Chapter 1

Introduction

This thesis describes a collection of works which study the cosmological signatures of
electromagnetic energy and high energy (& 100 GeV) neutrino/antineutrino injection
in high redshift Universe (z & 1000). We develop numerical codes for evolving high
energy particle cascades in the expanding Universe taking into account interaction of
high energy particles with the background particles (electrons, photons, nuclei, and
neutrinos/antineutrinos).
In this chapter, we give a general and brief overview of the standard model
of cosmology and the cosmological probes that are used in the subsequent chapters.

1.1 The standard model of cosmology

Our expanding Universe, being isotropic and homogenous on large length scales, can
be described by the Friedmann-Lemaitre-Robertson-Walker (FLRW) metric [1–7],

dr 2
" #
ds = −dt + a(t)
2 2 2
+ r (dθ + sin θdφ ) ,
2 2 2 2
(1.1.1)
1 − kr 2

where the speed of light is chosen to be unity, t is the proper time, a(t) is the scale
factor, (r, θ, φ) are the comoving spherical coordinates and k is the curvature of space
and can take values k=0, 1 and -1. For a spatially flat Universe like ours, k =0. The
scale factor captures the expansion history of our Universe which is normalized to
unity today, i.e. a(t 0 ) = 1 where t 0 is the age of the Universe. The wavelength of
light (λ) emitted at some time (t) in the past will be observed today at a wavelength
(λ 0 ) due to the expansion of the Universe, which is related to the scale factor by,
a(t 0 )/a(t) = λ 0 /λ. The redshift is defined as, z = (λ 0 − λ)/λ. Therefore, redshift can

3
be expressed as a function of scale factor as,

a(t 0 )
1+z = . (1.1.2)
a(t)

The evolution of the scale factor as a function of time can be calculated from Fried-
mann equation by [1, 2],
!2
ȧ 8πG
= ρtot, (1.1.3)
a 3
where G is the gravitational constant and ρtot is the total energy density of the
Universe. In the standard flat ΛCDM model [8], the total energy density of the
Universe consists of cold dark matter (ρcdm ), baryonic dark matter (ρb , combined
with cold dark matter called matter (ρm )), total radiation (ρr ), and dark energy
(ρΛ ). Eq. 1.1.3 can be rearranged to relate the total energy density of the Universe
and the Hubble constant, i.e.,
3H 2
ρc = , (1.1.4)
8πG
 
where H = ȧa is the Hubble constant. The evolution of energy density of various
components can be calculated from the continuity condition [1, 2],

ȧ
ρ̇i = −3 ( ρi + Pi ), (1.1.5)
a

where i is the index for matter, radiation or dark energy, and P is the pressure . The
pressure of each component can be related to its density through equation of state
parameter as,
Pi = wi ρi, (1.1.6)

where w for matter, radiation, and dark energy is 0,1/3, and -1 respectively. Using
Eq. 1.1.5, we see that ρm ∝ a−3 , ρr ∝ a−4 , and ρΛ is a constant with time. The
density parameter Ωi is defined by,

ρi
Ωi = . (1.1.7)
ρc

With this definition, eq. 1.1.3 can be rewritten as,

H (t) 2 = H02 (Ωm,0 a−3 + Ωr,0 a−4 + ΩΛ,0 ), (1.1.8)

where Ωi,0 s are the density parameters today and H0 = 67 kms−1 Mpc−1 [8] is the
Hubble parameter today. We can also define dimensionless Hubble parameter h ≡

–4–
 H0
.
100kms−1 Mpc−1
The above discussion assumes the Universe to be perfectly homogenous
and isotropic. In reality this is not so as we see structures which are fluctuations
on top of the homogenous and isotropic Universe. The seeds of these fluctuations
are given by the initial condition of the early Universe. The observation of these
fluctuations can be extremely accurately fitted by a six parameter model. These six
parameters are Ωb h2 , Ωc h2 , the angular scale of sound horizon as seen today (θ MC ),
amplitude of initial power spectrum (As ), power law index of power spectrum as a
function of wave number (ns ), and optical depth of reionization (τ)[8]. The energy
density of cosmic microwave background (CMB) photons was precisely measured
by Far-InfraRed Absolute Spectrophotometer (FIRAS) experiment onboard Cosmic
Background Explorer (COBE) [9, 10]. The background neutrino energy density can
be computed by knowing the CMB energy density and the neutrinos’ interaction
with the other standard model particles. The value of ΩΛ , in a flat Universe, is then
given by ΩΛ = 1 − Ωm − Ωr . The early Universe is radiation dominated until the
redshift, 1 + z eq = ΩΩmr , with redshift of radiation-matter equality being z eq ≈ 3200.
In the late Universe, dark energy starts to dominate when ΩΛ = Ωm (1 + zΛ ) 3 with
zΛ ∼ 0.3.

1.1.1 Neutrino energy density in the Universe and Ne f f

The total radiation energy density of the Universe in the radiation dominated era
can be written as,
π2
ρr = g ρ (T )T 4, (1.1.9)
30
where g ρ (T ) is the number of relativistic degree of freedom at temperature T. The
expression for g ρ (T ) can be written as,
X 7X
g ρ (T ) = g ρ,b + g ρ, f , (1.1.10)
b
8 f

where the sum is over the degenerate states of bosons (b) and fermions (f). The
factor 7/8 comes from the phase space integral of Fermi-Dirac distribution. When
all the standard model particles are relativistic g ρ (T )=106.75. Similarly, the total
number density of relativistic particles is given by,

ζ (3)
nr = gn (T )T 3, (1.1.11)
π2

–5–
where gn (T ) = b gn,b + 34 f gn, f . Once the temperature of Universe is below the
P P

mass (mi ) of a standard model particle i, the contribution of that particle to radiation
energy density is suppressed by e−mi /T .
The neutrinos/antineutrinos decouple from the background relativistic
plasma once the rate of weak interaction drops below the Hubble rate, which happens
to be around T ∼ MeV. At this temperature, in addition to neutrinos/anti-neutrinos,
only photons and electrons/positrons, out of all the standard model particles are
relativistic. After the neutrinos decouple from the relativistic plasma, their tem-
perature redshifts ∝ a−1 as they are still relativistic. The released energy from
electron-positron annihilation, which happens after neutrino decoupling, increases
the temperature of the CMB photons as compared to neutrinos/antineutrinos. The
two temperatures, in case of complete decoupling of neutrinos/anti-neutrinos before
electrons/positrons annihilation, are related by,
 4  1/3
Tν = Tγ, (1.1.12)
11

where Tν and Tγ are the temperatures of neutrinos/antineutrinos and CMB photons

respectively. The temperature of the CMB spectrum today is measured to be TCMB =
2.725K [11]. Therefore, the temperature of the neutrino background turns out to be
1.9 K. The contribution of neutrinos/antineutrinos to total radiation energy density
can then be parameterized by,

ρν
Neff =   4/3 , (1.1.13)
7
8
4
11 ργ

where ρ ν and ργ are the energy density of neutrinos/antineutrinos and CMB photons
respectively. If the decoupling of neutrinos is instantaneous, Neff = 3, as standard
model neutrinos/antineutrinos have 6 degree of freedom while photons have 2 po-
larizations. Since the neutrino decoupling is not instantaneous, some of the energy
from e− e+ annihilation leaks to the neutrinos making Neff = 3.046, which is slightly
higher than 3 [12–22], for a more recent calculation see [23]. The total relativistic en-
ergy density (or radiation energy density) affects the expansion rate of the Universe.
This makes Neff somewhat degenerate with the Hubble parameter. By requiring the
angular measure of the acoustic scale at the recombination epoch to be fixed (which
is precisely measured by Planck [8]), constraints on the Hubble parameter lead to
constraints on the total radiation energy density of the Universe at the recombina-
tion epoch i.e. on the Neff parameter. The current 2-σ constraint on Neff from CMB

–6–
anisotropy is 2.92±0.36 [8]. While our discussion has focused on the standard model
neutrinos, in general any relativistic energy density can contribute to Neff . The extra
neutrino species which are thermalized, in addition to the standard model neutrinos,
or other new relativistic particles such as dark radiation, can make Neff > 3.046
while addition of extra energy to photons after neutrino decoupling can decrease
the relative energy density in neutrinos with respect to the CMB and hence make
Neff < 3.046.

1.1.2 Big bang nucleosynthesis

Big bang nucleosynthesis (BBN) is the earliest reliable cosmological probe of the
Universe [24–27]. At temperatures above 1 MeV, weak interaction is in thermal
equilibrium. The ratio of number density of neutrons to protons is given by, n/p =
e−Q/T , where Q = 1.3 MeV is the mass difference between neutrons and protons.
Therefore, at T >> 1 MeV, number densities of neutrons and protons are equal. With
the expansion of the Universe, the rate of neutron-proton conversion drops below the
Hubble rate. Therefore, the abundance of neutrons freezes out and does not deplete
exponentially as given by the equilibrium solution. After freezeout, the neutrons
start to β-decay. The neutrons combine with the protons to produce deuterium,
n+ p ↔ D+γ, where D denotes deuterium. Since the photons outnumber the baryons
by a factor of 109 , deuterium starts to get produced only at T ∼ 0.1 MeV, when the
number density of high energy photons which can dissociate deuterium nuclei is
comparable to the number density of deuterium getting produced. By that time, the
ratio of number density of neutrons to protons is ∼ 1/7. Deuterium then combine with
protons to form helium-3 which can subsequently combine with another deuterium
to form helium-4. The reactions happen sequentially as the number densities of these
nucleons is low, so 3-body or higher processes are unlikely. As the Universe continues
expanding, the reaction rates of all these processes fall below the Hubble rate and
the number densities of these light elements get frozen.

1.1.3 Recombination

At T >> 1 eV, the Universe is completely ionized and electrons, protons, helium
nuclei, and the CMB photons are in thermal equilibrium. Even though the binding
energy of neutral hydrogen is 13.6 eV, they cannot form at the epoch when temper-
ature of the Universe is 13.6 eV. This is because photons outnumber the baryons by

–7–
more than 9 orders of magnitude and there are too many energetic photons which
can ionize the neutral hydrogen. Instead, neutral atoms start to form when the
temperature of the Universe is ∼ 1 eV. The recombination of electrons with protons
and helium nuclei happens in three stages as the binding energy of doubly ionized
helium, singly ionized helium, and hydrogen is 54.4 eV, 24.6 eV, and 13.6 eV re-
spectively. The number density of helium is approximately 8 percent of the number
density of hydrogen, therefore, the phase of hydrogen recombination is the most im-
portant. The problem of cosmological recombination has been studied in great detail
beginning with the first calculations of [28, 29]. The authors, in these first seminal
calculations, approximated hydrogen to be a three level atom where the three levels
are the ground state, the first excited state, and the continuum where the electron
is free. The higher excited states are assumed to be in thermal equilibrium. An
electron cannot recombine directly to the ground state as the emitted energetic pho-
ton can ionize another neutral hydrogen. Electron recombining to the first excited
state can jump to the ground state from 2p state. However, during the initial phase
of recombination, when the number densities of free electrons and protons is high,
the optical depth of this transition is very high. It is highly likely that the electron
in 2p level gets photo-ionized by the CMB photons before jumping to the ground
level. Therefore, two-photon transition (2s→1s) becomes as important as Lyman-
alpha transition (2p→1s). As the Universe expands and the recombination proceeds,
it becomes increasingly unlikely for the protons and electrons to recombine due to
dilution of their number densities. Therefore, the abundance of free electrons freezes
out.
The need for multilevel calculations without the assumption of thermal
equilibrium between excited levels of hydrogen was emphasized by [30] for precision
calculations needed by the CMB experiments in the 21st century. There has been
tremendous progress since then, culminating in the fast effective multilevel recom-
bination codes [31–38]. Most of the complicated dynamics of recombination can
be captured by the effective 3-level atoms originally proposed by [28, 29], with the
equations suitably modified by a fudge factor [30] or a fudge function [34]. In these
calculations, only the first excited levels with the principal quantum number n = 2 of
hydrogen and helium are resolved, in addition to the ground state and the continuum
or the ionized state of the atom. The evolution of population of any excited state

–8–
can be written as,

dx Hi X
= (R ji x Hj − Ri j x Hi ), (1.1.14)
dt j

where x Hi = nHi /nH , nHi is the number density of hydrogen with recombined electrons
in excited level i, nH is the total (ionized + neutral) hydrogen number density, and R ji
is the rate coefficient for transition of an electron from state j to i. Similar expressions
can also be written for helium. Accurate calculation of abundance of free electrons
as a function of redshift is critical as the scattering of the CMB photons with these
free electrons affect the shape of the CMB anisotropy spectrum.
In this work, we have used publicly available codes Recfast++ and Cos-
moRec [34]. CosmoRec solves exactly for 500 shells of hydrogen and 20 shells of
helium and provides output in the form of ionization fraction of free electrons and
fraction of hydrogen/helium atom with recombined electrons (x Hi, x Hei as defined
above) upto user defined number of excited levels, as a function of redshift. Rec-
fast++ is a much a simpler code which keeps track of the ground level, the first
excited level and the free electron fraction. To take into account the effect of higher
excited levels, it is calibrated with CosmoRec through a pre-computed redshift
dependent fudge function. Since, Recfast++ tracks only three levels of hydrogen
and helium, it is much faster than CosmoRec. In both of these codes, there is
a scope to add extra ionizations and excitations to hydrogen/helium recombination
history from exotic energy injections such as dark matter decay and annihilation.
We have modified these codes to study energy injections from dark matter decay
and primordial black hole evaporation (More discussion on this in Sec. 4.2).
Even after recombination, the CMB photons still scatter with the residual
charged particles, to keep them at the same temperature as the CMB until z ∼ 200,
when Compton scattering becomes increasingly inefficient due to dilution of number
densities of these particles. Thereafter, the temperature of baryons redshifts as ∝ (1+
z) 2 until the epoch of reionization at around z ∼ 10−20. During reionization, baryons
collapse to form structures with the emission of energetic photons which ionizes the
neutral baryons. Therefore, the ionization fraction of baryons and abundance of
free electrons start to increase. The interaction of the CMB photons with these free
electrons is captured by the optical depth of reionization (τ) though the accurate
evolution of free electrons during reionization requires detailed numerical simulations.

–9–
1.2 High redshift cosmological probes of new physics
The incredibly precise measurement of the cosmic microwave background (CMB)
anisotropy power spectrum allows us to not just measure the 6 parameters of the
standard ΛCDM cosmological model with high precision but also study extensions
to it and constrain new physics beyond the standard model of particle physics. One
such scenario can be an extended dark sector rather than just one dark matter
particle. The CMB is sensitive to new physics in the dark sector such as dark matter
self interactions [39, 40], dark matter-dark radiation interactions [41, 42], decay of
long-lived unstable particles [43, 44], evaporating primordial black holes [45, 46], and
annihilation of dark matter to standard model particles [47–49]. The physics in the
dark sector can affect CMB both gravitationally [42, 50, 51] and electromagnetically
[43, 44, 46, 47]. In particular, there are many particle physics motivated scenarios
where dark matter can annihilate or decay to standard model particles [43, 52–59],
see [60, 61] for reviews.

1.2.1 Possible energy injection scenarios in the high redshift Universe

The dark matter direct detection experiments, looking for Weakly Interacting Mas-
sive Particles (WIMPS) have so far not yielded any detection and the allowed param-
eter space for WIMPs is gradually being ruled out. This has motivated physicists
to look beyond the simplest thermal WIMP models. One way around the direct
detection constraints is Super-WIMPs (SWIMP), where instead of producing the
usual lightest stable particle in the theory in a beyond standard model theory like
supersymmetry or Kaluza-Klein models, the thermally produced WIMP is the next-
to-lightest particle which then decays to the lightest stable particle of the theory
with the right abundance needed for dark matter [55, 56]. For a review of dark mat-
ter models, see [60]. We can escape the direct detection constraints if the lightest
particle, which would be the dark matter today, has much weaker interactions with
the standard model particles i.e. is a SWIMP. We could in general have additional
unstable particles which were produced in the early Universe, with lifetime of few
years to hundreds of thousands of years, which would decay not into SWIMPs +
standard model particles but completely into standard model particles. More gen-
erally, we would like to constrain energy injection in the form of different standard
model particles over the entire history of the Universe and look for deviations from
standard cosmology as a probe of beyond standard model physics. Another interest-
ing example of energy injection in the form of energetic photons and other particles

– 10 –
is from evaporating primordial black holes [62], from accreting primordial black holes
[63–65], and cosmic strings [66, 67]. We note that there are examples of long lived
composite particles in the standard cosmology. During the big bang nucleosynthesis,
tritium (3 H) and beryllium (7 Be) are produced which decay to stable isotopes of
helium (3 He) and lithium (7 Li) at redshifts of z7Be = 3 × 104 and z3H = 2.5 × 105
respectively injecting energy into the CMB [68]. Unfortunately the abundance of
these elements in our Universe is too low for the resulting spectral distortions to be
observable.

1.2.2 Primordial black holes and cosmic strings

Primordial black holes (PBH) have recently gathered attention as a candidate of dark
matter [69] in light of the discovery of gravitational waves from the merger of black
holes with the mass of tens of solar mass [70]. Regardless of whether PBHs form a
dominant component of dark matter today, evidence for their existence at any time
in cosmic history, in any mass range, would give great insight into the initial condi-
tions and, in particular, the initial density fluctuations of the Universe (for a recent
summary of these constraints, see [71]). PBHs are formed in the radiation domi-
nated era when the radiation pressure is not able to resist the gravitational collapse
in overdense regions [72–75]. Mass of the black hole, produced in this scenario, is of
the order of horizon mass [62, 74]. Due to the formation at different epochs, mass of
PBHs can vary from Planck mass relics to ∼ 1010 times heavier than the mass of the
Sun [71]. Although PBH formation with extended mass spectrum is likely [76, 77],
we will restrict ourselves to monochromatic PBH mass function in this thesis so as
not to restrict ourselves to a particular model. Since PBHs are formed due to early
Universe small scale fluctuations, their abundance today puts constraints on the ini-
tial power spectrum on small scales [78, 79]. Therefore, deriving accurate constraints
on the allowed abundance of PBHs through various cosmological probes is a subject
of great interest today. There are constraints on accreting solar mass black holes
in the early Universe from CMB spectral distortions and CMB anisotropy power
spectrum [63, 65, 80], but we will not consider them in this thesis.
Cosmic strings are one dimensional objects which carry energy and appear
as topological defects due to symmetry breaking at phase transitions. In the early
Universe, they can form string networks and stretch across the Hubble horizon [81,
82]. There are high energy theories in which cosmic strings carry superconducting
current i.e. current flows without any resistance [83]. These superconducting strings

– 11 –
can radiate electromagnetic and gravitational energy which can be parameterized
by the string tension and current inside the strings [66, 67]. This electromagnetic
radiation can impact CMB spectrum [84] and reionization [85]

1.2.3 Modification to light element abundances from electromagnetic en-

ergy injections

Injection of energetic electromagnetic particles as well as neutrinos (anti-neutrinos)

gives rise to a particle cascade after interaction with the background particles. These
energetic particles modify the abundance of the light elements produced in the BBN,
freeze-out electron fraction after recombination and give rise to spectral distortions
in the CMB. Energetic electromagnetic particles, above photo-dissociation threshold
of the respective nuclei, can dissociate light elements produced in the BBN and
change their abundances. From the measured abundance of these light elements
and their theoretical prediction from the standard cosmology, we can constrain the
energy injection scenarios from the unstable particles. The BBN constraints on
decaying dark matter have been previously studied in [86–92]. The authors of [88–
92] calculate the change in abundance of light elements by solving the high energy
particle cascades assuming the rate of interaction of high energy particles with the
background particles to be much higher than the Hubble rate.

1.2.4 CMB spectral distortions from electromagnetic energy injections

In the standard ΛCDM cosmology, the cosmic microwave background (CMB) spec-
trum is given by the Planck spectrum to a very good approximation. Since the
ratio of the number density of photons to the number density of baryons is of the
order of 109 , recombination dynamics is controlled by the energetic photons in the
exponential tail of the Planck spectrum while the bulk of the photons in the CMB
pass through the recombination epoch unimpeded. If there is some mechanism in
the high redshift Universe (before the recombination epoch) which injects energy
into the tightly coupled baryon-photon fluid, it will leave its imprints in the spectral
distortion of the CMB.
Without any energy injection and ignoring the adiabatic cooling of the
background electrons [28, 29, 93, 94], the CMB photons and the background electrons
are in thermal equilibrium. The CMB spectrum is given by the Planck spectrum.
Energy injection without the addition of photons to the CMB gives rise to spectral

– 12 –
distortions as the CMB can not relax to a modified Planck spectrum without the ad-
dition of extra photons below z < 2×106 . In almost all previous works, the distortion
is assumed to be of thermal nature i.e. all injected energy, irrespective of the energy
of individual particles, ends up in heating the background electrons instantaneously.
These heated electrons boost the CMB photons through non-relativistic Compton
scattering, producing y-distortion (also known as Sunyaev-Zeldovich effect) instan-
taneously or µ-type distortion depending upon the redshift at which energy was
injected [93, 95–100]. The scattering of photons by thermal electrons redistributes
the energy of the photons modifying the photon spectrum and this process is called
Comptonization. If some energy is injected at sufficiently high redshift (z & 2.0×105 ),
the distorted spectrum can relax to Bose-Einstein spectrum with a constant chemi-
cal potential after a sufficient number of collisions between background photons and
electrons. It can not relax to a Planck spectrum since Compton scattering is a pho-
ton number conserving process. A y-type distortion is created at low redshifts when
the Comptonization process is not efficient enough for the photon spectrum to relax
to a Bose-Einstein distribution. It was emphasized in [100] that, there is an extended
epoch in the early Universe, corresponding to the redshift range 104 < z < 2 × 105 ,
in which an initial y-type distortion cannot relax to a Bose-Einstein spectrum but
evolves to an intermediate-type distortion (i-type), the shape of which depends upon
the redshift of energy injection. At z & 2 × 106 , photon-production process such as
double Compton scattering becomes efficient such that the CMB can partially relax
to a Planck spectrum. The efficiency with which spectral distortions in µ era can
relax to a Planck spectrum can be captured in an analytic window function called µ-
2.5
visiblity function. The expression for µ-visiblity function is given by ζ (z) = e−(z/zµ )
with z µ = 2 × 106 [96, 101, 102].
In Fig. 1.1, we plot the spectral distortion shapes for one-time energy
injections at different redshifts. The physics can be described as follows. The mag-
nitude of the energy injection (10−5 ) seems small, but it is small only when it is
compared to the CMB. The number of electrons are 9 orders of magnitude smaller
compared to the number of CMB photons. So, the electrons react immediately to
this energy injection with increase in temperature. The electron temperature is given
by [104, 105],
(n + n2 )x 4 dx
R
Te
= R , (1.2.1)
T 4 nx 3 dx
where, Te is the electron temperature, T is the background tempearture with T =
2.7(1 + z)K and n is the photon occupation number. The temperature increase

– 13 –
 ν(GHz) 100 1000

y=10-54

δIν(10-22Wm-2Hz-1ster-1)
1.5 z=2×104
z=3×104
1 z=5×10
z=7×1045
0.5 z=1×10
0

-0.5 z=2×1055
z=3×10
-1 µ=5.6×10-5

1 x 10

Figure 1.1: Comptonization of y-distortion to µ-distortion. This plot is reproduced

from [103].

10
(∆Te/T)×105

1
104 105
redshift (z)

Figure 1.2: Evolution of electron temperature (Te ) as a function of redshift for

one-time energy injection at z = 2 × 105 .

immeditaely after energy injection is ∆T e

T of electron immediately after the energy
Einj
injection is 5.4y where y= 41 ECMB is the magnitude of the y-distortion. Then equi-
librium between the CMB photons and the electrons is broken. The only way a
new equilibrium is again reached is if the photons redistribute their energy and at-
tain a Bose-Einstein distribution and electrons cool to the temperature of the Bose-
Einstein spectrum thus obtained. The electron temperature at equilibrium is given
∆TCMB
by ∆TT = TCMB =2.56y. The evolution of electron temperature is plotted in Fig. 1.2.
e

The final CMB Bose-Einstein spectrum has a non-zero chemical potential.

The measurement of the CMB monopole spectrum in 1990s by FIRAS
onboard COBE satellite [9, 10] remains the best constraint on the deviation of the
CMB monopole spectrum from a blackbody until today. The COBE-FIRAS data

– 14 –
has been used by numerous authors to constrain many exotic as well as standard
model energy injection processes in the early Universe (e.g. [56, 106, 107], see [108]
for a review).

1.2.5 Modification to recombination from electromagnetic energy injec-

tion

Energetic electrons, positrons, and photons, injected by new physics, deposit their
energy in the background baryon-photon plasma by heating the background electrons
and by ionization and excitation of neutral atoms. A fraction of energy escapes as
low energy photons below 10.2 eV (rest frame Lyman-alpha threshold) [47, 109]. In-
creased ionization rate of neutral hydrogen and helium due to energy injection around
recombination results in higher freeze-out number density of residual free electrons
after recombination compared to the standard recombination [28, 29]. Increased
number of scatterings of the CMB photons with the free electrons damp the tem-
perature anisotropy while giving a boost to polarization signal [43, 44, 47, 110, 111].
Precision measurement of the CMB anisotropy spectrum therefore puts constraints
on the amount of electromagnetic energy that can be injected during the epoch of re-
combination and hence on the parameters of the new physics such as the annihilation
cross-section of dark matter, fraction of decaying dark matter, lifetime of decaying
particles [49, 112, 113], and abundance of primordial evaporating black holes as a
function of black hole mass [46].

1.2.6 Brief description of previous works

Energy deposition by sub-keV electrons to hydrogen and helium gas as a function of

ionization fraction has been studied in [114, 115]. Monte Carlo simulation of higher
energy electrons including inverse Compton scattering (IC) was done in [116]. In
these calculations, the high energy (& keV) electrons are evolved until their energy
drops down to ∼eV, after depositing most of their energy to background particles,
at which point the rest of their energy is deposited as heat. Keeping track of the
evolution history from injected energy to ∼ eV energy for each injected particle makes
these calculations computationally expensive. An alternative recursive solution to
the above problem was provided in [47, 117, 118], who also included relativistic
processes of electron-positron pair production and photon-photon elastic scattering.
In this method, the pre-computed energy deposition history of lower energy particles

– 15 –
is used to compute energy deposition of higher energy particles. Using this recursive
method, the authors in [112] have calculated the constraints on fraction of decaying
dark matter (with respect to total dark matter) for decay to monochromatic electron-
positron and photon pairs as a function of dark matter mass (m X ) and lifetime (τX )
for τX & 1013 s (or corresponding decay redshift z X . 1200) using Planck 2015 [119]
CMB anisotropy power spectrum data. Similar constraints have been provided in
[113], where the authors have used the publicly available results of energy deposition
fractions from [120] to constraint particle lifetimes τX & 1012 s (z X . 4500). The
authors in [113] also provide an effective on-the-spot ansatz, which means that the
energy injected at a particular redshift is immediately deposited at that redshift.
They absorb the beyond on-the-spot corrections into an effective energy injection
history function. Their results show that the constraints on decaying dark matter
abundance with this on-the-spot ansatz agrees well with the full calculations of [112]
for τX & 1013 s but not for lower lifetimes.

1.2.7 Brief description of the work presented in this thesis

In this thesis, we study signatures of electromagnetic and neutrino energy injec-

tion on the CMB anisotropy power spectrum, distortion of the CMB spectrum, and
abundance of primordial elements. We develop a versatile code which solves the
high energy particle cascade evolution taking into account the scattering processes
of high energy particles with the background baryon-photon plasma before recom-
bination and partially neutral gas in the post-recombination era, in the expanding
Universe. CMB spectral distortion, CMB anisotropy, and abundance of light ele-
ments are the high redshift cosmological probes that we consider in this thesis. CMB
spectral distortion and BBN provide constraints for energy injection at redshifts less
than 2 × 106 while CMB anisotropy power spectrum provides stronger constraints
for redshifts z < 10000. We use publicly available COBE-FIRAS [9], Planck 2015
and 2018 [8, 119] data and theoretical and observational abundances of light ele-
ments [27, 121, 122] to constrain the energy injection scenarios. For concreteness,
we consider energy injection scenarios such as dark matter decay to standard model
particles, primordial black hole evaporation to the standard model particles, and
decay of cosmic strings.

– 16 –
1.3 Description of chapters

Chapter 2 We describe the physics of particle cascades triggered by the injection of

high energy electrons (positrons), photons, and neutrinos (antineutrinos). We detail
the cross-sections for the collision processes that are used in this thesis. We also
discuss the numerical techniques used to obtain particle spectra and evolve them
with time.

Chapter 3 In this chapter, we obtain the CMB spectral distortion shapes for high
energy monochromatic electron/positron and photon injection by explicitly evolving
the particle cascade. We show that the spectral distortion shapes from relativistic
particle injection can be very different from non-relativistic y, µ and i-type distor-
tions. We predict new spectral distortion shapes which are functions of energy and
type (electron/positron or photon) of injected particle.

Chapter 4 In this chapter, we first calculate the efficiency of energy deposition

from injected high energy electromagnetic particles to the background baryon-photon
plasma and subsequent modification to the recombination dynamics. Then we de-
scribe the physics of the modification to the abundance of light elements produced
in the BBN from energetic particle injections above the photo-dissociation threshold.

Chapter 5 In this chapter, we obtain constraints on dark matter mass and life-
time of decay from CMB spectral distortions, CMB anisotropy power spectrum and
abundance of the light elements, using the up-to-date cosmological data. We show
that the CMB spectral distortion constraints are function of dark matter mass as
opposed to independent of mass which was claimed in literature. We show that CMB
anisotropies are sensitive to energy injections at redshifts as high as z ≈ 10000, pro-
viding stronger constraints compared to light element abundances and CMB spectral
distortions.

Chapter 6 In this chapter, we provide CMB anisotropy, CMB spectral distortions,

and light element abundance constraints on primordial black holes. We take into
account the effect of hadronization and subsequent secondary electromagnetic par-
ticles from the evaporation of black holes. This strengthens previous CMB spectral
distortion constraints by a factor of 50. We also obtain the most accurate CMB
anisotropy constraints on evaporating primordial black holes by evolving the particle

– 17 –
cascades.

Chapter 7 In this chapter, we provide CMB spectral distortion constraints from

the decay of cosmic strings.

Chapter 8 In this chapter, we, for the first time, evolve the particle cascades from
high energy (> 100 GeV) neutrinos (anti-neutrinos) in the pre-recombination Uni-
verse. We calculate the neutrinos’ relative energy density with respect to the CMB
photons at recombination which is parameterized as Neff . We show that CMB spec-
tral distortions and light element abundance provide much stronger constraints on
Neff compared to the present day constraints from CMB anisotropies for these high
energy neutrinos.

Chapter 9 In this chapter, we end with conclusions and future directions.

– 18 –
1.4 Brief description of Appendix

A.1 Summary of the cross-section for electrons/positrons used in this work.

A.2 Summary of the cross-section for photons.

A.3 Description of our use of public code PYTHIA.

A.4 Expressions for y, µ and temperature shift distortions.

A.5 Plots for changes to CMB anisotropy from energy injection due to dark matter
decay.

A.6 ΛCDM parameters for various dark matter injection scenarios.

A.7 Comparison of the ΛCDM parameters for Planck-2015 [123] and 2018 [124]
data.

– 19 –