Large Scale Structure and

UNIT 15 LARGE SCALE STRUCTURE AND the Expanding Universe

THE EXPANDING UNIVERSE

Structure
15.1 Introduction
Objectives
15.2 Cosmic Distance Ladder
An Example from Terrestrial Physics
Distance Measurement using Cepheid Variables
15.3 Hubbles Law
Distance-Velocity Relation
15.4 Clusters of Galaxies
The Virial Theorem and Dark Matter
15.5 Friedmann Equation and its Solutions
15.6 Early Universe and Nucleosynthesis
Cosmic Background Radiation
Evolving vs. Steady State Universe
15.7 Summary
15.8 Terminal Questions
15.9 Solutions and Answers

15.1 INTRODUCTION
In previous units you have learnt about stars and galaxies and their properties. You
have learnt that there are billions of galaxies in the universe. It is now time to estimate
the size of the universe and discuss the current ideas about its major components and
its origin. The science which deals with the origin of the universe is called cosmology.

In this unit, we discuss some aspects of cosmology. First we learn how distances of
distant objects can be estimated so that we can get an idea of the size of the universe.
Estimation of distances is a very tricky problem because there is no way of verifying
these distances directly. So, we look for internal consistency in various methods,
which means that the distance estimates given by them agree with one another.

In this unit we discuss the distribution of matter on very large scales also called the
large scale structure of the universe. We also look at the kind of matter that may be
forming the bulk of the universe. This matter is not visible and is called the dark
matter. It shows itself up only through its gravitational effect.

Objectives

After studying this unit, you should be able to:

explain how astronomical distances are measured;
state Hubbles law and explain how Hubbles constant indicates the age of the
universe;
explain the need to postulate the existence of dark matter in the universe;
derive Friedmann equation and solve it in simple cases; and
explain why a hot and dense phase in the early universe is needed to explain the
existence of cosmic background radiation and to synthesise light elements.

35
Galaxies and the
Universe 15.2 COSMIC DISTANCE LADDER
An important step in this direction is to estimate the physical size of distant objects
and their distances from us. You know that to specify the position of an object in
three-dimensional space, we need 3 coordinates. Since we are observing from the
Earth (or its vicinity from satellites), it is best to use spherical polar coordinate system
with us as the origin. Two of the coordinates, namely, and (which are related to
the declination and right ascension, respectively) are easily fixed by pointing a
telescope in the direction of the object. Fixing the distance to an astronomical object is
relatively trickier.

In this section, we give an introduction to the basic principles that are used to measure
distances. The basic principle involved is to use the properties of the nearby objects
and deduce distances of similar objects farther off using these properties. Then we use
the latter objects to deduce the distances to objects still farther, and so on. This series
of steps which takes us from one step (in terms of distances) to the next step (in terms
of distances even farther) is termed the Cosmic Distance Ladder.

All this is best understood through an example of estimating distances on the Earth.

15.2.1 An Example from Terrestrial Physics

Consider the following situation. You are in a house and there are small plants around
the lawn in your house. Outside the house also there are similar plants and in addition
there are also some trees. We do not know the height of these trees. Very far away
there is the sea and on the beach also there are a number of such trees. The problem is
to estimate the distance of the sea from your house. To begin with we do not know
about the height of the trees.

The obvious thing would be to walk down to the sea (keeping in mind that all the
steps should approximately be of the same size) and count the number of steps. The
length of the step multiplied by the number of steps gives us the distance to the beach.
But you cannot use this method if you are not allowed to, or are unable to come out of
the house and go to the beach. The situation is more like this in the context of distance
measurement in astronomy. We can make direct measurement only on the Earths
surface. Since it will take about 1,00,000 years, even to reach the other end of the
galaxy by moving at the speed of light, we need to devise a better strategy.

Let us come back to our problem of measuring the distance to the sea beach. We will
put an extra condition that we are not allowed to come out of the gates of the house.
With this constraint we can proceed in the following way.

We first measure the heights h of the plants inside the garden of the house. Next we
measure the angle subtended by similar plants outside which are in the vicinity of
h
the trees. The distance of these plants from us equals d = . Since the trees are in the
vicinity of these plants we know that their distance from us is also d. Further, we can
measure the angle 1 subtended by the heights of these trees. The height h1 of these
trees is now given by h1 = d 1. In this way we have achieved the first step in the
ladder of finding distances. We next use this information to estimate the distance to
the trees near the seaside. Since the trees near the beach are similar to the ones nearby,
we assume that their height is also h1. By measuring the angle subtended by these
trees we can compute the distance of these trees from us. Because of the physical
association of these trees with the sea side we know the distance of the sea from your
house. This is the second step of the ladder. Like this we can add more steps and go
on to estimate the distances of far off objects.
36
15.2.2 Distance Measurement using Cepheid Variables Large Scale Structure and
the Expanding Universe

We will now give an example of using this principle to measure distances in

astronomy. In Unit 1 you have learnt the technique of measuring distances by parallax
method. In this section we will discuss how this method can be used to calibrate
another technique for distance measurements using Cepheid variable stars, which in
turn are used to measure even further distances. The brightness of Cepheid variable
stars is a periodic function of time (Fig. 15.1b).

3.0

Visual Magnitude
3.5

4.0

4.5

5.0
0 5 10
Days
(a) (b)

Fig.15.1: a) Cepheid variable stars in NGC 300; b) the brightness of Cepheid stars as a function of time

We have a large number of such stars in our neighbourhood. Their distances can be
measured by the parallax method. From these distances and from their observed
apparent magnitudes, their absolute magnitudes and luminosities can be calculated. It
turns out that their absolute magnitudes are directly proportional to their periods
(Fig. 15.2). This is called the period-luminosity relation for the Cepheids.
8

6
Absolute magnitude

0
0 0.5 1.0 1.5 2.0 2.5

Log period (days)

Fig.15.2: Period-luminosity relation for Cepheids
37
Galaxies and the This relation is firmly established for the local sample. Now we assume that the
Universe
distant sample of these stars also obeys this relation. So, the observed period of a
member of the distant sample is used to find its absolute magnitude. The apparent
magnitude can be observed directly. Using the relation between the absolute
magnitude and the apparent magnitude:

M = m + 5 5 log r, (15.1)

we can find the distance of this star. In this way, the Cepheid variable stars have been
used to find distances of nearby galaxies. Used in this manner, Cepheids are called
standard candles.

We can now use Cepheids to define some other objects, such as supernovae, to act as
standard candles to estimate even larger distances.

What is the result of these investigations? We find that distant galaxies are rushing
away from us with velocities which are proportional to their distances. This is called
Hubbles law.

Spend SAQ 1
5 min.
Explain the concept of a distance ladder.

15.3 HUBBLES LAW

Hubbles law is probably the single most important step in our attempt to understand
the Universe. This law was discovered by Edwin Hubble and it relates the distances of
galaxies with the velocities with which they are receding away from us (Fig. 15.3).

300,000

250,000

200,000
Velocity (km s )
1

150,000

100,000

50,000

0 500 1000 1500 2000 2500 3000 3500 4000

Distance (Mpc)
Fig.15.3: Hubbles law

The velocity of any object can be split into a component that is along the line-of-sight
and another component that is transverse (perpendicular to) the line-of-sight. The line-
of-sight component of the velocity can be determined very accurately by Doppler shift
of the light that we received from the object. Hence this gives the velocity with which
38 the object is coming towards us or receding away from us. Let us look at this process.
15.3.1 Distance-Velocity Relation Large Scale Structure and
the Expanding Universe

Using the methods similar to those mentioned above, Hubble estimated the distances
and velocities of a set of galaxies and plotted them. He found that the galaxies in
general seem to be receding away from us. This is popularly known as the expanding
universe. Further, the velocities with which they are receding away, are directly
proportional to their distances from us (Fig. 15.3). This prompted him to propose a
law, now known by his name.

Hubbles law

v = Hr, (15.2)

where v is the line-of-sight velocity of an object, r its distance from us and the
proportionality constant H is called the Hubble constant.

The importance of this relation is that, once we know the velocity of a galaxy (by red
shift measurement), we can calculate the distance at which it is located. It is important
to point out here that Hubbles law holds even if we were on some other galaxy. Our
location in the universe does not have any special importance.

Notice that in the above relation H is the slope of the curve shown in Fig. 15.3. Notice
also that 1/H has the dimensions of time. In a very simple picture, we can imagine that
all the galaxies which are today moving away from one another were at some time in
the past together at one point. Some event occurred at that time which triggered the
expansion of the universe. This event is usually called the Big Bang. The quantity 1/H
measures the time since that even, or the age of the universe.

SAQ 2 Spend
5 min.
In astronomy, the velocity is measured in km s1. The distance of galaxies is measured
in Mpc or million parsec. Find the dimensions of H.

Unfortunately the measured value of H has lots of errors. But we know today that it is
roughly 70 km s1/Mpc1. Estimate the age of the universe.

15.4 CLUSTERS OF GALAXIES

Galaxies mostly exist as members of large groups called galaxy clusters
(see Fig. 15.4). A cluster of galaxies contains about a thousand galaxies. A galaxy
cluster consists of a variety of galaxies. It is an observed fact that the gross features of
these clusters are very similar. This fact alone has a far reaching implication. We
expect that roughly the same kind of physical processes are responsible for their
evolution. Further, it is natural to assume that they were created at different times and
began to evolve. This is because, in order to initiate the process of creation and
subsequent evolution of these clusters one needs a certain combination of
astrophysical conditions.

These conditions need not be the same at all points at a given time. Let us illustrate
this situation in the following manner:

Consider a set of systems, A, B, C, Let the evolution in each of these systems be

governed by the same physical processes. If they started to evolve at different times,
we would expect that at any time, in particular today, they should be in different
stages of evolution. Hence, we would expect they should not show great similarity. 39
Galaxies and the
Universe

Fig. 15.4: Cluster of galaxies

Alternatively, if we find that they look similar today, we may naively think that they
must have got created at the same time, so that they have had the same time for
evolution. For the systems under consideration, namely, the clusters of galaxies, we
know that they look similar but at the same time we have reasons to believe that they
started their evolution at different times.

Our simple-minded reasoning leads to the suggestion that they may have been created
at different times but they have reached some kind of steady state today. Such systems
have a simple, but important relationship for their kinematic parameters such as their
mass and velocity. This relationship is called the virial theorem which we now
describe.

15.4.1 The Virial Theorem and Dark Matter

The virial theorem says that if a system is bounded and is in equilibrium, then its
moment of inertia does not change with time. Such a system will have the following
relation between its total kinetic and potential energies:

2T + V = 0, (15.3)

where T is the total kinetic and V is the total potential energy of the system.

Suppose we have a spherical system consisting of N particles that are interacting

gravitationally. Let the position of the ith particle be ri and its velocity vi. The total
kinetic energy is then

1
i=1 vi2 .
N
T= m (15.4)
2

The total potential energy is obtained by summing over the potential energy of all the
pairs. We have

Gmi m j
V= i> j ri r j (15.5)

40
For a spherically symmetric distribution, V will come out to be proportional to Large Scale Structure and
1 the Expanding Universe
M (R) R , where R is the radius of the system and M the total mass enclosed in a
sphere of radius R. The total kinetic energy can be estimated in the following manner.
If the velocities are in random directions, some of the particles of the system will
contribute a blue shift and some red shift. This will result in broadening of the spectral
lines. Hence, from the width of the spectral lines we can estimate the root mean
square (rms) velocity. This gives the total kinetic energy. If the system is in
equilibrium, we should have

T = V/2 (15.6)
or

2
v rms R 1 . (15.7)

When one tries to estimate the mass of the system in the above manner, one finds that
there is much more mass in the system as compared to that suggested by the luminous
mass alone. Hence, it is postulated that there should be a significant fraction of mass
in the form of dark matter. What form this dark matter takes, we do not yet know.
One possibility is that it consists of cold, burnt out stars which emit very little
radiation. Another possibility is that it is in the form of particles which interact very
weakly with normal matter. It is a very active area of research at present.

15.5 FRIEDMANN EQUATION AND ITS SOLUTIONS

Cosmology is the study of the overall features of the universe. In this framework, one
does not bother about local features of the universe like planets or stars. In cosmology,
we set up equations and solve them to obtain the very large scale features of the
universe.

To set up the basic equations governing the evolution of the universe as a whole,
Newtonian theory of gravity is inadequate and is, rigorously speaking, inapplicable.
The correct theory to use is Einsteins General Theory of Relativity. This, however, is
outside the scope of this course. All is, however, not lost. It so happens that if we go
ahead and apply Newtons laws (which strictly speaking is not the correct thing to do)
the final equations which we get are the same equations that result from the correct
theory, namely, the General Theory of Relativity.

The aim of this Section is to study the evolution of the Universe using these equations.
Hence, we will not be disturbed by the fact that the derivation of these is not rigorous.
We will happily go ahead and use these equations since we are aware that the final
equations are the correct ones.

We know that over a variety of scales the universe is not homogeneous. We have
planets, stars, galaxies and clusters of galaxies, which indicate that the universe is far
from being homogeneous. However, if we consider the universe as a whole, then at
large enough scales the universe seems to be homogeneous and isotropic. This last
point needs some explanation.

Consider a well maintained lawn. From a distance, the lawn appears uniformly green.
But as we start analysing the lawn on smaller scales, it begins to lose its homogeneity.
Consider a grass hopper sitting in the lawn. It observes the lawn at a scale which is
about the blade of grass. Clearly it will notice that the lawn is not at all homogeneous.
This means that the lawn is homogeneously green over distance scales which are
much bigger than the size of a blade of grass. In a similar way, we say that over scales
of sizes, much bigger than the size of galaxies, the universe is homogeneous.
41
Galaxies and the The equations that we now derive are the equations that govern the overall evolution
Universe
of a homogeneous universe. Consider two points A and O. With O as centre, and AO
as radius, draw a sphere. The force with which a test particle at A is gravitationally
pulled towards O can be calculated by just using the mass enclosed by the sphere. As
we know, the mass outside the sphere will not exert any net force on A.

Let the density of matter be . Notice that density does not depend upon space. This is
because of our condition of homogeneity. However, there is no such restriction on
time dependence. So we will include time dependence for generality. The mass of the
sphere of radius R = OA is

4
M= (t ) R 3 (t ), (15.8)
3

& be the
The gravitational potential at A due to this sphere is GM/R. Further, let v = R
velocity with which A is moving with respect to O. Depending on the magnitudes of
the kinetic energy and the potential energy, the particle at point A may keep moving
away from O or may turn back and fall towards O. This is the well known condition
for escape velocity.

The total energy per unit mass of a test particle at A is

GM v2
E= + (15.9)
R 2

or

2GM
R& 2 2 E = (15.10)
R

If the energy E is positive, then the distance between A and O will keep increasing and
if E is negative, A will attain a maximum distance from O and then begin to fall
towards O.

This condition on E can alternatively be expressed by replacing 2E by k .2 E ; k = +1

corresponds to the case E < 0.

An alternative way of writing the last condition is then

2GM
R& 2 + k .2 E = (15.11)
R

Dividing both sides by R 2 and expressing the mass M in terms of the density as

4 3
M= R
3

we get,

R& 2 k .2 E 8G
+ = (15.12)
2 2 3
R R

Defining a new variable a = R 2 E , we can now write the Friedmann equation

42 governing the evolution of the distance between two particles.
 Large Scale Structure and
the Expanding Universe
Friedmann equation

a& 2 k 8G
+ = (15.13)
a2 a2 3

We get the same equation from the general theory of relativity. The parameter k then
signifies the curvature of space. In the general theory of relativity, the effect of the
gravitational field is to make the space curved. The curvature of space is denoted by k,
which can take values +1, 0 or 1, depending on the overall density of the universe.
The quantity a is called the scale factor and the nature of a as function of t indicates
the nature of the expansion of the universe.

Fig. 15.5 shows the behaviour of a as a function of t for the three values of k, i.e., +1,
0 and 1. These curves are the solutions of the Friedmann equation. We see that
when k = 1 (which according to general theory of relativity implies that the overall
density of the universe is less than a certain critical density), or k = 0 (the overall
density of the universe equals the critical density), the universe keeps expanding.
When k = +1 (the overall density of the universe is greater than the critical density),
the universe expands up to a point and then starts contracting. Present observations
indicate that the universe will keep expanding, and its expansion will not be followed
by contraction.

In the early phase of the universe, the curvature must have been small, so it is
sufficient to consider the case of k = 0. The solution of the Friedmann equations, of
course, depends on the nature of energy density. If an, the equation can be solved
for k = 0 and the result comes out to be

a ( n / 2) 2 / n t 2 / n (15.14)

k<0
k =0
a(t)

k>0

t
Fig.15.5: The variation of a with time
43
Galaxies and the SAQ 3
Universe Spend
5 min.
n
Verify that Eq. (15.14) is a solution of Eq. (15.13) when a .

Time-temperature Relationship
We know from the thermodynamics of radiation that it has pressure and energy
density which we denote by prad and rad, respectively. They are related to each other
through the relation,

p rad = rad c 2 3 (15.15)

where c is the speed of light. The distance between points increases as R which, in
3
turn, is proportional to a. Hence, the volume increases as a and the energy contained
3
in it as a . Now, using the first law of thermodynamics, dU + pdV = 0, we have

d [ c 2 a 3 ] + pd [a 3 ] = 0. (15.16)

Applying it to radiation, we get

rad a 4 (15.17)

Using this relation in the expression for derived in the last section (with n = 4) we
get,

a t (15.18)

At the same time we know that the temperature of radiation is related to its energy-
density by

rad T 4 (15.19)

for isotropic and homogeneous radiation field.

From the last three equations, we get the relationship of temperature with time as

T 1/ t (15.20)

We see that t = 0 is both interesting as well as disturbing. This is because at that point
of time, the temperature shoots to infinity and so does the energy density. It is this
epoch which is termed as the Big Bang.

15.6 EARLY UNIVERSE AND NUCLEOSYNTHESIS

We saw that the temperature of radiation varies as inverse of the square root of time.
The immediate consequence of this is that the radiation temperature in early times
should have been very high. If matter and radiation were in thermal equilibrium, the
above statements imply that the temperature of matter was also high in early times.

We know from thermodynamics that temperature is a measure of the mean kinetic

energy which in turn implies high energy collisions between particles. The earlier the
epoch, the higher the energy with which these particles collide with each other. Today
we know about the physical phenomena at high energies from experimental
investigations using high energy particle accelerators and colliders. We should expect
44
that the same phenomenon must have taken place in the early universe. In fact, for this Large Scale Structure and
the Expanding Universe
reason, early universe is often called the poor mans laboratory.

An important class of reactions at high energy is those which lead to the synthesis of
nuclei of elements. Very early on, there were no complex nuclei. The only ones were
the hydrogen nuclei, i.e., protons. At those energies, even if the protons and neutrons
combined to form higher atomic number nuclei (e.g., the Helium nucleus), the kinetic
energies of the particles were so high that the collisions would have immediately
disintegrated them.

As the universe cools, a certain temperature is reached when the energies are low
enough that this backward reaction (namely disintegration) begins to get suppressed.
Hence, stable helium nucleus begins to get formed. As the temperature lowers further,
we expect that higher atomic number nuclei will begin to get synthesized. So the
question is, Can we proceed in this way and synthesize all the naturally occurring
nuclei? The answer unfortunately is no!

This line of reasoning works for only the elements with first few atomic numbers. As
the temperature decreases, we can form Lithium and some Boron. The problem comes
up when we need to form Beryllium. In the process of the formation of the stable
Beryllium nucleus, one passes through an intermediate stage where, spontaneous
disintegration is faster than the fusion. So even before there can be fusion the nucleus
which is supposed to participate in the fusion, disintegrates. Hence one cannot form
Beryllium by this procedure. Only after the formation of Beryllium, can the nuclei of
higher atomic numbers be formed. Hence this is called the Beryllium Bottleneck.
The universe has to wait for a very long time, namely, till stars form, in order to
synthesize elements of atomic number 5 and higher. (Refer to Unit 10 for details of
nucleo-synthesis inside the stars).

15.6.1 Cosmic Background Radiation

Do we have any signature of the early hot phase of the universe? The answer is yes. In
1965, two scientists at the Bell Telephone Laboratories in America discovered
accidentally a radiation at a very low temperature of only 3 K which seemed to come
from all directions. It was highly isotropic. It was suggested that the radiation fills the
whole universe. Since the wavelength of the peak radiation, ~ 1 mm, falls in the
microwave region, it was called the cosmic microwave background radiation
(CMBR). This is the relic of the era when the universe was very hot and dense. It is
argued that the radiation was once very hot and has been cooled to its present
temperature due to the expansion of the universe over billions of years (recall from
4
Eqs. (15.20) and (15.19) that T 1 t and rad T ). Put in another way, the same
energy fills an every increasing volume, so its energy density decreases and so does its
temperature.

The discovery of the CMBR has great significance for cosmology. (The discoverers of
the radiation were awarded the Nobel Prize for their work.) It shows that the universe
was once very hot and dense. This lends great support to theories of the universe
which maintain that the universe is changing with time, that is, it is evolving. Its
existence is a very powerful argument against theories which propose that the
universe is steady, that is, it is unchanging. In the theories of the latter type, it is
extremely hard to produce such a radiation.

CMBR is a topic of intense research today.

15.6.2 Evolving vs. Steady State Universe

You have just seen that CMBR points to a phase of the universe when it was very hot
and dense. This phase is generally known as the hot Big Bang. The idea is that some 45
Galaxies and the violent event took place at that time which sent the universe expanding. You have also
Universe seen above that if the early universe had not been hot and dense, it would not have
been possible to synthesise light elements, such as H2, H e4 and Lithium. In fact, the
prediction of the precise observed abundances of these light elements is a very
powerful argument in favour of the universe that changes with time: an evolutionary
universe.

Yet there is a set of scientists who believe in a steady state universe, a universe
which has no beginning and no end, which appears the same at all points in space and
at all times. Historically, the steady state theory emerged in the1940s and 1950s, when
observational techniques were not much developed, and so the Hubble constant H
could not be measured accurately. Recall that 1/H gives a rough time scale of the age
of the universe. So, the age of the universe inferred from the value of H at that time
turned out to be less than the age of some fossils on the Earth. This was quite
embarrassing. To overcome this age problem, the scientists proposed the steady state
universe. However, CMBR and synthesis of light elements in the early universe are
very powerful arguments against this theory and in favour of the hot Big Bang theory.

With this we come to an end of this unit in which you have studied about the large
scale structure of the universe. We now present its summary.

15.7 SUMMARY
Distances of far-off objects inform us about the large scale structure of the
universe. To find distances of other galaxies we employ the concept of the
distance ladder. The first step of the ladder is the Cepheid variable stars found in
nearly galaxies. Subsequent steps include objects such as supernovae which can
be detected even in galaxies which are very far off.

The outcome of the exploration of the universe is that the galaxies are rushing
away from one another and the universe is expanding.
The expansion of the universe is in accordance with Hubbles law, v = Hr, so that
1/H gives a rough estimate of the age of the universe.
The behaviour of the universe is governed by the Friedmann equation:

a& 2 k 8G
+ =
a2 a2 3

The need for a hot and dense early universe can be explained in connection with
the synthesis of light elements and the existence of the cosmic microwave
background radiation.
The present understanding is that if the universe was once hot and dense, then it
must be an evolving universe, and not a steady state universe.

15.8 TERMINAL QUESTIONS Spend 30 min.

1. Explain how Cepheid variables have been used to measure astronomical

distances.

2. State Hubbles law. How can this be used to get an estimate of the age of the
universe?

3. Explain why at one time, the steady state theory appeared necessary. What is its
46 status now?
 Large Scale Structure and
15.9 SOLUTIONS AND ANSWERS the Expanding Universe

Self Assessment Questions (SAQs)

1. See Section 15.2.

1 1
2. H = 70 km s Mpc

= 70 10 5 cm s 1 / 3 1018 10 6 cm

7 10 6
= s 1
24
3 10

1 3 10 24
Age of the universe = = s
H 7 10 6

3 10 24
= yr
7 10 6 3 10 7

= 1.4 1010 yr = 14 billion years.

a& 2 8G
3. Eq. (15.13) =
a2 3
a& 2 8G
Since = Ca n = C a n [C is a constant]
2 3
a

n
a& 8GC n da 1 8GC
= a 2 a& = = Aa 2 A =
a 3 dt 3

n2 n
a 2
a 2 da = A dt
n t a n ( 2) 2n . t 2n .
2

Hence, proved.

Terminal Questions

1. See Text.
2. See Text.
3. See Text.

47