Notes

Contents

1 The Decay Rate of Neutrino at The Tree Levers 2

2 Cosmological Standard Model 5

3 Boltzmann Equation 8

1
1 The Decay Rate of Neutrino at The Tree Levers

For Type-I seesaw, the SU (2)L × U (1)Y gauge-invariant Yukawa-interaction and mass terms about leptons
are

e R + 1 NRc MR NR + h.c. .
−L′lepton = LL Yl HER + LL Yν HN (1)
2
LL is left-spinor lepton-field. Yl and Yµ are 3-order complex matrices called Yukawa matrices. H is a
tow-component complex scalar field.
 +  0∗ 
ϕ ϕ
H= H ≡ iσ 2 ⋆
H = (ϕ− ≡ ϕ+∗ ) . (2)
ϕ0 −ϕ−
e

The mass matrix of right-handed Majorana neutrinos MR can be diagonalized by a unitary transformation
† ∗
UM , UM MR UM = Mc, and the eigenvalues of M c are the mass of Ni (Ni are the mass eigenstaes of
right-handed neutrinos, i = 1, 2, 3). Meanwhile, we diagonalize the Yukawa matrix of charged leptons,
†
UlL Yl UlR = Ybl . In this case, the Yukawa term of right-handed Majorana neutrinos is transformed
 
e R = Lα U † Yν UM ∗ e R Ni ≡ Lα Yν′ e R Ni = (Lα ) Yν′ ϵ H ∗P N ,



LL Yν HN lL HP αi
HP a αi ab b R i
(3)
αi
†
e R = N i Hb ϵab Yν′ ∗ PL (Lα ) = N i ϵab Yν′ ∗ Hb PL (Lα ) ,



h.c. = LL Yν HN αi a αi a (4)

where ϵ is a antisymmetric matrix (ϵ12 = −ϵ21 = 1). α is the flavor index of leptons (α = e, ν, τ ) and
a, b = 1, 2 represent “up, down” scalar of SU(2) tow-component sates.

According to the above equations, the decay of right-handed Majorana neutrinos are Ni → (Lα )a + Hb
and its CP conjugation N̄i → L̄α a + H̄b , where N̄ , L̄, H̄ are antiparticles. And there are N̄i = Ni , H̄b = Hb∗ .

The Feynman diagrams of the decay of right-handed Majorana neutrinos are

Firstly, we calculate Ni → (Lα )a + Hb . In tree diagrams level, the amplitude M of scattering matrix
elements satisfies
4
iM (Ni → (Lα )a + Hb ) (2π) δ (4) (pi − pa − pb )
(5)
= ūr (pa )ϵab Yν′ αi PR us (pi ) (2π) δ (4) (pi − pa − pb ) .

4

Considering “averaging initial states and summing final states”,

1X 1 1 h i
|M|2 = ϵ2ab | Yν′ αi |2 ūr (pa )PR us (pi )ūs (pi )PL ur (pa ) = ϵ2ab | Yν′ αi |2 tr (p



/αa + mαa )(p
/i + mi )PL
2 r,s 2 2
" #
1 2 ′
2 µ ν µ ν
1 − γ5 .
= ϵab | Yν αi | tr (pαa )µ (pi )ν γ γ + mi (pαa )µ γ + mαa (pi )ν γ + mαa mi
2 2
=ϵ2ab | Yν′ αi |2 (pαa · pi + mαa mi )

(6)

The differential decay rate dΓ of Ni decaying to different flavor is

!
X 1 Z d3 pαa 1 Z d3 pb 1 1X 4
dΓ (Ni → Lα + H) = 3 0 3 0 |M|2 (2π) δ (4) (pi − pαa − pb ) . (7)
2mi (2π) 2pαa (2π) 2pb 2 r,s
a,b

2
In a rest frame, there are
q q
pµi = (mi , 0) , pµαa = ( m2l + q 2 , q) , pµb = ( m2H + q 2 , k) . (8)
(9)
At the spherical coordinates, the differential volume is
d3 q = dq qdθ q sin θdϕ = q 2 sin θdq dθ dϕ . (10)

The integral of δ (4) (pi − pαa − pb ) can make dpαa and d3 pb disappear,
X 1 Z d3 q 1
Z
d3 k 1 4 (4) i |M|2
dΓ (Ni → Lα + H) = 3 3 (2π) δ (m − Ea,q − E b,k , q + k)
2mi (2π) 2Ea,q (2π) 2Eb,k 2
a,b
X 1 Z d3 q 1 i |M|2
= 2 δ(m − Ea,q − E b,−q ) ,
2mi (2π) 4Ea,q Eb,−q 2
a,b
Z 2
X 1 Z π δ(q − q0 ) |M|2
Z 2π
q dq 1
= dθ sin θ dϕ 2
2mi 0
a,b 0 (2π) 4Ea,q Eb,−q q a,q +Eb,−q 2
E
Ea,q Eb,−q
(11)
where q0 satisfies
q q
mi = m2l + q02 + m2H + q02 . (12)
So we can get
X 1 Z π Z 2π 2 ′
2
q0 ϵab | Yν αi |
q 
2 2
dΓ (Ni → Lα + H) = dθ sin θ dϕ 2 mi ml + q0 + ml
2mi 0 0 (2π) 4mi
a,b

X ϵ2ab | Yν′

2 q
|

αi 2 2 . (13)
= q0 ml + q0 + ml
8πmi
a,b

| Yν′ αi |2

q 
2 2
= q0 ml + q0 + ml
4πmi
At the high energy scale of early universe, the gauge symmetry is not broken, so particles of standard
model are massless. So we can get
′
2
| Yν′ αi |2


ml ,mH →0 | Yν αi | 2
dΓ (Ni → Lα + H) = q0 = mi . (14)
8πmi 16π
Summing the flavor of leptons
X mi
∗ X mi
†
Yν′ αi Yν′ αi = Yν′ iα Yν′ αi



dΓ (Ni → L + H) =
α
16π α
16π
. (15)
mi  ′ † ′ 
= Yν Yν
16π ii

Nextly, we calculate the CP conjugated decay, Ni → L̄α a + H̄b . The CP conjugated amplitude M is
4
iM Ni → L̄α a + H̄b (2π) δ (4) (pi − pa − pb )




∗ . (16)
= v̄ s (pi )ϵab Yν′ αi PL v r (pa ) (2π) δ (4) (pi − pa − pb )
4

3
As same as, we can get
1X 1
∗ 1
∗ h i
|M|2 = ϵ2ab | Yν′ αi |2 v̄ s (pi )PL v r (pa )v̄ r (pa )PR v s (pi ) = ϵ2ab | Yν′ αi |2 tr (p/i − mi )(p
/αa − mαa )PR
2 r,s 2 2
" #
1 2 ′
∗ 2 µ ν ν µ
1 + γ5 .
= ϵab | Yν αi | tr (pαa )ν (pi )µ γ γ − mi (pαa )ν γ − mαa (pi )µ γ + mαa mi
2 2
2 ′
∗ 2
=ϵab | Yν αi | (pαa · pi + mαa mi )
(17)

The decay rate is

∗
| Yν′ αi |2
q 

2 2
dΓ Ni → L̄α + H̄ = q0 ml + q0 + ml
4πmi
. (18)
′
∗ 2
ml ,mH →0 | Yν αi |
= mi
16π

m  
dΓ Ni → L̄ + H̄ = i Yν′ † Yν′ . (19)
16π ii

4
2 Cosmological Standard Model

The cosmological principle assumes that the observed universe is so homogeneous and isotropic on large scales,
so the spacetime should be maximally symmetric space. Therefore, one use Friedmann-Robertson-Walker
(FRW) metric
" #
2
dr  
ds2 = dt2 − a2 (t) + r2 dθ2 + sin2 θdϕ2 , (20)
1 − kr2
where a(t) is scale factor and k = 0, ±1 meaning

k=0 Flat Universe Euclidean Geometry

k=1 Closed Universe Spherical Geometry .
k = −1 Opened Universe Hyperbolic Geometry

The characteristics of observed universe supports k = 0, on large scale, so we will use it on the below. The
Einstein equations are
Gµν = 8πGN Tµν , (21)
where Gµν is the Einstein Tensor and Tµν is the stress-energy tensor. The simplest model is a perfect fluid
having energy density ρ(t) and pressure P (t)
T µν = P δ µν + (ρ + P )uµ uν = diag(ρ, −P, −P, −P ) , (22)
where uµ is the 4d velocity field of fluid, and in a comoving coordinate, uµ = (1, 0, 0, 0). The comoving
coordinate system is relatively static to the fluid. For the perfect fluid, the components of Einstein equations
are
k
3H 2 + 3 2 = 8πGN ρ , (23)
 a
k
− 3H 2 + 2Ḣ + 2 = 8πGN P . (24)
a
Using the Christoffel symbols and Riemann tensor, from (23) and (24), we can get
ρ̇ + 3H(ρ + P ) = 0 , (25)
where H is Hubble parameter defining
ȧ(t)
H(t) ≡ . (26)
a(t)
The most useful toy-model of cosmological fluid satisfies
P = wρ . (27)
Some values of w

Period of Universe Domination w

Early Radiation w = 1/3 ρa4 =constant .
Adolescent Matter w=0 ρa3 =constant
Adult Cosmological Constant w = −1 ρ =constant

5
Considering k = 0, w ̸= −1, from (23),(25), we can get
2
  3(1+w)
t
a(t) = a0 (28)
t0
 −2
t
ρ(t) = ρ (29)
t0
2 1
H= (30)
3(1 + w) t

Considering the cosmological temperature T , we want to ger T (t). So we need to know the relation
H(T ) of Hubble parameter H and temperature T . In early universe, the cosmological temperature is
very high and elastic scattering of particles is so strong. Therefore, early universe is a perfect thermal
equilibrium system and all particles are thermal equilibrium sates at every instant. The number density n,
energy density ρ and pressure P satisfy
d3 p
Z
n=g 3 f (p) , (31)
(2π)
d3 p
Z
ρ=g 3 E(p)f (p) , (32)
(2π)
d3 p |p|2
Z
P =g 3 f (p) , (33)
(2π) 3E(p)
where E 2 = p2 + m2 and g is internal degrees of freedom. f (p) is
( 1
E/T Fermi-Dirac statistics,
f (p) = e 1 +1 (34)
E/T Bose-Einstein statistics,
e −1

Using the integral of energy, we rewrite the above

1
(E 2 − m2 ) 2
Z ∞
g
n= dE E , (35)
2π 2 m eE/T ± 1
Z ∞ 2 2 21
g 2 (E − m )
ρ= dE E , (36)
2π 2 m eE/T ± 1
3
(E 2 − m2 ) 2
Z ∞
g
P = dE . (37)
6π 2 m eE/T ± 1
When universe is early, T ≫ m.
( ζ 3
gB 32 T (Boson)
n= π , (38)
ζ3 3 3
gF 2 T × 4 (Fermion)
π
2
(
gB π30 T 4 (Boson)
ρ= 2 , (39)
gF π30 T 4 × 78 (Fermion)
P = ρ/3 . (40)
where ζ3 ≈ 1.202 is the Riemann zeta function of three. Defining a new internal degrees of freedom g∗ (T ),
 4  4
X Ti 7 X Ti
g∗ (T ) = gi + gi . (41)
T 8 T
Bosons Fermions

6
T is the photon temperature and Ti is the i-specie temperature which can be different from the photons.

T Energy scale gF and Ti gB g∗

Gluons(Gaµ )
: 8 × 2(polarization)
u, c, t, d, s, b : 6 × 3(color)×4(spinor) ± 0
W , Z : 3 × 2(polarization)
T ≥300GeV e, µ, τ : 3 × 4(spinor)
Photons: 2(polarization)
νe , νµ , ντ : 3 × 2(Wely spinor) 106.75
Higgs: 4(H ± , H 0 , H 0∗ )
Gauge symmetry
All Ti are T
All Ti are T
e : 4(spinor)
νe , νµ , ντ : 3 × 2(Wely spinor)
100MeV≥ T ≥1MeV Photons: 2(polarization), T 10.75
All Ti are T
νe , νµ , ντ : 3 × 2(Wely spinor)
T ≪ 1MeV Photons: 2(polarization), T 3.36
4 1/3


Tν = 11 T
.

So we can rewrite ρ(T ) as

π2 4
ρ = g∗ T (42)
30
And then, Hubble parameter H(T ) is
s
4π 3 GN g∗ 2
H= T (43)
45

Using the relation H(t) of Hubble parameter H and time t, the function of the photon temperature T
depending t is
 1/4
5 1
T (t) = √ (44)
(1 + w)2 π 3 GN g∗ t

Defining cosmological entropy density s as

S ρ+P ρ
s≡ = = (1 + w) (45)
V T T
In the relativistic limit, we obtain

(1 + w)π 2
s= g∗S T 3 (46)
30
where
 3  3
X Ti 7 X Ti
g∗S (T ) = gi + gi . (47)
T 8 T
Bosons Fermions

7
3 Boltzmann Equation

Before beginning this section, we firstly need to understand some things. The scattering can change
the number of particles for each species by inelastic collision, but also bring all particles into thermal
equilibrium by elastic collision. In the early universe, the temperature is very hight, so the universe can
reach thermal equilibrium quickly. The dynamical force going thermal equilibrium is much greater than the
ability of particle creation and annihilation, so the expansion of the early universe is “quasi-static process”.
Therefore, For the universe, the thermodynamic characteristics are given by cosmological dynamics and
equilibrium statistical mechanics, but the particles characteristics are analyzed by scattering of particles
and equilibrium statistical mechanics.

Considering the particles A, we want to find the evolution of A depending time. Because of the expansion
of universe, the physical volume is V = V0 a3 (t) and V0 is the volume in comving coordinate. Defining the
number density nA of particles A, there is
d(nA V )
= V (ṅA + 3HnA ) , (48)
dt
where H is Hubble parameter. Considering a decay process A → B + C, the change of number density is
Z 3
d pA
−V 3 f (pA )nA Γ(A → B + C) ,
(49)
(2π)

where Γ is the decay rate and f (pA ) is distribution function of A. In the early universe, the temperature
is so hight, so we can assume that all particles are thermal equilibrium and satisfy Maxwell-Boltzmann
distribution
E−µ
f (pA ) = A0 e− T , (50)
R d3 p
where A0 is normalization parameter 3 f (p) = 1 and µ is chemical potential representing the change of
(2π)
particles numbers. It is emphasized that the degrees of degeneracy can be ignored in distribution functions,
because we can bring degeneracy into the decay rate (scattering matrix element M), which is equal to
an average over initial sates and a sum over the final states. We can define neq A as number density when
particles A are thermal equilibrium,
Z 3
d pA −E/T
neq
A = 3e . (51)
(2π)

and then rewrite f (pA ) as

1 − ET
f (pA ) = e , (52)
neq
A

We must emphasis the fact that the f (pA ) seems to not depend chemical
P potential
P µ, but when inelastic
scattering is enough quick and system is thermodynamic equilibrium, µin = µout , such as µA = µB +µC
for A ↔ B + C. At rest system, the decay rate Γ e is
Z 3 Z 3
1 d pB 1 d pC 1 4
Γ=
e
3 3 |M(mA → pB + pC )|2 (2π) δ (4) (mA − pB − pC ) . (53)
2mA (2π) 2EB (2π) 2EC
In any other frame, we only replace mA with EA and add p. Importantly, the Fermi blocking factors
should be considered. Assuming that particles B, C are Fermions and not considering the temperature of

8
particles, the distribution functions of B, C are 1. However, when particles B, C have finite
 temperature

and are thermodynamic equilibrium, the number of B, C having energy E ∼ E + ∆E is NB fB (∆pB )3
 
or NC fC (∆pC )3 . In phase space, there are (∆p)3 V = ℏ3 and ℏ = 1, so

N n E
N f (∆p)3 = f = nf = eq e− T (54)
V n
 
e−E/T nB /neqB particles B occupy E ∼ E + ∆E energy-states, which will block scattering to create B, so
 
the states which B created by the decay can take up are 1 − N f (∆p)3 called ”Fermi blocking factors”.
 
Similarly, if B is Boson, ”Bose stimulated factors” is 1 + N f (∆p)3 . Therefore, for the decay process
A → B + C, we rewrite the change of number density as
 
Y Z d3 pi 1 nA − ET A

nB − ET B

nC − ET C

−V   eq e A 1 − eq e B 1 − eq e C

i=A,B,C
(2π)3 Ei nA nB nC (55)
4
× |M(pA → pB + pC )|2 (2π) δ (4) (pA − pB − pC )
P P P P
For scattering A + B → C and reversed scattering C → A + B, the general form of the A
number is
Z ! !
YZ YZ X X
ṅA + 3HnA = dΠA dΠB dΠC (2π)4 δ (4) (pA + pB − pC )
B C B C
 !" # 2
 Y Y X X
× n C fC (1 ± nB fB ) (1 ± nA fA ) M( pC → pA + pB ) (56)

C B C B

!" # 2
Y Y X X 
−nA fA nB fB (1 ± nC fC ) M(pA + pB → pC )

B C B C

where scattering matrix elements M must consider an average over the initial states and a sum over the
final states. If the final state is Boson, we should choose 1 + nf , and if the final state is Fermion, choose
1 − nf . Noting that dΠi , fi of specie i are

d3 pi 1
dΠi ≡ (57)
(2π)3 Ei
1
fi ≡ eq e−Ei /Ti (58)
ni

In early universe, the temperature of all species

P are equal
P to the temperature of photons T , and all inelastic
scattering processes are enough quick µA + µB = µC . When the temperature is so high, all particles
are relativistic satisfying ⟨E⟩ = 3T , and Bose condensation or Fermi degeneracy are absent. What’s more,
n is created in initial period which is smaller than neq (n < neq ), and then n is thermodynamic equilibrium
in next period (n ≈ neq ). Therefore, 1 + nf ⪅ e−3 ≪ 1, the blocking and simulated factors can be ignored,

9
For simply the expression of Boltzmann equations, we can define “Interaction Density” as
Z ! !
X X YZ YZ −EC /T
γ( pC → pA + pB ) = dΠA dΠB dΠC e
C B B C
2 (59)
4 (4)
X X X X
× (2π) δ (pA + pB − pC ) M( pC → pA + pB )
B C C B
Z ! !
YZ YZ
dΠA e−EA /T dΠB e−EB /T
X X
γ(pA + pB → pC ) = dΠC
B C B C
2 (60)
4 (4)
X X X X
× (2π) δ (pA + pB − pC ) M(pA + pB → pC )
B C B C

The Boltzmann equations are rewritten as

ṅA + 3HnA
( ! ! )
X Y nC X X nA Y nB X X (61)
= eq γ( pC → pA + pB ) − eq γ(pA + pB → pC )
nC nA neq
B
C C B B B C

10