can think of this as a
2D rubber sheet. If we expand the sheet, the wave is stretched.

3
Deriving the equations that govern the evolution of the universe

Newtonian derivation

It was found in the 1920’s that the governing equations for the evolution of the uni-
verse can also be derived using Newtonian gravity.

Newton’s law of gravity:

GM m
F =
r2
G in this case is Newton’s constant and is 6.67 × 10−11 m3 kg −1 s−2 . Note: gravity is
attractive, and weak. Gravitational potential energy
GM m
V =−
r
Friedmann Equation

Alexander Friedmann (1888-1925). Assume that the universe is homogeneous and isotropic.
We can imagine the universe, and cut out a sphere of radius r with a test particle of
mass m. The sphere will have a mass density, ρ. Newton tells us m only feels the mass
inside of the spherical shell. Spherical region is centred at any point relative to which
we want to calculate the force (due to homogeneity). Mass inside the shell
4π 3
M= r ρ
3
Force
Mm 4π
F =G 2
= G mrρ
r 3
Potential
4πG 2
V =− mr ρ
3
Test particle has kinetic energy
1
T = mṙ2
2
Total energy of the system

U = T +V
1 2 4πG 2
= mṙ − mr ρ
2 3

4
Note: the total energy U is a constant of integration of the equations of motion (can be
different for different particles). Introduce comoving coordinates r = ax

ṙ = ȧx + |{z}
aẋ
0

get total energy

1 4πG
U = m (ȧx)2 − ρm (ax)2
3 3
Rearrange
ȧ2 8πG 2U 1
= ρ+
a2 3 mx2 a2
Finally, get Friedmann equation
 2
ȧ 8πG k 2 c2
= ρ− 2
a 3 a
where
2U
k=− (constant)
mc2 x2

Thermodynamics and Energy Conservation

First law of thermodynamics is

dE + P dV = T dS

where E is the energy, P is the pressure, V volume, T temperature and S entropy.

Consider a sphere with comoving radius x = 1

4π 3 4π 3
V = r = a
3 3
Energy E = mc2
4π 3 2
E= a ρc
3
Differentiate
dE 4π 3 2
= 4πa2 ρc2 ȧ + a c ρ
dt 3
dV
= 4πa2 ȧ
dt

5
Putting things together and assuming no heat exchange (adiabatic expansion) dS = 0
  
ȧ P
ρ̇ + 3 ρ+ 2 =0
a c
Energy conservation in an expanding universe.

Recall: fluid dynamics

~ (ρ~v ) = 0
ρ̇ + ∇

However homogeneity and isotropy ρ = ρ(t) and v = v(t). So ρ̇ = 0.

Next Step: supply an equation of state that is P = P (ρ).

General Relativistic Derivation

Einstein’s field equations are “simply”

8πG
Gµν = Tµν
c4
Here Gµν is the Einstein Tensor, Tµν is the energy-momentum tensor. The Einstein
tensor Gµν is defined as
1
Gµν = Rµν − gµν R
2
where Rµν is the Ricci Tensor, gµν is the metric tensor and R is the Ricci Scalar. Gµν
describes how spacetime curves due to the matter content, which is described by Tµν .
Rµν and R (and hence Gµν ) can be calculated from the metric tensor gµν (via the con-
nection coefficients Γα µγ ).

Hence we need to specify gµν and Tµν . Homogeneity and isotropy limit our freedom
to choose gµν and Tµν .

Recap
 2
2 ȧ 8πG kc2
H = = ρ− 2 Friedmann
a 3 a
  
ȧ P
ρ̇ + 3 ρ+ 2 =0 Energy Conservation
a c

6
to close system: need equation of state

The Metric Tensor: (encodes the geometry of spacetime)

Length of ds (in 2D) by Pythagoras

ds2 = dx2 + dy 2

We can introduce the metric tensor gµν . The metric can be written as a matrix (in this
case gmn , where n = 1, 2 and m = 1, 2 as we are working in 2 dimensions)
 
g g
gmn = 11 12
g21 g22

This can be written as

XX
ds2 = gmn dxm dxn
m n

Or in the usual Einstein summation convention

ds2 = dxm gmn dxn

Looking at our 2D example we find that

 
1 0
gmn =
0 1

We can extend to 3D (extend coordinate range), and then further to 4D (where we in-
clude time)

Back to Cosmology: Homogeneous and isotropic universe

The Friedmann-Robertson-Walker Metric (FRW)

The FRW metric is specified by the following line element

dr2
 
2 2 2 2 2 2
ds = −c dt + a (t) + r dΩ(2)
1 − kr2

Where k is the curvature, and r is the comoving radial coordinate and

dΩ2(2) = dθ2 + sin2 θdφ2

7
For flat space k = 0 this reduces to

ds2 = −c2 dt2 + a2 (dx1 )2 + (dx2 )2 + (dx3 )2

Which can be rewritten as

ds2 = −c2 dt2 + a2 δij dxi dxj

FRW is the most general spatially homogeneous and isotropic metric. H therefore has
constant spatial curvature.

Tµν : The energy momentum-tensor of a perfect fluid

The cosmological principle also limits the choice for Tµν

Tµν = (ρ + P ) uµ uν + P gµν

where ρ energy density, P pressure, uµ the fluid 4 velocity. Here

uµ = (1 0 0 0)

T00 :

T00 = (ρ + P ) + (−1) P = ρ

Tij :

Tij = P a2 δij Flat Case

The Einstein equations also give energy-momentum conservation:

Dµ Gµν = 0

where Dµ (or ∇µ ) is known as the covariant derivative and is used in tensor calculus.
By virtue of the Bianchi Identity:

Dρ R µνσ + Dν R µσρ + Dσ R µρν = 0

We can derive that

Dµ (Gµν − 8πGTµν ) = 0

So

Dµ Tµν = 0

8
The Acceleration Equation

Combining the Friedmann and energy conservation equation, we get

 
ä 4πG P
=− ρ+3 2
a 3 c

Usually ρ > 0 and P > 0. This implies ä < 0. So we have a de-accelerated expansion.

However: non-standard matter, e.g. dark energy, can have P < 0

Geometric Units

Very popular in cosmology c = 1

How does this work?

Instead of using m, we now use a new length unit n.l.u, such that 1 n.l.u = 3.108 m
n.l.u
c=1
s
Then, from special relativity, recall that space and time are equivalent, get

c=1

However in this course, usually, we will not use c = 1.