Can Fusing Dark Matter Explain the

Preponderance of Supermassive Black Holes?

Moira Gresham, Liz Meador, Brooke Neupert
August 06 2019

1 The Big Problem

In April 2018, the first ever picture of a black hole was taken and released to
the world, thanks to the efforts of astronomers and computer scientists at the
Event Horizon Telescope. This specific black hole is the supermassive black
hole (SMBH) in the center of the galaxy M87, an active black hole with the
mass of 6.5 ∗ 1010 solar masses, or 1.29 ∗ 1040 kg, making it the most massive
one we have observed.

Observations of numerous other galaxies, including our own, show that SMBH
are present in most, if not all galaxies. These SMBH generally have masses
between 106 and 1010 solar masses. If these black holes are active, it means
that they are accreting matter and powering what is known as an “active
galactic nuclei” (AGN).

AGN emit electromagnetic radiation across a wide spectrum; anything from

ultraviolet to gamma ray wavelengths. Quasars are a type of AGN that are
some of the most luminous objects in the observable universe, and they re-
quire a supermassive black hole to shine. In fact, it is the observation of
quasars that leads us to the central problem of this research.

It is important to point out that as we are observing these objects, we are

looking back in time. This is because the light from these sources takes time
to travel to earth, so the light that we are seeing from distant objects gives
us the image of something as it was long ago. Things that are farther away

1
are farther back in time, and vice versa.

We have observed highly luminous quasars as far back as 12.8 billion years
ago. Because we know that quasars need a supermassive black hole to power
them, there must be one there. But for one to have grown from a stellar mass
from gas accretion is impossible - the universe at that time is too young to
have grown such a large object.

The other, slightly more disturbing, fact is that we have not observed any
intermediate mass black holes (104 − 105 solar masses). If SMBH have grown
from stellar mass black holes (<103 solar masses), we should be able to find
some black holes in this mass range. But we haven’t.

We are then left with a great mystery: How did SMBH form?

This question has been agonized over by astronomers and astrophysicists

for many years, and many hypotheses to explain the formation and evolu-
tion of SMBH have come forward. This paper seeks to give an overview of
many of these, with a focus on how collisional models of self-interacting dark
matter may play a significant role.

In Section II we outline the dynamics of the galaxies around these SMBH

and their accretion disks, followed by an explanation of different formation
and evolution hypotheses in Section III. Section IV shows our models for col-
lisions between two particles in orbit around a black hole, before concluding
and discussing future directions for this research in Section V.

2 Galactic Dynamics and Disks

In an effort to better understand the processes that allow for the formation of
supermassive black holes it is essential to go back to the early universe. Our
universe, perhaps one of many, has gone through immense changes over its
13.8 billion year existence. The birth of our universe is believed to originate
from the fantastic phenomenon known as inflation. Inflation describes the
early exponential expansion of space. At the time of inflation there existed
slight variations (quantum fluctuations) in adjacent regions known as under-
densities and over-densities. These inhomogeneities were basically blown up

2
at the time of inflation and lead to a lot of the larger-scale structure we see
today by the force of gravity.

Post-inflation at earlier times in its history, our universe was much smaller,
denser, and hotter–it was essentially a hot quark soup. Past the time of
Big Bang Nucleosynthesis and the Cosmic Microwave Background, around
at an epoch near the Dark Ages, baryonic matter would have begun falling
into the structures created by dark matter. Constituents of our universe like
the stars, galaxies, you, and me, are all considered to be baryonic matter
and only consist of a mere five percent, compared to the abundance of dark
matter at approximately 27 percent. Many physicists and astronomers are
searching for dark matter in many forms as it appears to be an extremely
necessary component in understanding the evolution of our universe. Dark
matter was present throughout the early growth of our universe and played
a crucial role in the formation of large and small scale structures.

As the universe continued to expand and cool after the Dark Ages, our
galactic structures formed through gravitational instability. Gravitational
instability is a process where the overdense regions continue to acquire more
mass to form gravitationally bound objects and the underdense regions lose
that matter.The formation of galaxies is largely believed to have assembled
in a bottom-up formation where smaller galaxies are constructed first and
merge to become clusters. The evolution of these galaxies is indeed a chaotic
process.

The formation process varies for different types of galaxies and still is an
area of astrophysics that needs further clarification on specific evolutions,
disk formation, angular momentum transport, and more. The type of galaxy
that is of particular interest in this article are spiral or disk galaxies which
generally have a central bulge and a circular flow of material surrounding
the center. These galaxies start with a primordial cloud of gas that collapses
under its own gravity, where the time taken for a cloud to collapse is pro-
portional to √1% (where % = density of a gas cloud). As gas dissipates to form
the disk, star formation occurs in the inner disk where the surface density
is the highest and then propagates outwards as the disk continues to settle.
This disk formation occurs because the orbit of lowest energy for a given
angular momentum forms a circle. Most of today’s thin-disk stars occupy

3
nearly circular orbits because the gas that they originated from lost almost
as much energy as possible. This compares to thick disk stars which were
born from clouds that yielded less of their energy but still formed somewhat
flat systems. Over time the amount of stars being produced in the galaxy
decreases as the available gas in the galaxy diminishes. This of course may
change based on interactions of the galaxy and other celestial objects, for in-
stance the absorption of a dwarf galaxy would likely continue star formation
and inflate the bulge. Within the bulge of many spiral galaxies is a suspected
supermassive black hole that correlates to galaxy’s evolution.

3 The Formation of Supermassive Black Holes

Understanding the formation of supermassive black holes is a process that
isn’t well known in the field of cosmology and astrophysics. The issue that
we focus on in this article has many theories which attempt to resolve the
gap in understanding of supermassive black holes.

3.1 Non-DM Models

Before we discuss how dark matter may be involved in the formation and
evolution of SMBHs, it is important to look at hypotheses of how they may
have formed without it. Although it is thought that they should have formed
from smaller black holes, it is also possible that SMBH simply emerged into
the universe in its early stages. In this section, we will expand upon these
hypotheses and how plausible they may be.

3.1.1 Seed Models

In these models, supermassive black holes grow by accreting matter onto a
smaller black hole, known as a seed. What sets these two hypotheses apart
is the size of the black holes that they grow from. While light seeds may be
smaller, around the size of a stellar black hole, heavy black hole seeds may
be much larger, up to 106 solar masses.

Light Seed Models Light seeds seem to be the obvious starting point for
the growth of supermassive black holes. The collapse of early-universe Pop
III (no-metallicity) stars would leave them behind, and they could accrete

4
matter to become larger. The problem with this is that there is a limit to
how fast a black hole can grow. Unfortunately for this model, that rate is not
fast enough to produce the observed SMBH by the time we have observed
them.

The fastest that a black hole should be able to accrete matter is at the
Eddington Rate. The Eddington rate is determined by the Eddington limit,
which is how luminous (and massive) a body can become without upsetting
the balance between the radiation pressure outwards and the gravitational
force inwards. The Eddington rate is how fast matter can be accreted to
allow the star to shine at the Eddington limit. In order for a light seed to
form a SMBH, it would need to accrete at super-Eddington rates.

Super-Eddington accretion is possible under the right set of circumstances,

which would have been most likely in the early universe, when these light
seeds are forming. However, these conditions are still very difficult to come
by, as they need to be very specific.

First it is necessary that a large amount of Pop III stars collapse into seeds
of around 100 solar masses. These stars would then have to accrete at the
Eddington limit for their whole lifetime in order to grow to 109 solar masses
by the time we observe them at. This is unlikely, as it requires the seeds to
have a constant supply of dense gas for its entire lifetime. It has also been
shown that light seeds are likely to accrete at sub-Eddington rates, making
Eddington accretion itself difficult.

One way that light seeds can grow, however, is through mergers with each
other. Although it is unlikely that a SMBH will form through mergers alone,
if a light seed becomes big enough through a combination of mergers and
accretion, it may become massive enough to accrete as a heavy seed.

Overall, although the growth of light seeds seems like the most straight-
forward way to form SMBH, it doesn’t show very much promise as a model
due to the rarity of the conditions that are required for these seeds to form
the 109 solar mass black holes that we have observed at high redshift.

5
Heavy Seed Models Heavy seed models show much more promise as a
pathway for SMBH formation than light seeds. As black holes with masses
of around 104 − 106 solar masses, they are able to accrete faster than lighter
seeds due to their mass and would be able to accrete enough matter to form
a SMBH at the early time we observe them at.

The caveat with these heavy seeds is how they formed. They are not formed
by the collapse of Pop III stars, as light seeds are, but through the collapse
of dense gas clouds that have resisted undergoing fragmentation into stars.
Rather, they can be formed by the direct collapse of a dense gas cloud. These
black holes are known as “direct collapse black holes (DCBH).

In some cases, these clouds may briefly form a supermassive star (SMS)
that collapses into a black hole very quickly. In order for this to occur, there
must be specific conditions allowing for the collapse of a halo without it frag-
menting into Pop III stars. One of these conditions is that the cooling of a
halo must be prevented, in order for it to have a larger maximum mass for
collapsed objects. This cooling can occur as a result of the dissociation of
H2 molecules.

Photons with a specific energy, known as Lyman-Werner photons, coming

from a halo that has undergone Pop III fragmentation have the right amount
of energy to break hydrogen molecules apart, but not enough energy to ionize
the atomic hydrogen itself. The dissociation of hydrogen gas molecules heats
the halo, preventing fragmentation. When the gas in the halo does collapse,
it will collapse into a more massive object.

The issue with this model is that the two clouds, the one that had undergone
Pop III fragmentation giving off LW photons and the one forming the black
hole, would need to have very specific conditions for timing, evolution, and
separation in order to form the massive seed. It is also necessary that the
cloud has low metallicity in order to prevent fragmentation.

Although heavy seeds are DCBH in most models, it is also possible that
these heavy seeds grew in other ways. Galaxy mergers and high baryonic
streaming velocities also facilitate the growth of heavy seeds.

Baryonic streaming velocity is the relative velocity of baryons to dark mat-

6
ter particles. High streaming velocities may prevent the formation of H2
molecules in the halo as well as Lyman-Werner photons, preventing frag-
mentation.

The growth of heavy seeds seems to be the most popular and agreed upon
pathway for SMBH formation and growth at this time due to the plausibility
of it accreting enough matter to form a SMBH by the early times we observe
them at.

3.1.2 Primordial Black Holes

Another popular model for SMBH formation is that they are simply primor-
dial black holes. Primordial black holes were formed in the pre-stellar epoch,
before stars were formed. This is possible due to inhomogeneities that al-
lowed some denser regions in the early universe to gravitationally collapse
and form black holes.

This theory is popular because it eliminates the need for accretion onto a
smaller seed. These black holes may have accreted somewhat but having
formed at an already giant mass, there would be no question of how they
grew to such large masses in the early universe.

Many observations of black holes that challenge the model of accretion also
support the possibility that they are primordial black holes. We keep discov-
ering larger SMBH at early redshifts, making the accretion models less and
less reassuring. One especially disturbing finding is a SMBH at a redshift of
7.5 ( 13 billion years ago). As we find SMBH farther into the early universe,
the need for more massive seeds and higher accretion rates, which become
less probable as we go farther back.

It is possible, however, that the black hole seeds themselves are primordial
black holes that then accrete to become SMBH.

This hypothesis also has a very strong argument for it, as an opposition
of the seed models. However, it is possible that a combination of both seed
accretion and primordial formation are responsible for the observations of
SMBH at high redshifts.

7
3.2 Dark Matter Models
Many theories which focus on the formation of supermassive black holes uti-
lize dark matter as either a component that radiates and/or influences the
growth of central black holes.

The current cosmological model assumes that dark matter is fairly colli-
sionless, which means it does not interact with baryonic matter. For many
of the theories connecting dark matter and supermassive black holes there
is an assumption that part of the dark matter may interact with itself, re-
ferred to as self-interacting dark matter (SIDM). This is an assumption that
is possible to make as we have not directly seen dark matter. An instance
where self-interacting dark matter was considered was after the observation
of the Bullet Cluster. This was the collision of two galaxy clusters where the
intergalactic gas in each cluster interacted and released heat and detectable
X-ray emission.When astronomers mapped the distribution of matter they
noted its presence alongside the galaxies not the intra-cluster matter. In this
case the dark matter passing through untouched is what is most noticeable.
It makes sense to then ask the question of what dark matter interacts with,
if not baryonic matter, which leads to self-interacting dark matter.

There have been less theories consisting of dark matter that are entirely
self-interacting, and more which consider two-component dark matter, where
there is a subdominant component that is self interacting and the rest is col-
lisionless and cold. These theories may be referred to as PIDM (Partially
Interacting Dark Matter) or uSIDM (Ultra-Strongly Interacting Dark Mat-
ter), since the strength of self-interaction and the amount of dark matter that
interacts with itself are adjustable for different models. These two-component
dark matter theories expect accretion to occur at higher redshifts in order to
form the observed quasars with 109solar masses and assume the dissipation
of the dark matter via collisions.

The collisions of dark matter particles will have an effect on the efficiency
of the formation of a supermassive black hole. Under the assumption that
the gas of PIDM within the halo has only elastic interactions, core collapse
would continue until instability results and a black hole seed is formed. A
thick core of PIDM would then surround the formed black hole and would
contribute to its growth. For supermassive black hole formation the elastic

8
collisions require a large cross section, which compares to the dissipative (or
inelastic) scenario where cosmologists expect the direct collapse to be accel-
erated. There would be an increase in the amount of heat flowing from the
inner halo to the outer hastening the collapse of the dark matter halo possibly
through the emission of dark photons (or formation of dark matter bound
states) by these dissipative interactions. It is during this cooling process that
enough mass must be transferred to the supermassive seed producing a mass
that can evolve through normal accretion processes to match current quasar
observations. Factors that contribute to the growth of the initial seed are de-
pendent on the fraction of dark matter that is self interacting, the strength
of the interaction, and the degree of elasticity. For instance, in order for
an elastic collision to eventually produce a black hole seed that is able to
evolve to current observations, it requires a greater strength of interaction
and fraction of SIDM. This process of dissipative dynamics is expected to
be even more efficient for perfectly inelastic collisions, since these types of
collisions may still produce the necessary observation at lower interaction
strengths and smaller fractions of SIDM. Dissipative dark matter has also
been considered in other contexts such as the formation and collapse of a
dark disk similar to our baryonic one. Overall, the possibility of dissipative
dark matter represents an interesting approach to supermassive black hole
formation.

4 Collision Models
4.1 Why are our Models Important?
In the realm of astrophysics we know there are a vast number of collisions
which occur in the universe. These collisions occur at atomic levels and on
galactic scales between varying forms of matter. These collisions generate
and create many of the small and large structures we see today. An interest-
ing phenomenon which appears within galactic structures in our universe are
supermassive black holes. These giants live in the centers of most galaxies
and have a complex birth, one cosmologists have yet to understand. Their
birth may very well involve the elusive yet ever-present, dark matter.

As an investigation of dark matter’s potential influence on the formation

of supermassive black holes, our team has created a series of models aimed

9
at gaining a qualitative understanding of possible dark matter interactions.
This document includes a collection of different inelastic and elastic collision
models. Each model contains a statement about the parameters of the colli-
sion and specifically examines the relationship between energy and angular
momentum.

4.2 Inelastic Collisions

Model I Two masses are in a circular orbit around a central mass M. The
mass of the first object is twice that of the second.When they collide, it is
an inelastic collision. What happens to the orbit? What is the periapse and
apoapse?

Relevant Equations:
mb = 2ma
mc = mb + ma = 3ma
p = mv
E = KE + P E
GM m
PE =
r
L=r×p
L = mrv sin θ
s
GM
v=
r

The important things to realize about this problem are that momentum is
conserved and angular momentum is conserved, but kinetic energy is not.
It is also important to note that due to the masses being in a circular orbit,
their velocity will be equal in magnitude, although opposite in direction.

va = −vb

The first step in this problem is to find what we can from conservation of
momentum.

10
 pa + pb = p c
ma va + mb vb = (ma + mb )vc
ma va + 2ma vb = (3ma )vc
va = −vb
ma (−vb + 2vb ) = 3ma vc
ma vb = 3ma vc
vb = 3vc

From conservation of momentum, we find that the velocity of the new mass
immediately after the collision is one-third of the original velocity.

Next we look at angular momentum, using the velocity relation from the
conservation of momentum. Since this is a circular orbit,

θ = 90◦

L = La + Lb
L=r×p
L = mrv sin θ
La = ma ri va
Lb = mb ri vb
La + Lb = Lc
ma ri va + mb ri vb = (ma + mb )rf vc
−ma ri vb + 2ma ri vb = (3ma )rf vc
ri vb (−ma + 2ma = (3ma )rf vc
ri vb = 3rf vc
ri (3vc ) = 3rf vc

11
 ri = rf

From this sequence, we find that ri is equal to rf , due to the fact that the
velocity is entirely tangential to the radius vector, and the collision will not
move the final mass any closer or farther away from the mass at that point.

Remembering that our goal is to find rmin and rmax , it may be surprising
to find that we already know rmax . In this collision, kinetic energy is lost,
as seen from the fact that the final velocity is a third of the original, and
therefore the orbit will shrink. What we have found, is that

rmax = rf

However, we still need to find rmin we should be able to do this by finding a

quadratic equation for r.

Lrmax = Lrmin
mrmax vc = 3mrmin vq
rmax vc = 3rmin vq
rmax vc
vq =
3rmin
s
GM
vc =
rmax
2 GM
rmax rmax
vq2 = 2
9rmin
rmax GM
vq2 = 2
9rmin

Now that we have an expression for the velocity at periapse, we can use the
conservation of energy (within the orbit, not the collision) to find an expres-
sion for r.

1 1 GM (3m) 1 GM (3m)
E = (3m)( vc2 ) − = (3m)(vq )2 −
2 3 rmax 2 rmin

12
 1 vc2 3GM 3 3GM
− = vq2 −
2 3 rmax 2 rmin
1 GM 3GM 3 rmax GM 3GM
− = 2
−
2 3rmax rmax 2 9rmin rmin
−17 GM 3 GM
= 2
6 rmax 18 rmin
−17 1 1 1 1
2
= 2
−3
6 rmax 6 rmin rmin rmax
17 2 1 2
0= rmin − 3rmax rmin + rmax
6 6

Using the quadratic formula:

q
−3rmax ± (3rmax )2 − 4( 17
6
)( 61 rmax
2 )
rmin = 17
3

Solving for this, we get:

rmin = rmax
1
rmin = rmax
17

Since we were essentially solving for any r where the θ = 90◦ , we found the
farthest and closest points in terms of the maximum radius, or the radius
at which the collapse occurred. From this result we can see that this orbit
becomes very eccentric, as the periapse is 17 times as close as the apoapse.

Model II (a)
The purpose of the next two models is to begin to define the new orbit of a
system after an inelastic collision in a much more general case. Here we are
not given any information about the masses of the two orbiting objects, or
the nature of their orbits; they may not be circular in this case.
If all we are given about these masses is their angular momentum and energy,
can we determine their orbits? We look at two problems: The first being a
general problem to find the orbit of one particle, the second involving both
particles in order to find the distance of the collision from the central mass.

13
What are rmin and rmax as an expression in terms of L and E for a sin-
gle particle?

Relevant Equations:
L=r×p
p = mv
GM m 1 p2
E=− +
r 2m
L = rp sin θ

Our first step was to solve for p:

GM m 1 p2
E=− +
r 2m
s
GM m
p= 2m(E + )
r

The next step was to put this in the angular momentum equation. It is
important to recall that because we are finding the apoapse and periapse,
θ = 90◦ :
L = rp sin θ
s
GM m
L = r 2m(E + sin θ
r
GM m
L2 = r2 (2m(E + ))
r
2GM m2
L2 = r2 (2mE + )
r
Our next step is to solve for r, finding a quadratic as we did in model 1.

L2 = 2mEr2 + 2GM m2 r
0 = 2mEr2 + 2GM m2 r − L2
Finally, we solve for r using the quadratic formula:
q
−2GM m2 ± 4G2 M 2 m4 + 4(2mE)(L2 )
r=
4mE
14
Model II (b)
What is the distance at which the collision occurs (rc ) in terms of the angular
momentum, energy, and collision angle of both particles?

Relevant Equations:
L1 = rc × p1
L2 = rc × p2
p1 = m1 v1
p2 = m2 v2
GM m1 1 p21
E1 = − +
rc 2 m1
GM m2 1 p22
E2 = − +
rc 2 m2
L1 = rc p1 sin θ
L2 = rc p2 sin θ

We can solve this equation in a similar way to the last problem. First, we need
to solve for p1 and p2 from their respective angular momentum expressions.

L1 = rc p1 sin θ1
L2 = rc p2 sin θ2
L1
p1 =
rc sin θ1
L2
p2 =
rc sin θ2

Our next step is to substitute these into the energy equations:

GM m1 1 p21
E1 = − +
rc 2 m1
GM m2 1 p22
E2 = − +
rc 2 m2

15
 GM m1 L21
E1 = − +
rc 2m1 rc2 (sin θ1 )2
GM m2 L22
E2 = − +
rc 2m2 rc2 (sin θ2 )2

Next we attempt to get another quadratic by multiplying everything by rc2 :

L21
rc2 E1 = −GM m1 rc +
2m1 (sin θ1 )2
L22
rc2 E2 = −GM m2 rc +
2m2 (sin θ2 )2
L21
0= rc2 E1 + GM m1 rc −
2m1 (sin θ1 )2
L22
0 = rc2 E2 + GM m2 rc −
2m2 (sin θ2 )2

Setting them equal and simplifying:

L21 L22
rc2 E1 + GM m1 rc − = rc
2
E2 + GM m 2 rc −
2m1 (sin θ1 )2 2m2 (sin θ2 )2

L21 L22
0 = (E1 − E2 )rc2 + (GM m1 − GM m2 )rc + (− + )
2m1 (sin θ1 )2 2m2 (sin θ2 )2

Putting this into the quadratic equation:

r
−L2 L22
−(GM m1 − GM m2 ) ± (GM m1 − GM m2 )2 − 4(E1 − E2 )( 2m1 (sin1θ1 )2 + 2m2 (sin θ2 )2
)
rc =
2(E1 − E2 )

Model III

The third of our inelastic collisions contains similarities to the first, but more
specifically focuses on the values post-collision. For model III, we consider
two colliding distinct particles each carrying energy (E1 , E2 ), mass (m1 , m2 ),

16
linear (p1 , p2 ), and angular momentum (L1 , L2 ) orbiting mass M . The mass
of the two particles combine since this is an inelastic collision (mf ), and are
characterized by their final energy (Ef ), linear (pf ), and angular momentum
(Lf ). In this case, a question we are particularly interested in asks, what the
radius of collapse will be? The radius of collapse will aid in outlining what
our new orbit will be.
The inital energy and angular momentum equations for particles one are
described by:

−GM m1 p2
E1 = + 1
rc 2m1

L1 = rc p1 sin(θ1 )

The initial energy and angular momentum equations for particle two are
described by:

−GM m2 p2
E2 = + 2
rc 2m2

L2 = rc p2 sin(θ2 )

The equations below describe the particle’s final energy and momentum after
the collision has taken place:

−GM mf p2f
Ef = +
rc 2mf

Lf = rc pf sin(θf )

Now that we’ve outlined the initial and final energy and momentum equa-
tions, we can begin to work the problem. Our first step is to utilize the final
energy equation to solve for the final linear momentum:

−GM mf p2
Ef = + f
rc 2mf

17
 s
GM mf
pf = 2mf (Ef +
rc

Using this new equation for linear momentum, we can substitute it into the
angular momentum equation and solve for the radius of collapse. Through
this method the radius of collapse remains in terms of the final energy, angular
momentum, mass, and angle.

Lf = rc pf sin(θf )

L2f = rc2 p2f sin(θf )2

GM mf
L2f = rc2 (2mf (Ef + ) sin(θf )2
rc

q
(−2GM m2f )(sin(θ)2 ) ± ((2GM m2f sin(θf )2 )2 ) − 4(2mf Ef )(sin(θf )2 ))(−L2f )
rc =
4mf Ef sin(θf )2

4.3 Elastic Collisions

Model IV
The fourth model we considered, the first of the elastic collisions was com-
prised of two particles each of mass, m, in initially circular orbits around a
center of mass M (where M » m). This compares to the inelastic collisions
since the kinetic energy in elastic collisions is conserved. In this system,
the particles enter with initial momentum (p1i , p2i ) at radii (r1i , r2i ). They
collide and essentially scatter off each other at an angle (θf ) each carrying
momentum (p1f , p2f ).
The following conservation laws describe the energy, linear, and angular mo-
mentum of the system:

Energy

p21i GM m p22i GM m p21f GM m p2f

2
GM m
− + − = − + −
2m ri 2m ri 2m r1f 2m r2f

18
Linear Momentum

mv1i + mv2i = mv1f + mv2f

Angular Momentum
ri pi sin(θi ) = r1f p1f sin(θf ) + r2f p2f sin(θf + 180)

Given that the initial orbits are circular the angular momentum can be ex-
pressed as:
s
GM
ri m sin(θi ) = r1f p1f sin(θf ) + r2f p2f sin(θf + 180)
ri

For this specific problem we are interested in determining the values of closest
and farthest approach. To solve for this we will only focus on one of the
particles so our resulting angular momentum relation reduces to:
s
GM
ri m sin(θi ) = rf p1f sin(θf )
ri

The value for farthest approach would scatter at an angle of pi/2 where
sin(90) = 1, so:
s
GM
ri m sin(θi ) = rf p1f
ri

Or
s
GM
ri m sin(θi ) = rf mvf
ri

Using the relationship between the initial angular momentum and final an-
gular momentum we can solve for the final velocity.
q
GM
ri m ri
sin (θi )
vf =
rf m

19
Our final and initial velocities can be substituted into the energy equation:
1 2 GM m 1 GM m
mvi − = mvf2 −
2 ri 2 r1f
q
GM
1 GM GM m 1 ri m ri
sin (θi ) GM m
m − = m( )2 −
2 ri ri 2 rf m rf

Through simplification of this equation we obtain a quadratic. Using the

quadratic formula we are able to determine the final radius as a function of
the initial radii with a theta dependency where:

rf = ri ± ri cos(θi )

4.4 Discussion
The models presented in the previous sections were explored in order to
achieve a better qualitative understanding of elastic and inelastic collisions
of two particles in orbit around a center of mass M.

From derivations of basic equations, we have found that the shape of the
orbits (periapse and apoapse) and the collision distance depend heavily on
the angular momentum, energy, and mass of the two masses, as well as the
center of mass.

When we apply this to dark matter collisions, we would first have to consider
the dark matter model we would want to analyze. If we follow the ΛCDM
model, then none of these scenarios would be plausible, as the dark mat-
ter would not interact in any way. However, models of self-interacting dark
matter would allow for such collisions to occur.

5 Future Work
There are two layers to the future directions possible with the information
presented here. The first consists of the way in which this type of work is
currently being improved upon and the second concerns the greater applica-
tions to the field of astrophysics and cosmology.

20
This type of work being carried out by observationalists and theorists is being
continued in numerous ways, specifically through: advanced numerical sim-
ulations, large scale surveys, observational indirect and direct experiments.
Our team’s future directions include possibly expanding our toy model to
include a smooth distribution of dark matter rather than a point mass at the
center. A significant part of our future understanding of the formation of
supermassive black holes requires working on advanced collision models and
analyzing those results.
By gaining a better understanding of the formation of supermassive black
holes we can determine a number of things in the field of astronomy and
physics. Supermassive black holes are an essential component in galactic
structure and impact greatly how large and small scale structure appears.
Probing formation methods has the ability to better describe how and why
those structures exist in greater detail and opens up the possibility of the for-
mation of a dark disk and other compact objects. Additionally, the current
cosmological model is in great favor of ΛCDM, and if the models for super-
massive black holes correctly predict the size of quasars at high mass using
ΛCDM it would contribute to a higher confidence with that model. There
are also applications in black hole thermodynamics that an understanding of
this particular problem would progress. Ultimately, having a better grasp on
the formation of supermassive black holes aids numerous sub-fields in astro-
physics, cosmology, particle physics, thermodynamics, and chemistry which
ends up advancing the collective knowledge of the scientific community.

21