Física del Cosmos

Cosmology
Part II: cosmic kinematics & dynamics

Academic year 2015-16

 Outline of this lecture

• The cosmological principle.

• Cosmological spacetime: the FLRW metric.

• Propagation of light and Hubble’s law.

• The Friedmann equations and their solutions.

• The accelerating Universe.

• Inflation and the “dark” Universe.

• Cosmological problems & puzzles.

• Bonus slides: who discovered the expanding Universe?

Beginning: the cosmological principle (I)
• Having the theory of General Relativity at our disposal, how should we go on
and build a “cosmology”?

• As always, we shall look for a “simple” solution of the field equations.

This is dictated by necessity and practicality: the field equations are too
hard to solve unless some simplification(s) is made. We look for a
spacetime geometry endowed with certain symmetries.

• These are encapsulated in the “cosmological principle”:

On large spatial scales, the Universe is isotropic (there is no privileged

direction) and homogeneous (there is no privileged point).

• This principle is “Copernican” as it implies that our own location in the

Universe is not special in any sense.
 The cosmological principle (II)
• The cosmological principle assumes that spacetime can be “foliated” into
spacelike hypersurfaces of t = constant where t is a “cosmic time”.

• Then, spatial isotropy and homogeneity apply on each t=const. hyper

surface (which are 3-D).

• This also means that the curvature of each hypersurface is constant but it
may vary with time.

• The spacetime is spherically

symmetric about any point, but it
is not stationary.
 Weyl’s postulate
• This postulate singles out a family of “privileged” world-lines — also called
fundamental observers.

• These world-lines are geodesics with constant spatial coordinates (x1 , x2 , x3 ).

The cosmic time t coincides with the proper time τ of these observers.

fundamental observers
• By construction, the world-lines
are orthogonal to the t=const.
hypersurfaces.

• These special world-lines will

later be identified with those of
galaxies.
 The cosmic spacetime metric
• The previous postulates determine the form of the spacetime metric
(as always we use i, j = {1,2,3}):

2 2 i j
ds = dt + gij dx dx [ Weyl’s postulate ]

i 2 k
gij (t, x ) = S (t)hij (x ) [ cosmological principle ]

ds2 = dt2 + S 2 (t)dl2 , dl2 = hij dxi dxj

dr 2 [ constant curvature 3-D space

2
dl2 = 2
+ r 2
(d✓ 2
+ sin ✓d' 2
) &
1 Kr spherical symmetry ]

K = const., (> 0, < 0, 0)

 The FLRW metric (Ι)
p
• With a rescaling of the radial coordinate (when K 6= 0 ): r ! |K|r
p
a(t) ⌘ S(t)/ |K| (K 6= 0)
and the definition of the scale factor:
a(t) ⌘ S(t) (K = 0)
• The cosmological metric takes the famous Friedmann-Lemaitre-
Robertson-Walker (FLRW) form:
 2
2 2 2 dr 2 2 2 2
ds = dt + a (t) 2
+ r (d✓ + sin ✓d' )
1 kr
k = {0, + 1, 1}
• This metric depends only on the cosmic time t . The constant k specifies the
geometry of the t=const. hypersurfaces (3-D space). The origin of the
spatial spherical coordinates (r, θ, φ) can be any arbitrary point
(note that r is dimensionless and that a(t) has dimensions of length).
 The FLRW metric (ΙI)
• An alternative, widely used, form of the FLRW metric is:

2 2 2
⇥ 2 2 2 2
⇤
ds = dt + a (t) d + F ( ){d✓ + sin ✓d' }

where F ( ) = { 2 , sin2 , sinh2 } for k = {0, + 1, 1}

2 [0, +1) 2 [0, ⇡] 2 [0, +1)

r= r = sin r = sinh
 Topology of the FLRW spacetime
• The constant parameter k determines the
k=+1 closed, finite
geometry of 3-D space. k=+1represents a
closed (finite) universe without
boundaries, k={ 0,-1} represent open
(infinite) universes.
k=-1, open,
infinite
• The function α(t) is determined by the
field equations of General Relativity. As
we shall shortly see, its form depends on
the assumed large-scale properties of
matter and radiation (the “cosmic fluid”).

• The 3-D volume of the k=+1 model is:

Z Z Z
⇡ ⇡ 2⇡
3 2 2 3 k=0, open, infinite
V = d d✓ d' a (t) sin sin ✓ = 2⇡ a (t)
0 0 0
 Fundamental observers: kinematics
• The fundamental observes of the FLRW metric are described by constant
spatial coordinates (χ,θ,φ). As a result, the cosmic time t can be identified
with the proper time τ of these observers.

• The proper distance between two fundamental observers at any given

time t can be written as ( without loss of generality we can place one of the
observers at the origin and the other at (χ,θ,φ)= (χ,0,0) ):

D(t) = a(t)

• From this it follows that the relative proper velocity between them is:

dD(t) ↵˙ Hubble
v= = ↵(t)
˙ ) v = H(t)D(t), H(t) ⌘ parameter
dt ↵
1
• Note: the relative velocity v exceeds the speed of light for D(t) > H (t)
 Light propagation in the FLRW metric (I)
• We consider the emission of light from a source (e.g. a distant galaxy) and its
reception by an observer (us). The emitted light can be thought to propagate
radially outwards from the source to the observer.

• Both the source and the observer have fixed FLRW coordinates, = e , =0
respectively, (this implies that the only relative motion is due to cosmic
expansion).
= e =0
t
t = t0 + t0
A pulse of light is emitted at t = te
χ and a successive one at t = te + te .
t = t0
t = te + te Or, these times could represent the
successive “peaks” of a monochromatic
wave.
t = te
source’s world-line observer’s world-line
 Light propagation in FLRW metric (II)
• The radial propagation of light is described by (we temporarily restore “c”):
cdt = ±a(t)d
• Integrating this for the radially ingoing light of the previous figure:
Z t0 Z t 0 + t0 Z t0
cdt cdt cdt c t0 c te
e = = = +
te a(t) te + te a(t) te a(t) a0 ae

where a0 ⌘ a(t0 ), ae ⌘ a(te )

• Therefore:
we can write this in terms
of wavelength & frequency:

0 a0 The redshift of the source depends only

• The redshift is: 1 + z ⌘ =
on ae , a0 and not on a(t) in between.
e ae
 Deriving Hubble’s law (I)
• The idea is to approximate the redshift formula “near” the present epoch t0 .
a0
1 + z(t) =
a(t)
• The Taylor-expansion of the scale factor leads to:

where t = t0 t and we have defined:

ȧ0 ä0
H0 ⌘ , q0 ⌘
a0 H02 a0
Hubble constant deceleration parameter

• Note: the above expansion assumes H0 t ⌧ 1 , that is, the light emission
took place in the “recent” past.
 Deriving Hubble’s law (II)
• Then, a0 ⇣ q0 ⌘
= 1 + H0 t+ 1+ H02 ( t)2 + ...
a(t) 2

• The desired expansion for the redshift is:

⇣ q0 ⌘ 2
z(t) = H0 t+ 1+ H0 ( t)2 + ...
2
This expression clearly demonstrates that z can be used for labelling/measuring
the cosmic past.

• Using the above expansion for a(t) for the emission of light from a source
at = e we find:
Z t0 
cdt c t 1
e = = 1 + H0 t + ... where here t = t0 te
te a(t) a 0 2
 Deriving Hubble’s law (III)
• Therefore, for a source (e.g. galaxy) not too far away (=not too far in the
past), and working at leading order with respect to H0 t :
ze ⇡ H 0 t where t = t0 te
redshift of the source

• The proper distance to the source (at t = te and t = t0 ):

D e = ae e ⇡ a0 (1 H0 t) e = D0 (1 H0 t)

• Finally we need to relate these to the source’s radial velocity at t = te :

ve = ȧe e
From the previous Taylor-expansion: ȧ(t) ⇡ a0 H0 ) ȧe ⇡ a0 H0

distance-redshift relation:
cze
ve ⇡ H0 D0 ⇡ H0 De ⇡ cze De ⇡ D0 ⇡
Hubble’s law! H0
 More on Hubble’s law

• For the derivation of the Hubble law we have only resorted to the cosmological
principle and the FLRW metric that comes with it. The GR field equations
have not been used at all. Hence, Hubble’s law is a purely kinematical effect.

• In the derivation we have assumed H0 t ⌧ 1 which implies low redshift

sources, “slowly” receding:

ve
ze ⇡ H 0 t ⌧ 1 ) ⇡ ze ⌧ 1
c
• Hubble’s law may look like the usual Doppler formula but it is not a Doppler
effect since the observer and the source are not in the same inertial frame.
Hubble’s law is truly a consequence of the FLRW expanding spacetime.

• At the precision of Hubble’s law formula the two proper distances D0 , De are
indistinguishable.
 Cosmological particle horizon (I)
• Recall that the proper radial distance in the FLRW metric is: D(t) = a(t)

• Light propagation is described by: cdt = ±a(t)d

• This means that the speed of light in the FLRW spacetime is:

dD d
vlight = = ↵(t) =c
dt dt
• The finite speed of light means that at any given time t = t0 an observer can
see only a finite part of the Universe. This is the region inside the observer’s
past light cone.

• The size of this “observable Universe” defines the particle horizon:

Z t0
cdt
ph = Dph = a0 ph
0 a(t)
Cosmological particle horizon (II)
=0 = ph (t0 )
time
t = t0

universe not visible yet

past
l ig h
t co
ne
visible universe
t < t0

t=0
coordinate distance
 Light from superluminal galaxies!

redshift z

ph (t0 )

subluminal recession vrec < c superluminal recession vrec > c

particle horizon: Dph (t) = a(t) ph (t) Speed of incoming light relative to us:
c
Hubble sphere: DHS (t) ⌘ vlight (t) = vrec (t) c = H(t)a(t) c
H(t)
 Geodesics in the FLRW spacetime
• It can be shown that all geodesics are radial relative to some
fundamental observer. This means that we need only to consider radial
geodesics (i.e. motion with θ = const., φ = const.).

dD d
• In general, the radial velocity is: v= = (a ) = H(t)D + a ˙
dt dt
“Hubble flow” “peculiar motion”
• From the usual geodesic equation we have (for timelike worldlines):
2 2
✓ ◆2 ✓ ◆2
d 1 d dt ȧ d dt d t 0 d d
2
= 2 01 = 2 2
= 11 = aȧ
d⌧ d⌧ d⌧ a d⌧ d⌧ d⌧ d⌧ d⌧
✓ ◆2 ✓ 2
◆ 1/2
2d A A
2
dt A
) a = A = const. =1+ 2 ) vpec =a˙ = 1+ 2
d⌧ d⌧ a a a
• Conclusion: peculiar motion is attenuated by cosmic expansion,
a geodesic tends to a fundamental worldline ( χ= const. & τ=t ).
 Dynamics: using the field equations
• This time we use the full field equations (with G and c restored):

µ⌫ µ⌫ 8⇡G µ⌫
G + ⇤g = 4 T
c
• The non-zero Ricci tensor components and Ricci scalar are (using {t,r,θ,φ}):
3 ä Z
R00 = 2
, R11 = 2
, R22 = r2 Z, R33 = R22 sin2 ✓
c a 1 kr
✓ 2
◆
aä 2ȧ2 k ä ȧ
where: Z = 2k + 2 + 2 R= 6 2
+ 2+ 2 2
c c a ac c a

• The large-scale matter/radiation is assumed to be a perfect fluid:

⇣ 4-velocity (relative to a fundamental observer)

p⌘ µ ⌫
T µ⌫ = ⇢ + 2 u u + pg µ⌫ µ t r ✓ '
u = (u , u , u , u ) = (1, 0, 0, 0)
c
spatial velocity = 0: no preferred direction.
 The Friedmann equations
• The field equations lead to the following Friedmann equations for the
evolution of the scale factor:
✓ ◆2
2 ȧ 8⇡G kc2 1 2
H ⌘ = ⇢ 2
+ ⇤c
a 3 a 3

✓ ◆
ä 4⇡G 3p 1 2
= ⇢+ 2 + ⇤c
a 3 c 3

• These two equations can be combined to give:

3
d 2 3 da this is the 1st law of thermodynamics
(⇢c a ) + p =0
dt dt (conservation of energy)
dE = pdV
 “Einstein-Newton” equations
• Consider the dynamics of an expanding, self-gravitating sphere of radius
R(t) = a(t) = const.
• Newton’s second law for the mass m:

GM R̈ ä 4⇡G
R̈ = ) = = ⇢
R2 R a 3 m
• Conserved energy E (per unit mass): R(t) = a(t)

1 2 GM Ṙ2 8⇡G 2E
Ṙ =E ) 2 = ⇢+ 2
2 R R 3 R

ȧ2 8⇡G k
) 2 = ⇢ k⌘ 2E/
a 3 a2 4⇡ 3
M = ⇢V = ⇢ R
3
 Matter-dominated Friedmann universe
• As it is customary, we solve the Friedmann equations for two special but
physically relevant cases, representing two different stages of cosmic
evolution. For simplicity we assume k=Λ=0 (i.e. flat models with no
cosmological constant).

• Matter-dominated era:
on large scales, the cosmic fluid (galaxies) is pressureless “dust”, p=0.
We have:
constant 2 C 8⇡G
⇢mat = ȧ = , C= ⇢mat a3 = constant
a3 a 3

✓ ◆1/3
9C 2/3
constant
) a(t) = 4
t ) ⇢mat =
t2

Spacetime expands, density of matter decreases

 Radiation-dominated Friedmann universe
• Radiaton-dominated era:
the large-scale photon gas (including relativistic particles) has a pressure:

1 2
prad = ⇢rad c
3
• Using once again the Friedmann equations:

constant 2D 8⇡G
⇢rad = ȧ = 2 , D= ⇢rad a4 = constant
a4 a 3

1/4 1/2
constant
) a(t) = (4D) t ) ⇢rad =
t2

Spacetime expands, density of radiation decreases

 Why Big Bang?
2
• For a more general pressure-density relation p = wc ⇢ (constant w > 0)
we obtain:

3(1+w) 2 B 8⇡G 3(1+w)

⇢/a and ȧ = , B= ⇢a = constant
a1+3w 3
2/[3(1+w)]
) a(t) ⇠ t

• The essence of the “Big Bang”: There is an “origin” in time, t =0, where the
density, the spacetime curvature (Riemann tensor), and other physical
quantities diverge.
↵ ä 2
R ⇠ ⇠t
a
• In reality, GR itself breaks down at t ! 0 and a more advanced theory is
needed (quantum gravity) in order to explain the “birth” of our Universe.
 More on the decay of density
• There is a very intuitive way to understand the evolution of the energy
density with time.

• The volume of a spherical region of coordinate radius χ is: V / a(t)3

• Let N be the number of particles (they could be photons) inside V, then the
number density n (=number of particles per unit volume) is:
3
n/a
3
• For non-relativistic particles of rest mass m: ⇢ ⇡ mn / a
h⌫ 4
• For photons (or other highly relativistic particles): ⇢rad = 2 n / a
c
1
recall that: ⌫ / a
• For a photon gas in thermal equilibrium:
⇢rad / T 4 ! T /a 1
Cooling of the CMB photon gas
 Friedmann models with k, keeping Λ=0
• We define the “critical” density:
3H02
⇢crit ⌘ H0 = H(tnow ) = Hubble parameter at present epoch
8⇡G
• We also introduce:
⇢0 = ⇢(tnow ) = total matter+radiation density at present epoch
• The first Friedmann equation can be written as:
⇢0 c2 k
⌦0 ⌘ =1+ 2 2
⇢crit a 0 H0

• We can see that the topology k={-1, 0,+ 1} of each cosmological model is
directly related to the density ratio ⇢0 /⇢crit .

⇢crit ⇠ few protons/m3

Friedmann models with k, and Λ=0
Matter-dominated models

k= 1, ⌦0 < 1
scale factor α(t)

k = 0, ⌦0 = 1

k = +1, ⌦0 > 1
 Λ-dominant: the de Sitter model
• A growing a(t) implies a decaying density ⇢ ⇠ a n . At some “late” epoch
the Friedmann equations will approximately behave as if the Universe
were devoid of matter (this is the de Sitter model):
✓ ◆2
ȧ 8⇡G kc2 1 2 1 2
= ⇢ 2
+ ⇤c ⇡ ⇤c
a 3 a 3 3
a(t) / exp[(⇤c2 /3)1/2 t]
✓ ◆ )
ä 4⇡G 3p 1 2 1 2 ä(t) > 0
= ⇢+ 2 + ⇤c ⇡ ⇤c
a 3 c 3 3

• That is, a Λ> 0 constant drives accelerated, exponential expansion!

• It is clear that all Λ> 0 expanding models will asymptotically tend to

the de Sitter universe!
 Λ-term: a negative pressure “fluid”
• Accelerated expansion ä > 0 requires a negative pressure fluid or
a Λ-dominated Universe.

• In fact, the cosmological constant can be viewed as an effective

matter contribution:
8⇡G µ⌫ µ⌫ 8⇡G µ⌫ µ⌫
G µ⌫
+ ⇤g µ⌫
= 4 T , G = 4 (T + Te↵ )
c c

• Seen as a “fluid”, the cosmological constant has negative pressure and

can therefore drive accelerated expansion:

2 ä 4⇡G 3
p⇤ = ⇢⇤ c = ⇢ + ⇢⇤ + 2 (p + p⇤ )
a 3 c
c2 ⇤
⇢⇤ = 2 8⇡G kc2
8⇡G H = (⇢ + ⇢⇤ )
3 a2
Summary of the Friedmann universes

⇤=0

k = +1 k=0 k= 1

⇤>0
 A “Game of Densities”

radiation matter “dark energy”

dominates dominates dominates

4 3 ⇢⇤ = const.
⇢rad ⇠ a ⇢mat ⇠ a
 Estimating the age of the Universe
• For most of its lifetime, the Universe has been matter-dominated.
Radiation was dynamically important for a tiny fraction ( ⇠ 105 yr ) of the
total age.

• Thus, we can use the matter-dominated model (for k=0):

2/3 ȧ(t) 2 2
a(t) = At ) H(t) = = ) tage =
a(t) 3t 3H0
1 1
using: H0 ⇡ 68 km s Mpc ) tage ⇡ 9.6 Gyr
• This is close to the correct result (13.8 Gyr) but still wrong: stars in
globular clusters are older than this age!

• Conclusion: the cosmological constant must be an important dynamical

factor!
 The present epoch: the “ΩΩ” equation
• We return to the general Friedmann equations, and we consider the
present cosmic epoch t = t0 :

8⇡G kc2 ⇤c2

2
H0 = ⇢0 2 + ) ⌦0 + ⌦ ⇤ + ⌦ k = 1
3 a0 3

we have
⇢0
defined: ⌦0 ⌘ = ⌦mat + ⌦rad ⇡ ⌦mat
⇢crit
kc2
⌦k ⌘
H02 a20
⇤c2
⌦⇤ ⌘
3H02
 A more general “ΩΩ” equation
• Again considering the Friedmann equation and treating everything as a fluid:

8⇡G kc2 2
H (t) ⇢ ⇣⇣↵ ⌘
a 2
2 0
H (t) = ⇢ ) 2 = + ⌦kk
3 a2 H0 ⇢crit ↵
a
• Assume that the cosmic fluid is made of (non-relativistic) matter and some
arbitrary “X” component:
⇢ = ⇢mat + ⇢x ⇣ a ⌘3
0 3(1+w)
) ⇢mat = ⇢0 , ⇢x / a
px = w⇢x c2 a

H 2 (t) ⇣ a ⌘3 ⇣ a ⌘3(1+w) ⇣ a ⌘2
0 0 0
2 = ⌦mat + (1 ⌦mat ⌦k ) + ⌦k
H0 a a a
 Accelerating expansion?
• Up until the mid-1990s, the cosmological constant Λ was systematically
ignored, believed to be equal to zero.

• The Friedmann equations, without the Λ terms, predict:

✓ ◆
ä 4⇡G 3p decelerating expansion since “normal”
= ⇢+ 2 <0 matter/radiation has ⇢ > 0, p > 0.
a 3 c

• Accelerating expansion requires some exotic type of matter (“dark energy”)

with negative pressure:
1 2
ä > 0 , p < ⇢c
3

• What observations have to say about the sign of ä ?

 Beyond Hubble’s law
• Hubble’s linear law is accurate for low redshifts (z << 1), that is, for “nearby”
galaxies.

• The theory generally predicts

deviations from the linear relation:

cz
d= + O(z 2 )
H0

ä0
depends on q0 =
H02 a0

• Reasonable expectation: ä < 0

the expansion slows down, after all
it does work against gravity.
 Evidence for accelerating expansion!
Using distant Type Ia Supernovae as “standard candles” observations
seem to suggest an accelerating expansion, i.e. ä0 > 0

This discovery won the

2011 Nobel prize in
Physics.
 Further evidence for acceleration
• Three independent observations,
modelled with the Friedmann
cosmological models: Hubble’s law,
CMB and light element abundances.

• Key findings:

✓Large scale acceleration. The cosmic

dynamics is presently dominated by a
“dark energy” (= a “fluid” with negative
pressure).

✓The local Universe is almost flat, ⌦k ⌧ 1

but k remains unspecified.
The content of the present day Universe

Our present understanding can be summarised as follows:

⌦mat + ⌦⇤ ⇡ 1

⌦⇤ ⇡ 0.67

⌦mat ⇡ 0.33

Constraints on k
Present day Universe: (topology of the Universe):
locally flat, global topology unknown
0.007 . ⌦k . 0.006
Note: a k=0 Universe can be mimicked by k=±1 and
H0 a0 /c >> 1 (which is the expected state of affairs!)
 Cosmic acceleration: explanations
• Cosmological constant.
✓simplest explanation, so far agrees with observations. Attempts to produce
an effective cosmological constant using quantum field theory leads to an
enormous discrepancy with the observed Λ.

• Dynamical dark energy.

✓it implies the existence of some unknown “substance” with negative
pressure.

• Modified gravity.

✓it implies that General Relativity is not the correct theory of gravity.
• Deviations from the homogeneity of the FLRW spacetime.
✓no new physics required, seems unlikely though.
 The theory challenge
• Is really Λ one of Nature’s fundamental constant (like G) ?

• Or is Λ a form of “dark energy”, a kind of exotic matter, with uniform

distribution and negative pressure, driving the cosmic expansion ?

• Answering these questions poses one of the biggest theoretical

challenges of modern physics.

• Future observations should be able to

test with higher precision Λ’s pressure-density
relation. If it turns out that:

p no cosmological constant,
2
6= 1 )
⇢c dark energy instead
 Does the vacuum gravitate?
• Heisenberg’s uncertainty principle describes a
fundamental property of Nature: particles and fields are
never at rest.

• Even at zero temperature, atoms in a crystal have

a “zero point energy” and oscillate around some average
position.

• Similarly, the vacuum is filled by the zero-point energy of

fields. The existence of this quantum vacuum has been
verified experimentally (the Casimir effect). 2
⇤QFT ⇠ LP

• Early on, there were attempts to “explain” Λ as the ⇤obs 120

gravitational contribution of the quantum vacuum. ⇠ 10 !
⇤QFT
Unfortunately, there is an enormous disagreement
between the predicted value and the observed one.
 The elusive “dark matter” (I)
• The study of the rotational profile of galactic disks revealed something
unexpected: the existence of some kind of “dark matter”!

• Dark matter does not emit any light or any other electromagnetic signal, it
only interacts gravitationally. Its exact nature remains unknown.

Predicted profile: v ⇠ 1/r

Observed profile: v ⇠ constant

 The elusive “dark matter” (II)
• We now know that most of the mass in a galaxy (about 90%) takes the form of
a dark matter halo.

• Additional evidence for the existence of

dark matter is provided by observations
of colliding galaxies and by gravitational
lensing.

• The alternative interpretation, based on

a modification of Newtonian gravity, is
now out of fashion.

• Dark matter dominates over normal,

luminous baryonic matter:

⌦mat ⇡ ⌦DM
 The “horizon problem”

• There two main problems attributed to the standard FLRW cosmological

model we have discussed in this lecture:

✓The homogeneity problem, also known as the “horizon problem”:

spacetime regions that could not have been in causal contact in a normal
radiation/matter dominated Universe appear to have the same
properties (e.g. the same CMB temperature).
 The “flatness problem”

✓The flatness problem:

the local Universe appears to be almost flat, that is “almost k=0”.
✓ ◆2
ȧ 8⇡G kc2 1 2 1 2
= ⇢ 2
+ ⇤c ⇡ ⇤c
a 3 a 3 3

3 4
⇠a ,a

Unless we truly have k=0, it seems difficult to explain why the k-term is so
much smaller than the matter/radiation term, given its slower decay
during matter/radiation-dominated eras.
 Inflation? (I)
• The inflation model provides a resolution of the horizon and flatness
puzzles by invoking a very brief phase of exponential expansion in the
very early universe ( t ⇠ 10 35 s ).

ȧ
• During the inflationary phase: a(t) / exp(Ht), H(t) = = constant
a
 Inflation? (II)
• Solving the horizon problem: an initially small, causally connected region,
can grow and encompass the entire observable Universe (and much more!).

• Solving the flatness problem: inflation makes the k-term tiny because it
makes α(t) very large.

• The required growth in a(t) is enormous:

afin 60
⇡e
ain
• The inflation model relies on the phase transition of a “scalar field”. Its
predictions are in agreement with existing high precision cosmological
observations (like the CMB map) but the physics details of the inflationary
mechanism remain highly uncertain.
More exotic: a Multi-verse!
 Cosmic fine tuning? Gravity & Entropy

• Gravity makes a fundamental change in the high entropy state of a

system. It promotes non-uniformity rather than uniformity.

• Our Universe, being

homogeneous on large
scales, is in fact in a state of
extremely low entropy!

• In this sense, our Universe

appears to be fine-tuned.

• Nobody knows why is that.

Could it be the result of
quantum gravity physics at
the moment of the big bang
(t=0)?
The Big Question is still open ...
 Addendum:
Who discovered the expanding Universe?

Friedmann was the first to discover (1922)

the model of an expanding Universe filled
with matter.

However, he made no attempt to relate his

mathematical model with the real world (and
observations).

Alexander Friedmann
Who discovered the expanding Universe?

Using the world’s most powerful telescope,

Hubble was able to show (1929) that
“extragalactic nebulae” (galaxies) obey a
linear velocity-distance law, what later became
known as Hubble’s law.

However, Hubble himself never claimed that

he discovered an “expanding Universe”. He
actually never associated his empirical law
with any theoretical models for the Universe.

Edwin Hubble
Who discovered the expanding Universe?

After rediscovering the Friedmann models,

Lemaitre was the first to make the connection
between theory and actual observations,
suggesting the notion of an expanding Universe.

Unfortunately, his original paper (1927) was

written in French (and published in a Belgian
journal) and was largely ignored by the
scientific community. Only later, in the 1930s,
when an english translation was available,
Lemaitre’s work became widely known.

George Lemaitre