Neutrinos

Lecture I: theory and phenomenology

of neutrino oscillations

The 2018 European School on High Energy Physics

Maratea
21-24 June 2018

Silvia Pascoli
IPPP - Durham U.
Shaped by the past, creating the future

mass
1 Institute for Particle
Physics Phenomenology
@Silvia Pascoli
 What will you learn from these lectures?

● The basics of neutrinos: a bit of history and the basic

concepts

● Neutrino oscillations: in vacuum, in matter,

experiments

● Nature of neutrinos, neutrino less double beta decay

● Neutrino masses and mixing BSM

● Neutrinos in cosmology
2 @Silvia Pascoli
 Today, we look at
● A bit of history: from the initial idea of the neutrino to
the solar and atmospheric neutrino anomalies

● The basic picture of neutrino oscillations (mixing of

states and coherence)

● The formal details: how to derive the probabilities

● Neutrino oscillations both in vacuum and in matter

● Their relevance in present and future experiments

(with some additional slides about experiments to
illustrate the latest results. Not discussed in detail.)
3 @Silvia Pascoli
 Useful references
● C. Giunti, C. W. Kim, Fundamentals of Neutrino Physics and
Astrophysics, Oxford University Press, USA (May 17, 2007)

● M. Fukugita,T.Yanagida, Physics of Neutrinos and applications

to astrophysics, Springer 2003

● Z.-Z. Xing, S. Zhou, Neutrinos in Particle Physics, Astronomy

and Cosmology, Springer 2011

● A. De Gouvea,TASI lectures, hep-ph/0411274

● A. Strumia and F. Vissani, hep-ph/0606054.

● Talks at the Neutrino 2018 conference (10 days ago)

4 @Silvia Pascoli
 A brief history of neutrinos
● The proposal of the “neutrino” was put forward
by W. Pauli in 1930. [Pauli Letter Collection, CERN]
Dear radioactive ladies and gentlemen,

…I have hit upon a desperate remedy to save the

… energy theorem. Namely the possibility that
there could exist in the nuclei electrically neutral
particles that I wish to call neutrons, which have
spin 1/2 … The mass of the neutron must be … not
larger than 0.01 proton mass. …in β decay a
neutron is emitted together with the electron, in
such a way that the sum of the energies of neutron
and electron is constant.

● Since the neutron was discovered two years later by

J. Chadwick, Fermi, following the proposal by E. Amaldi,
used the name “neutrino” (little neutron) in 1932 and
later proposed the Fermi theory of beta decay.
5 @Silvia Pascoli
 ● Reines and Cowan discovered the neutrino in
1956 using inverse beta decay. [Science 124, 3212:103]

● Madame Wu in 1956 The Nobel Prize

demonstrated that P is in Physics 1995
violated in weak interactions.
● Muon neutrinos were discovered in 1962 by L.
Lederman, M. Schwartz and J. Steinberger.

The Nobel Prize in

Physics 1988

6 @Silvia Pascoli
 ● The first idea of neutrino oscillations
was considered by B. Pontecorvo in 1957.
[B. Pontecorvo, J. Exp. Theor. Phys. 33 (1957)549.
B. Pontecorvo, J. Exp. Theor. Phys. 34 (1958) 247.]

● Mixing was introduced at the beginning

of the ‘60 by Z. Maki, M. Nakagawa, S. Sakata,
Prog. Theor. Phys. 28 (1962) 870,Y. Katayama, K. Matumoto, S. Tanaka, E.Yamada, Prog. Theor. Phys.
28 (1962) 675 and M. Nakagawa, et. al., Prog. Theor. Phys. 30 (1963)727.

● First indications of ν oscillations came from solar ν.

● R. Davis built the Homestake

experiment to detect solar
ν, based on an experimental
technique by Pontecorvo.
7 @Silvia Pascoli
 ● Compared with the predicted solar neutrino fluxes
(J. Bahcall et al.), a significant deficit was found. First
results were announced [R. Davis, Phys. Rev. Lett. 12 (1964)302 and R. Davis
et al., Phys. Rev. Lett. 20 (1968) 1205].

● This anomaly received further confirmation (SAGE,

GALLEX, SuperKamiokande, SNO...) and was finally
interpreted as neutrino oscillations.
SLAC Summer Institute on Particle Physics (SSI04), Aug. 2-13, 2004

8
φ µτ (10 cm -2 s -1)

SNO SNO
φ ES φ CC
7
6
6

5
SNO
4 φ NC
3 φ SSM
2
1
0
The Nobel Prize
0 1 2 3 4 5
6
6
φ e (10 cm -2 s -1 )
in Physics 2015
SNO, PRL 89 2002
8
gure 5: SNO’s CC, NC and ES measurements from the D2 O phase. The x- and y-axes are the inferred fluxes of electron
@Silvia Pascoli
 An anomaly was also found in atmospheric neutrinos.

● Atmospheric neutrinos had been observed by various

experiments but the first relevant indication of an
anomaly was presented in 1988 [Kamiokande Coll., Phys. Lett. B205 (1988)
416], subsequently confirmed by MACRO.

● Strong evidence was presented

in 1998 by SuperKamiokande
(corroborated by Soudan2 and
MACRO) [SuperKamiokande Coll., Phys. Rev. Lett.
81 (1998) 1562]. I consider this the start

of “modern neutrino physics”! The Nobel Prize

in Physics 2015
9 @Silvia Pascoli
 Neutrinos in the SM

● Neutrinos come in
3 flavours,
corresponding to the
charged lepton.

● They belong to SU(2) doublets:

electron

W
electron antineutrino
10 @Silvia Pascoli
 Neutrino mixing
Mixing is described by the Pontecorvo-Maki-Nakagawa-
Sakata matrix: |⇥ ⇤ = U i |⇥i ⇤
i Mass states
Flavour states g
|⇥ ⇤ = LCC U=i |⇥i ⇤ ⇧ (U ¯kL ⇥
+ h.c.)
k⇥ l L W⇥
which enters in thei CC interactions
2 k
g
LCC = ⇧ (U k ⇥¯kL l
⇥
L W⇥ + h.c.)
2 k
This implies that in an interaction with an electron, the
corresponding (anti-)neutrino will be produced, as a
superposition of different mass eigenstates.
Positron

W X
electron neutrino = Uei ⌫i
11
i
 ● 2-neutrino mixing matrix depends on 1 angle only.
The phases get absorbed in a redefinition of the
leptonic fields (a part from 1 Majorana phase).
⇥
cos sin
sin cos

● 3-neutrino mixing matrix has 3 angles and 1(+2)

CPV phases.
⇤ ⌅⇤ ⌅⇤ ⌅⇤ ⌅
i⌅e1
0 0 i⇥e
0 0
⇥¯1e ⇥¯µ2 ⇥¯⇤3
⇥
ei⇧ ⇧
e
0 ei⌅µ2
.
0 ⌃⇧ . CKM-
.
.
. e
. ⌃⇧ 0 ei⇥µ
e
0 ⌃⇧ µ ⌃
0 0 1 . type
. . 0 0 1 ⇤

Rephasing e e i(
the kinetic, NC and mass
e +⇥)
e
µ e i( µ +⇥)
µ terms are not modified:
⇥ e i⇥
⇥ these phases are unphysical.
12
 For Dirac neutrinos, the same rephasing can be done.
For Majorana neutrinos, the Majorana condition forbids
such rephasing: 2 physical CP-violating phases.
0 10 i
1
1 0 0 c13 0 s13 e
U = @ 0 c23 s23 A @ 0 1 0 A
i
0 s23 c23 s13 e 0 c13
0 10 1
c12 s12 0 1 0 0
@ s12 c12 0 A @ 0 ei↵21 /2 0 A
0 0 1 0 0 ei↵31 /2

For antineutrinos, U U

CP-conservation requires U is real = 0, ⇥

13
 Neutrinos oscillations: the basic picture
Contrary to what
expected in the SM,
neutrinos oscillate: after
being produced, they
can change their
flavour.
⌫1 ⌫1 ⌫1

muon electron
neutrino ⌫2 ⌫2 ⌫2 neutrino

Neutrino oscillations imply that neutrinos

have mass and they mix.
14
First evidence of physics beyond the SM.
 Neutrino oscillations and Quantum Mechanics analogs

Neutrino oscillations are analogous to many other

systems in QM, in which the initial state is a coherent
superposition of eigenstates of the Hamiltonian:

● NH3 molecule: produced in a superposition of “up”

and “down” states

● Spin states: for example a state with spin up in the z-

direction in a magnetic field aligned in the x-direction
B=(B,0,0). This gives raise to spin-precession, i.e. the state
changes the spin orientation with a typical oscillatory
behaviour.
15
 Neutrino oscillations: the picture
e

µ
X

Production Propagation Detection

Flavour Massive states Flavour

states (eigenstates of the states
Hamiltonian)
At production, coherent superposition of massive states:
| µ = Uµ1 | 1 + Uµ2 | 2 + Uµ3 | 3
16
 ⌫1 ⌫1 ⌫1

muon neutrino electron

neutrino
⌫2 ⌫2 ⌫2
Production Propagation Detection:
: e iE1 t projection over
| µ = Uµi | i 1
i 2 : e iE2 t
h⌫e |
3 : e iE3 t

As the propagation phases are different, the state

evolves with time and can change to other flavours.
17
 Neutrinos oscillations in vacuum: the theory

Let’s assume that at t=0 a muon neutrino is produced

| ,t = 0 = | µ = Uµi | i
i

The time-evolution is given by the solution of the

Schroedinger equation with free Hamiltonian:

| ,t = Uµi e iEi t
| i
i
In the same-momentum approximation:
E1 = p2 + m21 E2 = p2 + m22 E3 = p2 + m23

18 Note: other derivations are also valid (same E formalism, etc).

 At detection one projects over the flavour state as these
are the states which are involved in the interactions.
The probability of oscillation is
2
P( µ ⇥) = |⇥ ⇥ | , t⇤|
2
⇥
= Uµi U⇥⇥j e iEi t
⇥ j | i⇤
ij
2
⇥
= Uµi U⇥⇥i e iEi t

i
m2i
Typically, neutrinos are very relativistic: Ei p+
2p
2
⇥ m2
i 2Ei
= Uµi U⇥⇥i e t
2
i mi1
2
⇥ m2 m2
i 2E 1
i
= Uµi U⇥⇥i e t Exercise
Derive
i
19
 Implications of the existence of neutrino oscillations

The oscillation probability implies that

2
⇥ m2i1
P( ⇥) =
⇥
U 1 U⇥1 e i 2E L

i
● neutrinos have mass (as the different components
of the initial state need to propagate with different
phases)

● neutrinos mix (as U needs not be the identity. If

they do not mix the flavour eigenstates are also
eigenstates of the propagation Hamiltonian and they
do not evolve)
20
 General properties of neutrino oscillations
● Neutrino oscillations conserve the total lepton
number: a neutrino is produced and evolves with times
● They violate the flavour lepton number as expected
due to mixing.
● Neutrino oscillations do not depend on the overall
mass scale and on the Majorana phases.

● CPT invariance: P( ⇥) = P (¯⇥ ¯ )

2 2
⇥ ⇥
● CP-violation: ⇥
U i U⇥i e iEi t
= U⇥i U ⇥i eiEi t
i i
P( ⇥) ⇥= P (¯ ¯⇥ ) requires U = U ( = 0, ⇥)
21
 2-neutrino case
Let’s recall that the mixing is
⇥ ⇥ 2 ⇥
U 1 Ucos sin m2
P (⇥ ⇥ ⇥⇥ )⇥ = = ⇥1
21
⇥
⇥1 + U ⇥
2 ⇥2 e
U i 2E L
⇥⇥ sin cos ⇥2
m2
2
21
We compute the probability of oscillation
= cos sin cos sin e i L 2E

m2
2
21 2
P (⇥ ⇥ ⇥⇥ ) = U 21 ⇥1 +
U ⇥
U U ⇥
2 ⇥2 e i 2E 2 L
m21 m221
= cos sin 2
1 cos( L) i sin( L)
2E 2
2E
⇥ m2 ⇤
= 1cos 2sin cos sin
2i
m21 2E
e
21 L

= sin (2 ) 1 cos( L)
2 2E
2 2 2
m 2 m21 m21
=
= cos
sin2 (2sin
2
) sin2 ( 1 cos(
2 21
L) 2E L) i sin( 2E L)
⇥ 4E ⇤
1 2
m21
= sin (2 ) 1 cos(
2
L)
m2212 m221 [eV2 ]2E
L = 1.27 m 2 L[km] Exercise
4E sin2 (2 ) sin24( E[GeV]
= 21
L) Derive

22 4E
 1
P (⇥ ⇥⇥ ) ⇥ sin2 (2 )
2
P( ⇥) ⇥0

Thanks to T. Schwetz

23
First oscillation maximum
 Properties of 2-neutrino oscillations

● Appearance probability:
2
m
P (⇥ ⇥⇥ ) = sin (2 ) sin (
2 2 21
L)
4E
● Disappearance probability:
2
m
P (⇥ ⇥ ⇥ ) = 1 sin2 (2 ) sin2 ( 21
L)
4E
● No CP-violation as there is no Dirac phase in the
mixing matrix
P( ⇥) = P (¯ ¯⇥ )
● Consequently, no T-violation (using CPT):
P( ⇥) = P( ⇥ )
24
 3-neutrino oscillations
They depend on two mass squared-differences
2 2
m21 m31
In general the formula is quite complex
m2 m2
2
21 31
P( ⇥) = U ⇥
1 U⇥1 +U ⇥
2 U⇥2 e
i 2E L
+U ⇥
3 U⇥3 e
i 2E L

Interesting 2-neutrino limits

For a given L, the neutrino energy determines the
impact of a mass squared difference.Various limits are
of interest in concrete experimental situations.
2
m21
L 1
● 4E , applies to atmospheric, reactor (Daya
Bay...), current accelerator neutrino experiments...
25
 The oscillation probability reduces to a 2-neutrino limit:

m2
2
31
P( ⇥ ⇥) = ⇥
U 1 U⇥1 + U 2 U⇥2
⇥
+ U 3 U⇥3
⇥
e i 2E L

We use the fact that U 1 U⇥1 + U 2 U⇥2 + U 3 U⇥3 = ⇥

m2
2
31
= U ⇥
3 U⇥3 +U ⇥
3 U⇥3 e
i 2E L

m2
2
⇥ 2 31
= U 3 U⇥3 1+e i 2E L

The same we have encountered in the 2-neutrino case

2
2 m
= 2 |U 3 U⇥3 | sin2 ( 31
L) Exercise
4E Derive

26
 m231
● 4E L 1 : for reactor neutrinos (KamLAND).
The oscillations due to the atmospheric mass squared
differences get averaged out.
⇥
m221 L
P (¯
⇥e ⇥ ⇥¯e ; t) ⇤ 4
c13 1 sin (2
2
12 ) sin
2
+ 4
s13
4E

27 Thanks to T. Schwetz
 CP-violation will manifest itself in neutrino oscillations,
due to the delta phase. Let’s consider the CP-asymmetry:
P (⇥ ⇤ ⇥⇥ ; t) P (¯
⇥ ⇤ ⇥¯⇥ ; t) =
m2 m2
2
21 L 31 L
= U 1 U⇥1
⇥
+ U 2 U⇥2
⇥
e i 2E + U 3 U⇥3
⇥
e i 2E (U ⇤ U ⇥ )
m2
21 L m2
21 L
=U ⇥ ⇥
1 U⇥1 U 2 U⇥2 e
i 2E + U 1 U⇥1 U
⇥ ⇥
2 U⇥2 e(U ⇤ U ) + · · ·
i 2E ⇥
⌅ ⇥ 2
⇤ ⇥ 2
⇤ ⇥ 2
⇤⇧
m21 L m23 L m31 L
= 4s12 c12 s13 c13 s23 c23 sin sin
2
+ +
2E 2E 2E

● CP-violation requires all angles to be nonzero. Exercise**

Derive

● It is proportional to the sine of the delta phase.

● If one can neglect , the asymmetry goes to zero as
2
m21
we have seen that effective 2-neutrino probabilities are
28
CP-symmetric.
 Neutrinos oscillations in matter
● When neutrinos travel through a medium, they
interact with the background of electron, proton and
neutrons and acquire an effective mass.

● This modifies the mixing between flavour states and

propagation states and the eigenvalues of the
Hamiltonian, leading to a different oscillation probability
w.r.t. vacuum.

● Typically the background is CP and CPT violating, e.g.

the Earth and the Sun contain only electrons, protons
and neutrons, and the resulting oscillations are CP and
CPT violating.
29
 Effective potentials
2
Inelastic scattering and absorption processes go as GF
and are typically negligible. Neutrinos undergo also
forward elastic scattering, in which they do not change
momentum. [L. Wolfenstein, Phys. Rev. D 17, 2369 (1978); ibid. D 20, 2634 (1979), S. P.
Mikheyev, A. Yu Smirnov, Sov. J. Nucl. Phys. 42 (1986) 913.]

Electron neutrinos have CC and NC interactions, while

muon and tau neutrinos only the latter.

For a useful discussion, see E. Akhmedov, hep-ph/0001264; A. de Gouvea, hep-ph/0411274.

30
 We treat the electrons as a background, averaging over
it and we take into account that neutrinos see only the
left-handed component of the electrons.
⇥e · pe
⇥ē 0 e⇤ = Ne ⇥ē e⇤ = ⇥ve ⇤ ⇥ē 0 5 e⇤ = ⇥ ⇤ ⇥ē 5 e⇤ = ⇥⇥e ⇤
Ee

For an unpolarised at rest background, the only term is

the first one. Ne is the electron density.
The neutrino dispersion relation can be found by solving
the Dirac eq with plane waves, in the ultrarelativistic limit
⇤
E ⇥p± 2GF Ne

Strumia and Vissani

31
 The Hamiltonian
Let’s start with the vacuum Hamiltonian for 2-neutrinos
⇥ ⇥ ⇥
d | 1 E1 0 | 1
i =
dt | 2 0 E2 | 2

Recalling that | = U i| i , one can go

into the flavour basis i

⇥ ⇥ ⇥
d |⇥ ⇥ E1 0 |⇥1 ⇥
i = U U †
dt |⇥⇥ ⇥ 0 E2 |⇥2 ⇥
⇤ 2 2
⌅ ⇥
m
cos 2 m
sin 2 |⇥ ⇥
= 4E2 4E2
m
4E sin 2 m
4E cos 2 |⇥⇥ ⇥

We have neglected common terms on the diagonal as

they amount to an overall phase in the evolution.
32
 The full Hamiltonian in matter can then be obtained by
adding the potential terms, diagonal in the flavour basis.
For electron and muon neutrinos
⇥ ⇤ 2 ⌅ 2
⌅ ⇥
d |⇥e ⇥ m
cos 2 + 2GF Ne m
sin 2 |⇥e ⇥
i = 4E
m2
4E2
4E sin 2 cos 2
dt |⇥µ ⇥ m |⇥µ ⇥
4E

For antineutrinos the potential has the opposite sign.

In general the evolution is a complex problem but there

are few cases in which analytical or semi-analytical
results can be obtained.

33
 2-neutrino case in constant density
⇥ ⇤ 2 ⌅ 2
⌅ ⇥
d |⇥e ⇥ m
cos 2 + 2GF Ne m
sin 2 |⇥e ⇥
i = 4E
m2
4E2
4E sin 2 cos 2
dt |⇥µ ⇥ m |⇥µ ⇥
4E

If the electron density is constant (a good

approximation for oscillations in the Earth crust), it is
easy to solve. We need to diagonalise the Hamiltonian.

34
 Effective Hamiltonian Mixing angle

( )
vacuum
tan 2✓ ⇠
2

( )
matter suppression (Sun, SN)
M
2
tan 2✓ ⇠ ⌧ tan 2✓
+

35
( ) tan 2✓
MSW resonance (Sun, SN)
M 2
⇠
-
⇠1
 2-neutrino case in constant density
⇥ ⇤ 2 ⌅ 2
⌅ ⇥
d |⇥e ⇥ m
cos 2 + 2GF Ne m
sin 2 |⇥e ⇥
i = 4E
m2
4E2
4E sin 2 cos 2
dt |⇥µ ⇥ m |⇥µ ⇥
4E

If the electron density is constant (a good

approximation for oscillations in the Earth crust), it is
easy to solve. We need to diagonalise the Hamiltonian.
● Eigenvalues:⇤
⇥2 ⇥2
m2 ⇥ m2
EA EB = cos(2 ) 2GF Ne + sin(2 )
2E 2E
● The diagonal basis and the flavour basis are related by
a unitary matrix with angle in matter Exercise
Derive

m2
sin(2 )
tan(2 m) = m2
2E
⇥
2E cos(2 ) 2GF Ne
36
 ⇥ m2
● If 2GF Ne cos 2 , we recover the vacuum
2E
case and
m
⇥ m2
● If 2GF Ne 2E
cos(2 ) , matter
effects dominate
and oscillations are suppressed.
m2
● If 2GF Ne = cos 2 : resonance and maximal
2E
mixing
m = ⇥/4

● The resonance condition can be satisfied for

- neutrinos if m > 0
2

- antineutrinos if m < 0
2

(EA EB )L
P (⇥e ⇥ ⇥µ ; t) = sin (2 2
m ) sin
2

37
2
 In long baseline experiments
m2 p p
2E
cos(2✓) ⌫ + 2GF Ne ⌫¯ 2GF Ne

For neutrinos

( )
m2 > 0

tan 2✓ M
⇠
enhancement
2
- ++
For antineutrinos

( )
m2 > 0 suppression
M 2
tan 2✓ ⇠
- + -
38
 Matter effects modify the oscillation probability in LBL
experiments.
m
L
P = sin 2
23 sin 2
2 m
sin
2 13
µ e 13
2
The probability enhancement happens for

- neutrinos if m >0
2

- antineutrinos if m <0
2

The impact of matter effects is stronger at

higher energies and at longer baselines.

39
 2-neutrino oscillations with varying density
Let’s consider the case in which Ne depends on time.
This happens, e.g., if a beam of neutrinos is produced
and then propagates through a medium of varying
density (e.g. Sun, supernovae).
⇥ ⇤ 2 ⌅ 2
⌅ ⇥
d |⇥e ⇥ m
cos 2 + 2GF Ne (t) m
sin 2 |⇥e ⇥
i = 4E
m2
4E2
4E sin 2 cos 2
dt |⇥µ ⇥ m |⇥µ ⇥
4E

At a given instant of time t, the Hamiltonian can be

diagonalised by a unitary transformation as before. We
find the instantaneous matter basis and the instantaneous
values of the energy. The expressions are exactly as
before but with the angle which depends on time, θ(t).
40
 We have
| = U (t)| I , U (t)Hm,f l U (t) = diag(EA (t), EB (t))
†

Starting from the Schroedinger equation, we can

express it in the instantaneous basis
⇥ ⇤ 2 ⌅ 2
⌅ ⇥
d |⇥A ⇥ m
cos 2 + 2GF Ne (t) m
sin 2 |⇥A ⇥
i Um (t) = 4E
m2
4E2 Um (t)
4E sin 2 cos 2
dt |⇥B ⇥ m |⇥B ⇥
4E

⇥ ⇥ ⇥
d |⇥A ⇥ EA (t) i ˙(t) |⇥A ⇥
i =
dt |⇥B ⇥ i ˙(t) EB (t) |⇥B ⇥

The evolution of νA and νB are not decoupled. In

general, it is very difficult to find an analytical solution
to this problem.
41
 Adiabatic case
In the adiabatic case, each component evolves
independently. In the non adiabatic one, the state
can “jump” from one to the other.
If the evolution is sufficiently slow (adiabatic case):
| ˙(t)| ⇥| EA EB |

we can follow the evolution of each component

independently.

Adiabaticity condition
˙ m2
2|⇥| sin(2⇥) 2E
1
⇥ = 3
⇤1
|V̇CC |
|EA EB | |EA EB |
m MeV
2
In the Sun, typically we have
42 10 eV
9 2 E
 Solar neutrinos: MSW effect
The oscillations in matter were first discussed by L.
Wolfenstein, S. P. Mikheyev, A. Yu Smirnov.

● Production in the centre of the Sun: matter effects

dominate at high energy, negligible at low energy.
⇥A is cos 2
The probability of νe to be m
⇥B is sin2 m

If matter effects dominate, sin2 m 1

1
● P (⇥e ⇥ ⇥e ) = 1 (averaged vacuum
sin2 (2 )
2
oscillations), when matter effects are negligible (low
energies)
● P (⇥e ⇥e ) = sin2 (dominant matter effects and
43
adiabaticity) (high energies)
 Neutrinos oscillations in experiments
Neutrino production
In CC (NC) SU(2) interactions, the W boson (Z boson)
will be exchanged leading to the production of neutrinos.
p (u quark)
n (d quark)
W
electron
Beta decay.

electron
antineutrino
W muon
pion
Pion decay muon
44 Decay into electrons is suppressed. antineutrino
 Neutrino detection
Neutrino detection proceeds via CC (and NC) SU(2)
interactions. Example:
electron electron
neutrino

n p

Notice that the leptons have different masses:

me = 0.5 MeV < mmu = 105 MeV < mtau= 1700 MeV

A certain lepton will be produced in a CC only if the

45
neutrino has sufficient energy.
 We are interested mainly in produced charged particles as
these can emit light and/or leave tracks in segmented
detectors (magnetisation -> charge reconstruction).

Super-Kamiokande
detector

T2K experiment

NOvA
detector MINOS experiment

46
 Neutrino sources

J. Formaggio and S. Zeller, 1305.7513

47
 Solar neutrinos
Electron
2014
neutrinos are copiously produced in the Sun.
pp-chain
http://www.sns.ias.edu/∼jnb/

2012

2007,2008,2011
O. Smirnov, for 2012 (d/n)
Borexino,2010 2014 (seasonal)

Neutrino 2018
014+ Phase II data used
SNO
● Energies: 0.1-10 MeV.
● One can observed CC
νe and NC: measuring
the oscillation
disappearance and the
48 Super-Kamiokande overall flux.
 the observed spectrum (MC)
BOREXINO (in operation from May,2007) Solar neutrino
14
electron recoil spectra
Irreducible C and other internal
radioactive contaminants : 210Po,210Bi (both
•278 t of liquid organic not in secular equilibrium), 85Kr, 11C
External γ (high energy)
scintillator PC + PPO
(1.5 g/l)
• (ν,e)-scattering with
low threshold (~200
keV)
•Outer muon detector

Super-Kamiokande
• Data-set:• Dec Data-set:
14 2011-
th DecMay
th 14 21
st 2011- 2016May
esults
Results
MC input counting rates are quoted in cpd/100 t

50000 tons of Kamioka mine

13.7m • Total exposure:
•Water Cherenkov 1291.51 Elastic
Totaldetector
exposure:days1291.51
x 71.3 days
ton
~1km
18m
iv : 1707.09279
arXiv
O. Smirnov, for
: 1707.09279
• Fit range:
• (0.19-2.93)
Fit range: neutrino
(0.19-2.93)
MeV ~3km
MeV ~2km
(2700 mwe)
Borexino, scattering
For Solar in
neutrino a
Neutrino 2018
WC:
Cherenkov light Livetime Fiducial
Phase Period
(days) (kton
ID
41.4 m directional
SK-I
1996.4 ~
1496
2001.7
22.5
OD
Charged
particle
information
SK-II
2002.10 ~
791
2005.10

2006.7 ~ 22.5 (>5.5

SK-III 548
M. Ikeda, for
2008.8 13.3 (<5.5

Super- 22.5 (>5.5

SK-IV 2008.9 ~ 2860 13.3 (4.5<E
Kamiokande, 8.8 (<4.5M

4 39.3 m Neutrinototal2018
5695 days
Neutrino
 Solar neutrinos have energies
which go from vacuum
oscillations to adiabatic
resonance. MSW effect at high
energies, vacuum oscillations at
low energy.

Borexino
SAGE, GALLEX
SuperKamiokande
SNO
50 Strumia and Vissani
 as a function of neutrino energy

solar Δm2
Kamland Δm2 Super-K

SNO

O. Smirnov, for Borexino, Neutrino M. Ikeda, for Super-Kamiokande,

2018 Neutrino 2018
MSW errors (1σ) are shown by rose band
Total error on Pee:
• for pp and pep neutrinos, contribution of experimental errors dominates (easy to predict, dif
measure) Borexino
• for 7Be and 8B theoretical predictions of the Solar model are worse than measurements

SAGE, GALLEX
SuperKamiokande
SNO
51 Strumia and Vissani
 Solar ν Angle θ12 & Mass2 Difference
Super-K data best constrains Δm221
SNO data best constrains sin2θ12
complementarity makes combined
fit beneficial
correlation via 8B flux further
SNO tightens constraints

M. Ikeda, for Super-

Kamiokande,
Neutrino 2018
SK+SNO

Super-K
14

Solar experiments best constrain the “solar mixing”

angle of theta12 to be large (but non-maximal). The
mass squared difference is around 7x10-5 eV2.
 Atmospheric neutrinos
Cosmic rays hit the atmosphere and produce pions
(and kaons) which decay producing lots of muon and
electron (anti-) neutrinos.
● Typical energies: 100 MeVdetector
Super-Kamiokande - 100 GeV
● Typical distances: 100-10000 km.
50,000 tons Ring imaging Water Cherenk
1,000m und
Inner detector ~11
41.4m

Outer detector ~ 1
SK-I started in Apr. 1st, 1
SK-IV finished on
39m
53 1996 2002 SuperKamiokande Coll.
2006 2008
Prof. Kajita gave a talk on the “evidence for nm oscillation”
at Neutrino 1998. (June 5th, exactly 20 years ago.)
SK reported the
first evidence of
n e u t r i n o
oscillations in
Search for nt appearance at Sup
1Zenith
9 9angle
8 distribution
w i t h Ex
a t m o sSK-I+II+III+IV
p 5326
h edaysric
neutrinos
(numu->nutau).
( 33 kt yr8 )
http://www-sk.icrr.u-tokyo.ac.jp/nu98/
T. Kajita’s talk at1711.09436v1[hep-ex]
arXiv: Neutrino 1998

SK and MINOS went on to

measure the atmospheric mixing
Fitted Excess
Atm n BKG MC

angle to be large (mainly maximal) # of tau events

338.1 ± 72.7 (stat.+ sys.) events
and the atm mass squared different Reject no-tau-appearance @ 4.6s.
( Exp. significance is 3.3s )
at ~2.5x 10-3 eV2.
 Reactor neutrinos
Copious amounts of electron antineutrinos are
produced from reactors.
● Typical energy: 1-3 MeV;
● Typical distances: 1 km (Daya Bay, DCHOOZ, RENO)
—100 km (KamLAND). Antineutrino Detection
• Antineutrinos are detected via the Inverse Beta Decay (IBD) reaction:
● At these energies inverse beta decay interactions
dominate.
νe + p → e + n
+

These detectors
~30μs ~8MeV
signal use liquid
scintillator.
J. P. Ochoa-Ricoux, for Daya Bay,
55 at Neutrino 2018
At ~few km, the disappearance probability is
Oscillation
2
m L
Re
P (¯
⇥e ⇥ ⇥¯e ; t) = 1 sin (2 2
13 ) sin
2 31
• 4Esin22θ13 and |Δm2
Measure

Sensitivity to θ13. Reactors played an important role in P (

10
10 Entries

Δ χ2
Mean
RMS

8
8

the discovery of θ13 and in its precise measurement.

6
6

4
4

Neutrino mixing at reactors

2
2

2 4
0
0.065 0.07 0.075 0.08 0.085 0.09 0.095 0.1 0.105

2.8 2.8

2.7 2.7

2.6 2.6

Δm2ee [eV2]
Short baseline
experiments θ13 experiments:
Double Chooz 2.5
„Reactor anomaly“,
2.5

Daya Bay θ12

CENNS
RENO 2.4 2.4

2.3 2.3

KamLAND:
„Solar“ 2.2
preliminary 2.2

Parameters,
0.07 0.08 0.09 0.1
JUNO
sin (2θ13)
2

J. P. Ochoa-Ricoux,sinfor2 2θ
results with
C. Buck, for DoubleCHOOZ Coll., at Daya Bay, at Neutrino 13
% $m13(eV )L(m) ( 1958 days
Δm 2
2 2
P(Neutrino 2018
! e ! ! e ) " 1# 2 2
sin 2"13 sin '1.27 * 2018 | ee
|= (
 In 2012, previous hints The Big Bang
(DoubleCHOOZ,T2K, Theory: The
Speckerman

MINOS) for a nonzero third Recurrence

mixing angle were confirmed

by Daya Bay and RENO:
important discovery.

Daya Bay: reactor neutrino

experiment in China, Courtesy of Roy Kaltschmidt

RENO Experimental Set-up

Double-CHOOZ, A. 120 m.w.e.

Near Detector

Cabrera

29
4m

13
83
m
T2K event in 2011 RENO Far Detector

I.Yu 450 m.w.e.

This discovery has very important implications for the future

57 neutrino programme and understanding of the origin of mixing.
At ~100 km, the disappearance probability is
2 2
⇥
Solar4 ν Angle2 θ12 & 2Mass Differen m21 L
P (¯
⇥e ⇥ ⇥¯e ; t) ⇤ c13 1 sin (2 12 ) sin
4
+ s13 ⨉
4E
Sensitivity to Delta m212.
⨉

SK+SNO
⨉
Solar+
KamLAND

Solar

M. Ikeda, for Super-Kamiokande, Neutrino 2018

 Accelerator neutrinos
Conventional beams: muon neutrinos from pion decays

Neutrino production. T2K event

Credit: Fermilab

MINOS event

● Typical energies:
MINOS: E~4 GeV; T2K: E~700 MeV; NOvA: E~2 GeV.
OPERA and ICARUS: E~20 GeV.
● Typical distances: 100 km - 2000 km.
MINOS: L=735 km; T2K: L=295 km; NOvA: L=810 km.
OPERA and ICARUS: L=700 km.
59
 At these energies, one can detect electron, muon (and
tau) ν via CC interactions.

MINOS,T2K, NOvA:
m231 L
P( µ ⇥ µ ; t) = 1 4s223 c213 (1 s223 c213 ) sin2
4E

m231 L
T2K, NOvA: P (⇥µ ⇥e ; t) = 2
s23 sin (2
2
13 ) sin
2
+...
4E
<latexit sha1_base64="/xccffGbtcu+lCBiKkxNSWLJCuA=">AAAB/3icbVDLSsNAFJ3UV62vqks3g0UQhJCIoMuiG5cVTFtoQ5lMb9qhM5MwMxFC6cIvcKtf4E7c+il+gP/htM3Cth64cDjnXu69J0o508bzvp3S2vrG5lZ5u7Kzu7d/UD08auokUxQCmvBEtSOigTMJgWGGQztVQETEoRWN7qZ+6wmUZol8NHkKoSADyWJGibFScIFd1+1Va57rzYBXiV+QGirQ6FV/uv2EZgKkoZxo3fG91IRjogyjHCaVbqYhJXREBtCxVBIBOhzPjp3gM6v0cZwoW9Lgmfp3YkyE1rmIbKcgZqiXvan4n9fJTHwTjplMMwOSzhfFGccmwdPPcZ8poIbnlhCqmL0V0yFRhBqbz8KWGHIp0onNxV9OYZU0L13fc/2Hq1r9tkiojE7QKTpHPrpGdXSPGihAFDH0gl7Rm/PsvDsfzue8teQUM8doAc7XL6a2le8=</latexit>

OPERA (and m231 L

P (⇥µ ⇥⇥ ; t) = c413 sin2 (2 ) sin2
ICARUS): 23
4E

Sensitivity to 2
m31 , 23 , 13

60
 At these energies, one can detect electron, muon (and
tau) ν via CC interactions.

MINOS,T2K, NOvA:
m231 L
P( µ ⇥ µ ; t) = 1 4s223 c213 (1 s223 c213 ) sin2
4E

m231 L
T2K, NOvA: P (⇥µ ⇥e ; t) = 2
s23 sin (2
2
13 ) sin
2
+...
4E
<latexit sha1_base64="/xccffGbtcu+lCBiKkxNSWLJCuA=">AAAB/3icbVDLSsNAFJ3UV62vqks3g0UQhJCIoMuiG5cVTFtoQ5lMb9qhM5MwMxFC6cIvcKtf4E7c+il+gP/htM3Cth64cDjnXu69J0o508bzvp3S2vrG5lZ5u7Kzu7d/UD08auokUxQCmvBEtSOigTMJgWGGQztVQETEoRWN7qZ+6wmUZol8NHkKoSADyWJGibFScIFd1+1Va57rzYBXiV+QGirQ6FV/uv2EZgKkoZxo3fG91IRjogyjHCaVbqYhJXREBtCxVBIBOhzPjp3gM6v0cZwoW9Lgmfp3YkyE1rmIbKcgZqiXvan4n9fJTHwTjplMMwOSzhfFGccmwdPPcZ8poIbnlhCqmL0V0yFRhBqbz8KWGHIp0onNxV9OYZU0L13fc/2Hq1r9tkiojE7QKTpHPrpGdXSPGihAFDH0gl7Rm/PsvDsfzue8teQUM8doAc7XL6a2le8=</latexit>

OPERA (and m231 L

P (⇥µ ⇥⇥ ; t) = c413 sin2 (2 ) sin2
ICARUS): 23
4E

Sensitivity to 2
m31 , 23 , 13

<latexit sha1_base64="qp/gctUp99aYyZZgbJkzRDC1pdQ=">AAACGHicbZDLSgMxFIYz9VbrrepSF8EiVJAyUwVdFnXhsoK9QFtLJj3ThiaZIckIZejGx/AJ3OoTuBO37nwA38P0srCtPwQ+/nMO5+T3I860cd1vJ7W0vLK6ll7PbGxube9kd/eqOowVhQoNeajqPtHAmYSKYYZDPVJAhM+h5vevR/XaIyjNQnlvBhG0BOlKFjBKjLXa2cNmB7ghp1h3Zb55M2IsHort5MwbnrSzObfgjoUXwZtCDk1Vbmd/mp2QxgKkoZxo3fDcyLQSogyjHIaZZqwhIrRPutCwKIkA3UrGvxjiY+t0cBAq+6TBY/fvREKE1gPh205BTE/P10bmf7VGbILLVsJkFBuQdLIoiDk2IR5FgjtMATV8YIFQxeytmPaIItTY4Ga2BDCQIhraXLz5FBahWix4bsG7O8+VrqYJpdEBOkJ55KELVEK3qIwqiKIn9IJe0Zvz7Lw7H87npDXlTGf20Yycr1+d1p+a</latexit>
, sgn( m231 )
61
 Combi
nedFi
tRe
sul
ts
|Dm232|
NH:( 2
IH:( 2

Me
asu
OSa
ndMI
NOS+Ov
erv
iew Combi
nedFi
tRe
su
● MINOSandMI
NOS+pr obemuon-
neut
ri
nodi
sappe
ara
nceo
verabr
oadr
ang
eof
ndMI NOS+ ene
rgi
es Fa
rDe
tec
tor
ignedto Underg
r
●oundinSouda
nmine s
in
utri
no ● Co
nsi
st
enc
ywi
th7th
3
5re
k
m e
●
f
f
rl
oa
mvo
t
arr
gp
tr
e edic
t i
ont
ight
lyc
ons
tra
insa
lte
rna
teos
cil
l
ati
ons
nso ve
rlong ● 5.
4ktonma s
s 0
.3
hypot
hes
es
usi
ngt wo
thatare:
nti
ll
ator Det
ect
orsar
eon-axisfor
4J
une2
018 A
NuMInd
ea
um
t
ri
nA
our
i
bs
ea
n
amo-Uni
ver
sit
yofCi
nci
nna
ti 9
cal
orimet
ers
i
nmuo ns
nal
lyide
nti
c a
l
emati
c Bestfit Ex
clus
ionofmaxi
malmixi
ng:
int
yreduct
ion -3
i
zedfors
ign
Dm232=2 .
42x1
0 eV2 Pr
efer
encef
orl
oweroct
ant:
nandener
gy sn2q23=0
i .4
2 Pr
efer
encef
ornor
malhie
rarc
h
i
on
4J
une2
018
Ne
●
arDe
tec
tor
AtFer
mi l
ab
A. Aurisano, for Ada
m Aur
isa
no-Uni
ver
sit
yofCi
nci
nna
ti

● 1km fr
om tar
get MINOS and MINOS+,
1kto
nma s
s
at Neutrino 2018
●
 neutrino candidate interactions
νμ
p
νµ CC Signal

μ
𝜇+p
M. Sanchez, NOvA Coll.,
at Neutrino 2018 M. Wasko, T2K Coll. , at
Neutrino 2018
SK data f
1.49e21 POT νµ/ν̅μ disappe
p
νe CC Signal see poster #75
e+p
PREDICTING THE FD
M UOOB
νe
NSRun
T2K E RE V
N1-9c
e
AT OO
UPreliminary
TRIN ND
π
NC neutrino
• Each quartile for the Signal orand
Background
antineutrino beams gets
• Theunfolded
combinedindependently andand
data of neutrino
Near ratio is used to obtain a FD prediction from ND data.
theantineutrino
true Far/ bea ⌫µ
π0
γ
• We observe 113 events and expect 126 at this comb
ν
• We estimate
p
cosmic background
𝜋0 +𝜋 + p rate from the timing sidebands
γ of the
and observe 65NuMI
eventsbeam triggers
and expect 52and
at the best fit
1m

cosmic trigger data.

1m
10 102 10 3
10 10 2
q (ADC) 10 3 q (ADC) • If fit separately, the antineutrino beam mode prefers
10 102 10 3 q (ADC)

Mayly Sanchez - ISU 9

! neutrino beam mode. However, the χ2s are consisten
Neutrino beam NOvA Preliminary Antineutrino
parametersbeam NOvA
with p > 4%. Also Preliminary
allowed region compa
FD Data FD Data
12
All Quartiles
Prediction 8
All Quartiles νe/ν̅ appear
Prediction
e
1-σ syst. range 180
T2K Run1-9c Preliminary 1-σ syst. range T2K Run1-9c P
No Feldman-Cousins 0.9 NOvA180
Preliminary

θ (degrees)

θ (degrees)
Number of Events
10
Wrong Sign:νµCC 160 3.0
νe (QE) Wrong
160 Sign:νµCC νe (1 −2
Events / 0.1 GeV

Events / 0.1 GeV

0.8
NOvA
NOvA NH
NH 90%
90% CLCL0.7
140 140
Total bkg. 6 ν µ + ν µ 2018 Total bkg.
ν µ + ν µ 20180.6
8 120 120
Cosmic bkg. 100
2.8 ν µ 2017 0.5
Cosmic
100
bkg. −2

Δm232 (10 eV2)

∆m232 (10 eV2)

80 0.4 80
6
-3

-3
4
60 0.3 60
2.6 −2
40 0.2 40
20 0.1 20
4
0 0 0
0 0.2
2.4 0.4 0.6 0.8 1 1.2 0 0.2 0.4 0.6 0.8 −2
2 ν Reconstructed Energy (GeV) ν Reconstructed Energ
2 Imperial College
Best fit
London
2018 / 06 / 04 2.2 Neutrino 2018−2
0.3 0.4 0.5 0.6 0.7
0 0 sin2θ23
0 1 2 3 4 5 0 1 2 3 4 5
Reconstructed Neutrino Energy (GeV) Reconstructed Neutrino Energy (GeV)
 A. Aurisano,
for MINOS
and MINOS+,
at Neutrino
2018

The
Pos
ter
: accelerator
We dne
sday#53
, muon
T
.Ca
rrol
lneutrino disappearance channel
4J
une2
018 has allowed to
Ada
mmeasure
Aur
isa
no-Uni
vewith
r
sit
yofCi
ncgood
i
nna
ti precision
|Delta m231| and theta23.
 Measurement of oscillation parameters
Atmospheric neutrinos
SK (MINOS, IceCube)
LBL exp numu
✓23 disapp.:
MO 2
m31 MINOS, MINOS+
<latexit sha1_base64="VIvUvolC0yYZAFAQwGNHBsLGq7k=">AAACC3icbVBLSgNBFOyJvxg/GXXppjEIrsJMFHQZ1IXLCOYDSQw9nTdJk+6eobtHCEOO4Anc6gnciVsP4QG8hz1JFiax4EFR9R71qCDmTBvP+3Zya+sbm1v57cLO7t5+0T04bOgoURTqNOKRagVEA2cS6oYZDq1YAREBh2Ywusn85hMozSL5YMYxdAUZSBYySoyVem6xcwvcECweK7303J/03JJX9qbAq8SfkxKao9Zzfzr9iCYCpKGcaN32vdh0U6IMoxwmhU6iISZ0RAbQtlQSAbqbTh+f4FOr9HEYKTvS4Kn69yIlQuuxCOymIGaol71M/M9rJya86qZMxokBSWdBYcKxiXDWAu4zBdTwsSWEKmZ/xXRIFKHGdrWQEsJYijjrxV9uYZU0KmXfK/v3F6Xq9byhPDpGJ+gM+egSVdEdqqE6oihBL+gVvTnPzrvz4XzOVnPO/OYILcD5+gWGgprN</latexit>

T2K, NOvA
✓12
Reactor CPV LBL exp nue app.:
neutrinos: T2K, NOvA
JUNO
✓13

65
Also: Tests of standard neutrino paradigm
 -2.2 ★

sin
[1

δC
0.02 -2.4

Current status of neutrino parameters

Dm32
2
-2.4

∆m32
2
-2.6 90
-2.6
0.01
-2.8
0.3 3.0
NuFIT 0.4(2016)
0.5 0.6 0.7 360

3 sizable mixing
-2.8
0.3 0.4 0.5 0.6 0.7 360 2
2.8 2 sin q23
sin θ23 0 0

angles
8.5
∆m31

2.6
2

★ 0.03 ★ 270
0.03
2.4 270

2 mass squared
★
H
sin2 θ13[10-3 eV2]

δCP[10 eV ]
2
2.2
0.025
180
differences
-2.2
0.025

sin q13
180

-5
H
7.5 ★ ★

dCP
-2.4
32
2

★
0.02
∆m

Δm21
0.02
NuFit 3.0: M. C. Gonzalez-Garcia et al., 90
2
-2.6
90
7 1611.01514, Pre-Neutrino 2018
-2.8 0.015
0.015 0.3 0.4 0.5 0.6 0.7 360
2
sin θ23
0
06.5 8.5
8.5 0.2 0.25 0.3 0.35 0.4 0 0.01 0.02 0.03 0.04
See also F. Capozzi
2 et al., 1703.04471 2
0.03 270 sin θ12 sin θ13
★

neutrinos have mass

8
●
Dm21 [10 eV ]

8
2
θ13 eV ]
2

0.025 Figure 1: Global 3⌫ oscillation analysis. Each panels shows two-dimensional projection of
7.5
-5

neutrinos mix (Misaligned

-5

●
allowed 180
six-dimensional region after marginalization
H with respect to theH undisplayed paramete
[10

7.5 ★ ★
The di↵erent contours correspond to the two-dimensional allowed regions at 1 , 90%, 2 , 9
22

δCP

★
2
21
sin

flavour and massive states)

∆m

0.02 and 3 CL (2 dof). Results for di↵erent assumptions concerning the analysis of data from reac
7
experiments90 are shown:6.5
full regions correspond to analysis with the normalization of reactor flu
0.015
6.5
left free and data from short-baseline (less than 100 m) reactor experiments are included. F
0.2 0.25 0.3 0.35void0.4 0.015 0.2
0.02 0.25
0.025 0.3
0.03are0.35 0.4 0.015 0.02 0.025 0.03
regions short-baseline reactor data 2 not included but reactor fluxes
2 as predicted in [42]
2 2 sin q sin q

First evidence of physics

sin θ12 assumed. Note 0 sin θ13 12
8.5 that as atmospheric mass-squared splitting we use m231 for13 NO and m232 for

8
beyond the –Standard Model.
0 eV ]
2

4–
66
-5

7.5 ★ ★
 Conclusions

● 20 years ago, Neutrino oscillations were discovered.

They play a critical role in the study of neutrino
properties:their discovery implies that neutrinos have
mass and mix.

● Three mixing angles, two mass squared differences

have been measured with good precision. The data are
also giving the first hints in favour of MO and LCPV.

● A wide experimental program has taken place and

new experiments are underway. Stay tuned!
67
 Further theoretical issues on neutrino oscillations

Energy-momentum conservation
Let’s consider for simplicity a 2-body decay: ⇤ µ ⇥¯µ .

Energy-momentum conservation seems to require:

E⇥ = Eµ + E1 with E1 = p2 + m21

?
E⇥ = Eµ + E2 with E2 = p2 + 2
m2

How can the

picture be
consistent?
68
 Further theoretical issues on neutrino oscillations

Energy-momentum conservation
Let’s consider for simplicity a 2-body decay: ⇤ µ ⇥¯µ .

Energy-momentum conservation seems to require:

E⇥ = Eµ + E1 with E1 = p2 + m21
E⇥ = Eµ + E2 with E2 = p2 + 2
m2

These two requirements seems to be incompatible.

Intrinsic quantum uncertainty, localisation of the initial
pion lead to an uncertainty in the energy-momentum and
allow coherence of the initial neutrino state.
69
 The need for wavepackets
● In deriving the oscillation formulas we have implicitly
assumed that neutrinos can be described by plane-
waves, with definite momentum.
● However, production and detection are well localised
and very distant from each other. This leads to a
momentum spread which can be described by a wave-
packet formalism.
Typical sizes:
- e.g. production in decay: the relevant timescale is the
pion lifetime (or the time travelled in the decay pipe),
t ⇥ E⇥ p x
For details see, Akhmedov, Smirnov, 1008.2077; Giunti and Kim, Neutrino Physics and
70 Astrophysics.
 Decoherence and the size of a wave-packet

● The different components of the wavepacket, ν1, ν2

and ν3, travel with slightly different velocities (as their
mass is different).

● If the neutrinos travel extremely long distances, these

components stop to overlap, destroying coherence and
oscillations.

● In terrestrial experimental situation this is not

relevant. But this can happen for example for
supernovae neutrinos.

71