Cottingham W.N., Greenwood D.A. An Introduction to the Standard Model of Particle Physics
Подождите немного. Документ загружается.
20.4 KamLAND 199
best-fit oscillation
sin
2
2q = 1.0
∆m
2
= 6.9 × 10
−5
eV
2
2.6 MeV
analysis threshold
positron energy (MeV.)
25
20
15
10
5
0
2 468
Events
Figure 20.3 KamLAND data (K. Eguchi et al. Phys. Rev. Lett 90, 021802 (2003)).
The energy distribution of the observed positrons in bins of 0.425 MeV (solid
circles with error bars), along with the expected no oscillations distribution (upper
histogram) and the best fit including oscillations using (20.5) (lower histogram).
The shaded bands indicate the systematic error in the best fit distribution. The
vertical dashed line corresponds to the analysis threshold at 2.6 MeV.
and from the parameterization (14.16)
4|U
e1
|
2
|U
e2
|
2
= cos
4
θ
e3
sin
2
2θ
e2
≈ sin
2
2θ
e2
.
As at Chooz, MSW effects are negligible. The measured positron energy spectrum
is compared with the positron energy spectrum that would be expected if there
were no antineutrino oscillations. This spectrum can be very well estimated from
knowledge of the various reactor characteristics.
Some results from KamLAND are shown in Fig. 20.3. The energy spectrum of
the positrons is clearly below what it would be without oscillation. The best fit to
the data using an expression based on (20.5) has
|m
2
21
|=6.9 × 10
−5
eV
2
,
0.84 < sin
2
2θ
e2
< 1.
The KamLAND analysis took some account of systematic errors arising from the
simplifying assumption |U
e3
|
2
= 0.
200 Neutrino masses and mixing: experimental results
20.5 Atmospheric neutrinos
The Earth is continually bombarded by cosmic rays, which consist for the most part
of high energy protons and electrons. The protons, in their collisions with nuclei
in the upper atmosphere, produce π mesons. The π mesons decay by the chains
(Section 9.2, Section 9.4):
π
+
→ µ
+
+ ν
µ
, π
−
→ µ
−
+
¯
ν
µ
.
→
e
+
+ ν
e
+
¯
ν
µ
→
e
−
+
¯
ν
e
+ ν
µ
The neutrinos and antineutrinos are produced at a mean height ∼ 20 km, with
energies extending to the multi-GeV region. The ratio of the flux ν
µ
+
¯
ν
µ
to the
flux of ν
e
+
¯
ν
e
is evidently about 2.
In water detectors, such as Super Kamiokandi, charged leptons are produced
through reactions essentially of the form
ν
e
+ n → e
−
+ p,
¯
ν
e
+ p → e
+
+ n;
ν
µ
+ n → µ
−
+ p,
¯
ν
µ
+ p → µ
+
+ n.
The charged leptons emit Cerenkov radiation, which provides information on the
energy, direction and identity of the incident neutrino.
Figure 20.4 shows some results from the Super-Kamiokandi detector. The plots
show the ratio of observed ν
e
- and ν
µ
-like events to Monte Carlo calculations
in the absence of oscillations, as a function of D/E
ν
. E
ν
is the neutrino energy
and D the distance from the point of production ∼ 20 km above the Earth’s sur-
face, to the detector. D is then inferred from the measured neutrino direction.
For multi-GeV electron neutrinos, the MSW modification to the equations has
to be included for those neutrinos passing through the Earth on their way to the
detector.
The ν
e
data show no sign of oscillation, but there is a clear deficit of muon
neutrinos. The best fit to the data has m
2
At
= 2.2 × 10
−3
eV
2
, and like K2K has
sin
2
2θ
µ3
= 1, where for D/E
ν
< 10
3
km/GeV the m
2
32
and m
2
31
oscillations
are combined into one m
2
At
oscillation. The absence of discernible ν
e
→ ν
e
oscil-
lations in the data was the first indication of the smallness of |U
e3
|
2
, which again
implies that the m
2
At
oscillations are predominantly between ν
µ
and ν
τ
.
20.6 Solar neutrinos
The nuclear and thermal physics of the Sun is well understood. The solar neutrino
spectra predicted by the Standard Solar Model and shown in Fig. 20.5 may be
assumed with confidence.
20.6 Solar neutrinos 201
1.5
0.5
1
0
11010
2
10
3
10
4
10
5
D/E
ν
(km/GeV)
Measured events/Expected events in the absence of oscillations
e-like
µ-like
Figure 20.4 Data from Super Kamiokande (Y. Fukunda et al. Phys. Rev. Lett.
82, 1562 (1998). The ratio of measured events to expected events in the absence
of oscillations. The lines show the expected shape for ν
µ
↔ ν
τ
with m
2
At
=
2.2 × 10
−3
eV
2
and sin
2
(2θ
µ3
) = 1. There is no significant ν
e
↔ ν
e
oscillation
observed.
The first measurements of the spectra were made by R. Davis and his collabora-
tors in the deep Homestake mine in the U.S.A, (Davis 1964). The detection of the
neutrinos was made through the reaction
ν
e
+
37
17
Cl + 0.81 MeV → e
−
+
37
18
Ar.
The Super-Kamiokande detector also made measurements of the solar neutrino
flux with E
ν
greater than about 6 Mev (Fukuda et al. 1996).
Because of the high energy threshold these measurements were blind to the
principal flux from the ‘pp’ reaction. The GALLEX (Italy) and SAGE (Russia)
202 Neutrino masses and mixing: experimental results
Ga
15
O
10
16
10
14
10
12
10
10
10
8
10
6
0.1 0.2 0.5 1 2 5 10 20
17
F
7
Be
8
B
pep
hep
13
N
Cl SNO
pp
Flux at 1 AU (m
−2
s
−1
MeV
−1
) (for lines m
−2
s
−1
)
Neutrino ener
gy
(
MeV
)
Figure 20.5 The solar neutrino spectra predicted by the standard solar model.
Spectra for the pp chain are shown by solid lines and those for the CNO chain by
dashed lines. (See Bahcall, J. N. and Ulrich, R. K. (1988), Rev. Mod. Phys. 60,
297.)
experiments were designed to remedy this, and examine the pp flux through the
reaction (Hampel et al., 1999;Gavrin et al., 2003)
ν
e
+
17
31
Ga + 0.23 MeV →
71
32
Ge + e
−
.
The SNO (Sudbury Neutrino Observatory, Canada) is a heavy water detector. Neu-
trinos, with E
ν
greater than about 5 Mev, are detected through the reactions, (Ahmad
et al. 2002)
ν
e
+ D
2
+ 1.44 MeV → e
−
+ p + p,
ν
e
+ D
2
+ 2.22 MeV → p + n + ν.
The first of these reactions is a charged current interaction and can be initiated only
by an electron neutrino. The second is a neutral current interaction, initiated with
20.7 Solar MSW effects 203
equal probability by an electron, muon or tau neutrino. The SNO experiment also
measured the reaction rate of elastic neutrino scattering from electrons,
ν + e
−
→ ν + e
−
.
Again, this reaction can be triggered by a neutrino of any type. Measurements can
be used to infer both the ν
e
flux φ(ν
e
)) and the total flux φ(ν
e
+ ν
µ
+ ν
τ
).
The early results from the Homestake detector gave a measured flux of only
about one third of that expected from the standard Solar Model without oscillation.
Super Kamiokande, GALLEX and SAGE gave about half the expected rate. SNO
found that
φ(ν
e
)
φ(ν
e
+ ν
µ
+ ν
τ
)
= 0.306 ± 0.05.
The measured total neutrino flux was consistent with that expected from the Stan-
dard Solar Model and clearly, since the Sun produces only electron neutrinos, many
have made the transition to ν
µ
and ν
τ
.
20.7 Solar MSW effects
We showed in Section 19.4 that plane wave neutrino mass eigenstates depended on
functions f
i
(z), that satisfied
i
d f
i
dz
=
m
2
i
2E
f
i
+ V (z)U
∗
ej
U
ei
f
j
. (20.6)
The source of solar neutrinos is the central region of the Sun, where the Standard
Solar Model gives V (o) = 7.6 × 10
−12
eV. Comparing this with m
2
21
/2E, which
with the ‘reference parameters’ of Section 20.1 equals 3.5 × 10
−12
(10 MeV/E
ν
), it
is clear that the interpretation of the data from solar neutrino experiments requires
a serious consideration of the MSW effect.
As a starting approximation we again neglect the small term U
e3
.With U
e3
= 0
the solution of (20.6) for f
3
(z)is
f
3
(z) = e
im
2
3
z/2E
f
3
(0),
independent of V (z). With U
e3
= 0, and since the initial neutrino is an electron
neutrino, f
3
(0) = 0 and it follows that f
3
(z)iszero for all z:itplays no part in
the oscillations. The approximation again reduces the analysis to a two-neutrino
phenomenon in f
1
(z) and f
2
(z). After some algebra it can be shown that the solar
204 Neutrino masses and mixing: experimental results
neutrino data can be analysed with the equations
i
d f
e
dz
=
m
2
21
2E
(−cos(2θ
e2
) f
e
+ sin(2θ
e2
) f
x
) + V (z) f
e
(20.7)
i
df
x
dz
=
m
2
21
2E
(sin(2θ
e2
) f
e
+ (cos 2θ
e2
) f
x
).
f
x
= c
µ3
f
µ
− s
µ3
f
τ
is a combination of f
µ
and f
τ
, V (z)isknown from the Standard
Solar Model. The equations have to be integrated numerically.
All the solar neutrino data is consistent with the oscillation interpretation, and
analysis of the data gives 3 × 10
−5
eV
2
<m
2
21
< 1.9 × 10
−4
eV
2
, 30.2
◦
<θ
e2
<
34.9
◦
with high probability (95% confidence level). The best fit is with m
2
21
=
6.9 × 10
−5
eV
2
,θ
e2
= 32
◦
.
The solar neutrino data give a tighter constraint on θ
e2
than KamLAND. Also,
with the MSW effect, the solution of equations (20.7) depends on the sign of m
2
21
.
It is found to be positive, as is indicated in Figure 20.1
20.8 Future prospects
There are several planned experiments that will make a more thorough investigation
of neutrino masses and mixing phenomena. Apart from the possibility of sterile
neutrinos, indications of which have not been confirmed, there is no evidence to
contradict the three-neutrino theory of Chapter 19.However, it can be seen from
the quality of the data presented in this chapter that the neutrino mass theory is not
as well established as other branches of The Standard Model. Within the theory
experiments are planned to make more precise measurements of the m
2
and the
parameters of the neutrino mixing matrix.
The principal focus of experimental activity is on the construction of muon
neutrino beams as in the K2K experiment. An advantage of accelerator-generated
neutrinos is the control that one has on the flux and energy distribution. K2K is
an ongoing experiment but by late 2006 the muon neutrino experiments CNGS
and MINOS (Main Injector Neutrino Oscillation Search) will be in operation. The
CNGS neutrinos are generated at CERN and detected at the GRAN SASSO under-
ground laboratory in Italy. The MINOS beam is generated at Fermilab and detected
in the Soudan mine in Minnesota. Both experiments will look for evidence of the
rare ν
µ
→ ν
e
transition and for the expected ν
µ
→ ν
τ
oscillations. If the theory of
Chapter 19 is not challenged it is expected that by 2010 we will have much tighter
bounds on both sin
2
(2θ
e3
) and |m
2
At
|.
In the more distant future a new very high intensity proton accelerator will be
built at Tokai, Japan. The experiment T2K will take over from K2K with a neutrino
beam of much higher intensity. Detection at Super Kamiokande will give a base line
20.8 Future prospects 205
0.01
0.005
0.5 1.0
E
ν
P
D
ν
µ
→ ν
e
ν
µ
→ ν
e
Figure 20.6 The upper curve is P
D
(ν
µ
→ ν
e
).
The lower curve P
D
(
¯
ν
µ
→
¯
ν
e
).
The parameters are m
2
32
= 2 ×10
−3
eV
2
,m
2
21
= 7 ×10
−5
eV
2
,
cos θ
e2
= 0.84, cos θ
µ3
= 1/
√
2, sin θ
e3
= 0.05,δ = π/4, D = 295 km.
The MSW effect, which depends on the local geology will be significant but
calculable. It is not included here.
D ≈ 295 km. An upgrade to higher intensity for MINOS is also planned with a new
experiment NOνA. By 2015 with T2K and NOνAitisexpected that if sin
2
(2θ
e3
) >
0.01 then it will be detected. The MSW effect will influence these measurements and
the sign of m
2
32
could be established, and hence the mass ordering. If sin
2
(2θ
e3
) can
be measured then it is also possible to have a measurement of the CP violating phase
δ. Figure 20.6 shows the transition probabilities P
D
(ν
µ
→ ν
e
) and P
D
(
¯
ν
µ
→
¯
ν
e
)
as a function of E
ν
with the T2K baseline. δ is taken as 45
◦
and the other parameters
are a plausible set. Although the probabilities are small, the particle and antiparticle
probabilities differ considerably (see Section 19.3).
21
Majorana neutrinos
Majorana fields were introduced in Section 6.6.Ifneutrino fields are Majorana,
then there is no distinction to be made between neutrinos and antineutrinos. As
explained in Section 6.7, the smallness of neutrino masses makes the differences
between Dirac and Majorana neutrinos difficult to discern experimentally.
In this chapter we elaborate on the theory of Majorana neutrinos and show
how they can be accommodated within the Standard Model. Finally we describe
experiments on ‘double β decay’ that may determine the nature of neutrinos.
21.1 Majorana neutrino fields
We shall denote left-handed and right-handed Majorana neutrino fields by ν
L
(x)
and ν
R
(x). From (6.28 and 6.29), making the identifications
b
p+
= d
p+
, b
p−
= d
p−
we have for a Majorana neutrino field carrying mass m
ν
L
=
1
√
V
p
m
2E
p
b
p+
e
−θ/2
|
+
+ b
p−
e
θ/2
|
−
e
i(p·r−Et)
+
b
∗
p+
e
θ/2
|
−
− b
∗
p−
e
−θ/2
|
+
e
i(−p·r+Et)
,
(21.1)
ν
R
=
1
√
V
p
m
2E
p
b
p+
e
θ/2
|
+
+ b
p−
e
−θ/2
|
−
e
i(p·r−Et)
+
−b
∗
p+
e
−θ/2
|
−
+ b
∗
p−
e
θ/2
|
+
e
i(−p·r+Et)
.
(21.2)
The fields ν
L
(x) and ν
R
(x) are not independent. It is easily shown, using Problem
6.5, that
iσ
2
|
−
∗
=
|
+
,
iσ
2
|
+
∗
=−
|
−
,
206
21.2 Majorana Lagrangian density 207
and then that
ν
R
= (iσ
2
)ν
∗
L
and ν
L
=−(iσ
2
)ν
∗
R
. (21.3)
Thus either field may be derived from the other. As a consequence, only left-handed
Majorana fields or only right-handed Majorana fields need appear in any theory.
The charge conjugate field ν
c
L
was defined in (7.11b)by
ν
c
L
=−(iσ
2
)ν
∗
R
.
But by the results above −(iσ
2
)ν
∗
R
= ν
L
,sothat
ν
c
L
= ν
L
. (21.4)
Thus the charge conjugate of a Majorana field is identical to the field. There is no
room in the theory of Majorana neutrinos for a distinguishable antineutrino. For
agiven momentum, there are two basic particle states, which we may take to be
one with helicity +
1
/
2
, the other with helicity −
1
/
2
. (In these respects, Majorana
neutrinos are somewhat similar to photons, but with photons having helicities ±1).
21.2 Majorana Lagrangian density
The Majorana field is constructed from solutions of the Dirac equation. We saw in
Section 5.2 that the Lagrangian density for a free Dirac particle of mass m is
L
Dirac
= iψ
†
L
˜σ
µ
∂
µ
ψ
L
+ iψ
†
R
σ
µ
∂
µ
ψ
R
− m
#
ψ
†
L
ψ
R
+ ψ
†
R
ψ
L
$
.
In the case of a Majorana field, ν
R
is determined by ν
L
, and given by (21.3) above.
We choose to work with ν
L
, and therefore take the Majorana Lagrangian density to
be
L
M
=
1
2
iν
†
˜σ
µ
∂
µ
ν + i(iσ
2
ν
∗
)
†
σ
µ
∂
µ
(iσ
2
ν
∗
) − m
!
ν
†
(iσ
2
)ν
∗
+ ν
T
(−iσ
2
)ν
"
,
where ν = ν
L
.For the remainder of this chapter we shall drop the subscript L,
for clarity of notation. ν is a two component left-handed neutrino field. We have
introduced a factor of
1
/
2
to compensate for double counting.
The second dynamical term in L
M
is equivalent to the first (Problem 21.1), so
that the Lagrangian density may be written
L
M
= iν
†
˜σ
µ
∂
µ
ν −
m
2
!
ν
†
iσ
2
ν
∗
+ ν
T
−iσ
2
ν
"
. (21.5)
It is interesting and important to note that, with finite mass m and with the
Majorana constraints, we lose the U(1) symmetry that gave neutrino number
208 Majorana neutrinos
conservation in the Dirac case (Section 7.1). We shall see that with Majorana
neutrinos the overall lepton number is no longer conserved.
Noting the factor
1
/
2
in the Lagrangian density, the Hamiltonian operator H and
momentum operator P for Majorana neutrinos are (see Section 6.5)
H =
1
2
p,ε
b
∗
pε
b
pε
− b
pε
b
∗
pε
E
p
=
p,ε
b
∗
pε
b
pε
E
p
,
P =
1
2
p,ε
b
∗
pε
b
pε
− b
pε
b
∗
pε
p =
p,ε
b
∗
pε
b
pε
p,
(21.6)
where ε =±1isthe helicity index.
21.3 Majorana field equations
Avariation δν
∗
in the Majorana action yields the field equation
i˜σ
µ
∂
µ
ν = m
iσ
2
ν
∗
.
(Note that there are two contributions from the mass term in the Lagrangian density.)
In a frame K
in which the Majorana neutrino is at rest, p
i
ν
=−i∂
i
ν
= 0(i =
1, 2, 3), and the field equation reduces to
i
∂ν
∂t
= m
iσ
2
ν
∗
(21.7)
It is easy to verify that this equation has two solutions of the form
ν
1
= be
−iEt
1
0
+ b
∗
e
iEt
0
1
and ν
2
= be
−iEt
0
1
− b
∗
e
iEt
1
0
,
with E = m. (21.8)
We may then, as in Section 6.3, transform to a frame K in which the Majorana
neutrino is moving with velocity ν>0inthe Oz direction:
ν
1
= M
−1
ν
1
=
e
−θ/2
0
0e
θ/2
be
−imt
1
0
+ b
∗
e
imt
0
1
= be
−mt
e
−θ/2
1
0
+ b
∗
e
imt
e
θ/2
0
1
.
Substituting t
= t cosh θ − z sinh θ,
ν
1
= be
−θ/2
1
0
e
i( pz−Et)
+ b
∗
e
θ/2
0
1
e
i(−pz+Et)
. (21.9)