ASTR/PHYS 4080: Introduction to Cosmology Spring 2019

Homework 1 Solutions
1. Cosmic strings are hypothetical ∼1-dimensional topological defects in space-
time that may have formed in the early universe. In the novel Tides of Light by
Gregory Benford (also a physicist), an advanced alien civilization uses a cosmic
string to bore a hole through the center of a planet to mine heavy metals in
its core. The main character is tossed down the shaft through the planet and
calculates how long the journey will take him, which is given as this equation
in the 1989 paperback edition:
" #
π −1 v 3
time = − tan q , (1)
2 R 4π Gρ 4πGρ
3

where v is his initial velocity, R is the radius of the planet, G is the gravitational
constant, and ρ is the planet’s average density. Use this equation (and common
sense reasoning) to estimate the length of the journey for an astronaut falling
through an Earth-like planet.

Solution

Common sense immediately tells us there’s a problem, as the units of 1/Gρ is

not seconds, but s2 . The expression √ v4π must be unitless (since it’s inside
R 3
Gρ
q
the tan−1 () function), and since v/R has units of t−1 , the term 4π 3
Gρ must
also have inverse time units. The inverse square of that is outside the brack-
eted expression, giving units of t2 . Since this equation is found in a paperback
science fiction novel, the most likely mistake is that the editor accidentally
dropped a 1/2 power around the expression outside of the brackets.

To check this, we can attempt to rederive the equation. The equation of mo-
tion is mg = ma, where g = GM r2
and M = ρV = 4πr3 ρ/3. Putting that
together and recognizing a = r̈, we get r̈ − 4π 3q
Gρr = 0, which is the equa-
tion of a harmonic oscillator of frequency ω = 4π 3
Gρ. The solution is of the
form r = R cos(ωt − φ). From here, with some manipulation and taking the
derivative of r, the equation can be derived, but it should be clear enough that
indeed, it’s simply that the square root is missing.

1
ASTR/PHYS 4080: Introduction to Cosmology Spring 2019

If the character takes the equation at face value and lazily plugs and chugs and
assumes his answer will be in seconds, given an Earth-like average density of
∼ 5500 kg m−3 , he would estimate his trip to take over 200 hours (assuming v
is small—it would have to be dangerously large to matter). Neglecting v, the
π
time elapsed is 2ω , which from the harmonic oscillator equation is obviously
one-fourth of the total cycle, or the time it takes to reach the center of the
planet, so to calculate how long it takes to get to the other side, we need to
double the number given by the equation, so ∆t = ωπ .

Considering the high temperature at the centers of planets, his miscalculation

would lead to a prediction of an ordeal that would not only be longer, but
also quite worse, than James Franco’s in that movie about the idiot who goes
hiking alone without telling anyone. Based on this calculation, the character
may have chosen to commit a more painless suicide rather than slowly bake to
death, when the correct form of the equation would have told him to just wait
out the ∆t ∼ π/ω ∼ 42 min trip (where 1/ω = ( 4π 3
· 6.67 × 10−8 g−1 cm3 s−2 ·
5.51 g cm−3 )−1/2 ∼ 800 s).

The equation assumes the planet’s density is constant, which is generally not
true and therefore the equation will not provide an exactly correct answer.
However, the density of the Earth only varies by a factor of a few: (3 − 13) ×
103 kg m−3 , and the variation with density goes as ρ1/2 , so corrections should
be on the order of (131/2 − 31/2 )/6 or ∼30%, which pales in comparison to the
error invoked by a missing 1/2 power.

2. Suppose you are in an infinitely large, infinitely old universe in which the
average number density of stars is n∗ = 109 Mpc−3 and the average stellar
radius is equal to the Suns radius of R∗ = R = 7 × 108 m.

(a) How far, on average, could you see in any direction before your line of
sight struck a star? (Assume standard Euclidean geometry holds true in
this universe.) Put your answer in units of Mpc.

Solution

To decide how far, on average, one can see in a universe filled with spher-
ical objects of radius R, it is simplest to think of a long cylinder along the

2
ASTR/PHYS 4080: Introduction to Cosmology Spring 2019

line of sight. If an object is closer than R to the line of sight, then the
line of sight intersects the object’s surface.

For a distance l, the cylindrical volume which would contain such objects
is πR2 ln, where n is the density of objects. Therefore, the volume that
will, on average, contain one object is defined by πR2 ln = 1, and the
average distance to which one can see is given by:
1
l= . (2)
πR2 n
You have probably seen a similar equation for the mean-free-path between
collisions of, for example, atoms in a gas. Another way to derive equation 2
is to calculate the probability of each mean-free-path when the probability
of collision in each dl along the line of sight is dlπR2 . Calculating the
average value of l using the probability distribution yields equation 2.
Given a stellar density of n = 109 Mpc−3 and converting the stellar radius
of R = 7 × 108 m to Mpc the average distance is:
1
l= 7×108 m
= 6.2 × 1017 Mpc
π( )2 (109 Mpc−3 )
3.086×1022 m
Mpc

This distance is much larger than the distance light has traveled since the
Big Bang, ∼ 6000 Mpc, which is a large part of the reason why the night
sky is dark.

(b) If the stars are clumped into galaxies with a number density ng = 1 Mpc−3
and average radius Rg = 2000 pc, how far, on average, could you see in
any direction before your line of sight hit a galaxy? Put your answer in
units of Mpc.
Solution
If galaxies have an average density of 1 Mpc−3 and an average radius of
R = 2000 pc= 2.0 × 10−3 Mpc, then the average distance to a galaxy
along the line of sight is:
1
l= π(2.0×10−3 Mpc)2 (1Mpc−3 )
= 8.0 × 104 Mpc
Therefore, we might expect to see a galaxy everywhere on the sky. How-
ever, this distance is large enough that the expansion and geometry of the
universe must be taken into account.

3
ASTR/PHYS 4080: Introduction to Cosmology Spring 2019

3. A hypothesis once used to explain the Hubble relation is the tired light hy-
pothesis. The tired light hypothesis states that the universe is not expanding,
but that photons simply lose energy as they move through space (by some
unexplained means), with the energy loss per unit distance being given by the
law
dE
= −KE , (3)
dr
where K is a constant.

(a) Show that this hypothesis gives a distance-redshift relation that is linear
in the limit z  1.

Solution

If photons lose energy according to equation 3, then

dE
E
= −Kdr.
Integrating the left side of the equation from E0 to E(r), and the right
side from 0 to r yields
E(r) = E0 e−Kr
The energy and frequency of a photon are related by Planck’s constant,
E = hν (or E = hf in the textbook’s nomenclature), so the relation
between energy and wavelength is given by:
hc
E= λ

Therefore, in the “tired light hypothesis” the wavelength of light increases

with distance as λ(r) = λ0 eKr . This equation implies a relation between
the redshift, z, and the distance traveled:
λobs − λem λ(r) − λ0
z= = = eKr − 1 (4)
λem λ0

In the limit Kr << 1 (hence, z << 1), eKr approximates as 1 + Kr + . . .

and equation 4 reduces to z = Kr to first order.

(b) What must the value of K be in order to yield a Hubble constant of

H0 = 70 km s−1 Mpc−1 ?

Solution

4
ASTR/PHYS 4080: Introduction to Cosmology Spring 2019

Since z = Hc0 r (see Ryden eq. 2.8), the “tired light hypothesis” requires
K = Hc0 = 4380
1
Mpc−1 , where I have used the value for Hc0 from page 17
of Ryden. More explicitly, K = 2.3 × 10−4 Mpc−1 .

4. In a Steady State Universe, H0 is predicted to be constant with time and

matter must be created at a constant rate to maintain a constant density. In
this model, matter must be created in the solar system at the same constant
rate. The solar system is virialized (gravitationally bound) and does not follow
cosmological expansion. Instead, whatever is created in the solar system stays
in the solar system and contributes to the mass of the sun. For a Hubble
constant of H0 = 70 km s−1 Mpc−1 , assume that all matter created in a sphere
with radius 1000 AU was used to create the sun. How many years does it take
to generate enough material in this steady state universe to form the sun?

Solution

From the Ryden textbook (equation 2.23),

dM
ṀSS = = 3ρ0 H0 V (5)
dt
kg
where ρ0 = 3×10−27 cm 3 (the critical density). We want a volume corresponding

to R = 1000 AU = 1.496 × 1014 m radius to account for the mass of the sun,
M = 1.98 × 1030 kg. So the time it takes to form the sun is the mass divided
by dM
dt
M 4π
t = dM = M (3ρ0 H0 R3 )−1 (6)
dt
3
Using values for the solar mass, critical density, Hubble constant, and approx-
imate radius used to create the sun, t = 2.2 × 1014 Gyr. This is a very long
time compared to the age of the sun, indicating that structure formation is
very difficult to explain with a steady state universe.