Petcov PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 14

Neutrinoless double beta decay and H

-l
/
l

decays in the Higgs triplet model


S. T. Petcov,
1,2,
*
H. Sugiyama,
1,
and Y. Takanishi
3,
1
SISSA and INFN-Sezione di Trieste, I-34014 Trieste, Italy
2
IPMU, University of Tokyo, Tokyo, Japan
3
Physik-Department, Technische Universitat Munchen, D-85747 Garching, Germany
(Received 21 April 2009; published 10 July 2009)
The connection between the neutrinoless double beta [()
0
] decay effective Majorana mass }M
ee
},
and the branching ratios of the decays H

-l

l
/
, l, l
/
e, , of the doubly charged Higgs boson
H

is analyzed within the Higgs triplet model of neutrino mass generation. We work in the version of the
model with explicit breaking of the total lepton charge conservation, in which H

-l

l
/
, l, l
/
e, ,
are the dominant decay modes of H

. It is assumed also that H

are relatively light so that they can


be produced at the CERN LHC and the branching ratios of interest measured. Taking into account the
current and prospective uncertainties in the values of the neutrino mixing parameters most relevant for the
problem studiedthe atmospheric neutrino mixing angle
23
and the CHOOZ angle
13
, and allowing the
lightest neutrino mass and the CP violating Dirac and Majorana phases to vary in the intervals [0, 0.3 eV]
and j0; 2|, respectively, we derive the regions of values of BR(H

-e

) and BR(H

-e

)
for which }M
ee
} 0:05 eV, or }M
ee
} <0:05 eV. This is done for the normal mass ordering case, the
inverted mass ordering case, and the case when the ordering has not been determined. The analysis is
performed both without using possible additional data on BR(H

), as well as using
prospective data on these branching ratios. In the latter case, results for several values of BR(H

) are presented.
DOI: 10.1103/PhysRevD.80.015005 PACS numbers: 12.60.Fr, 14.60.Pq, 14.80.Cp
I. INTRODUCTION
Determining the origin of neutrino masses and mixing is
one of the major challenges of future research in neutrino
physics. It is well known that the existence of nonzero
neutrino masses can be related to the presence of a more
complicated Higgs sector in the standard theory, involving
additional Higgs elds beyond the single doublet eld.
Actually, it was realized a long time ago [13] that a
Majorana mass term for the left-handed (LH) avor neu-
trino elds can be generated by SU(2)
L
U(1)
Y
invariant
Yukawa couplings of two lepton doublet elds to a Higgs
triplet eld, carrying two units of the weak hypercharge,
}Y} 2. Such a Higgs eld has electrically neutral, singly
charged, and doubly charged components. The Majorana
mass term for the active avor neutrinos arises when the
neutral component of the Higgs triplet eld acquires a
nonzero vacuum expectation value (vev), breaking the
SU(2)
L
U(1)
Y
symmetry. There are several possible
realizations of this scenario. The realization in which the
global U(1)
L
symmetry associated with the conservation of
the total lepton charge L is broken only spontaneously by
the Higgs triplet vev [4], was ruled out by the LEP data on
the invisible decay width of the Z
0
boson. If, however, the
U(1)
L
symmetry is broken explicitly in a manner that leads
to a nonzero vacuum expectation value of the neutral
component of the Higgs triplet eld (see, e.g. [5,6]) one
obtains a viable model of neutrino mass generation. This
model has been investigated in detail recently [79] (see
also, e.g. [10,11]) and was shown to have a rich and
physically interesting phenomenology owing to the fact
that (i) the couplings of the doubly and singly charged
Higgs elds to the avor neutrinos and charged leptons are
proportional to the elements of the Majorana mass matrix
of the (avor) neutrinos, M
l
/
l
, and can be relatively large,
and that (ii) the physical doubly charged and singly
charged Higgs elds, H

and H

, can have masses in


the range from 100 GeV to 1 TeV and thus can, in
principle, be produced and observed at the CERN LHC.
Point (i) implies that the indicated couplings are deter-
mined essentially by the elements of the Pontecorvo-
Maki-Nakagawa-Sakata (PMNS) neutrino mixing matrix
[12] and by the neutrino masses. In [79], it was shown that
by studying the decays H

-l

l
/
, l, l
/
e, , , it
might be possible to obtain information on the absolute
neutrino mass scale, on the type of neutrino mass spectrum
[which can be, e.g. normal hierarchical (NH), inverted
hierarchical (IH), and quasidegenerate (QD)], and on the
Majorana CP violating phases [13] present in the neutrino
mixing matrix.
In the present article we investigate the possibility to use
the information on the neutrino mass spectrum and the
Majorana CP violating phases from the measurements of
the H

-l

l
/
decay branching ratios, BR(H

-
l

l
/
), l, l
/
e, , in order to obtain predictions for the
effective Majorana mass in neutrinoless double beta
*
Also at: Institute of Nuclear Research and Nuclear Energy,
Bulgarian Academy of Sciences, 1784 Soa, Bulgaria.

sugiyama@sissa.it

yasutaka.takanishi@ph.tum.de
PHYSICAL REVIEW D 80, 015005 (2009)
1550-7998=2009=80(1)=015005(14) 015005-1 2009 The American Physical Society
[()
0
] decay, }M
ee
} (see, e.g. [14]). Our study is moti-
vated by the fact that most probably the searches for the
doubly charged scalars H

and the decays H

-l

l
/
will be carried out at LHC before the next generation of
()
0
-decay experiments will be operative. Among the
different decay channels H

-l

l
/
, l, l
/
e, , , the
easier to observe are those with two electrons (positrons),
two muons (antimuons), or an electron (positron) and a
muon (antimuon), e

, and e

, in the nal
state. If the mass of H

does not exceed approximately


400 GeV, the branching ratios of the H

decays into
e

, e

, and

can be measured at LHC with


a few percent error [15]. We will show that if the doubly
charged Higgs bosons H

will be discovered at LHC and


at least the three branching ratios BR(H

-e

),
BR(H

-e

), and BR(H

) will be
measured with a sufcient accuracy, one can obtain unique
information on the ()
0
-decay effective Majorana mass
}M
ee
}. This information will be extremely important, in
particular, for the upcoming next generation of
()
0
-decay experiments.
II. THE HIGGS TRIPLET MODEL
In the Higgs triplet model (HTM) [13] a I 1, Y 2
complex SU(2)
L
triplet of Higgs scalar elds is added to
the standard model (SM) Lagrangian. In the 2 2 repre-
sentation, the Higgs triplet eld has the form


+
=

2
_

++

+
=

2
_
!
; (1)
where
0
,
+
, and
++
are neutral, singly charged, and
doubly charged scalar elds. In the avor basis in which
the charged lepton mass matrix is diagonal and which we
are going to use throughout this article, a Majorana mass
term for the LH avor neutrinos can be generated (without
the introduction of a right-handed neutrino eld) by the
SU(2)
L
U(1)
Y
gauge invariant Yukawa interaction:
L h
l
/
l
c
T
l
/
L
C(i
2
)c
lL
+H:c: (2)
Here h
l
/
l
h
ll
/ , l
/
, l e, , are complex Yukawa cou-
plings forming a symmetric matrix h, C is the charge
conjugation matrix,
2
is a Pauli matrix for SU(2)
L
indices,
and c
T
lL
(
lL
l
L
)
T
, l e, , is the LH lepton doublet
eld. A nonzero triplet vacuum expectation value, (
0
)
v

0, gives rise to a Majorana mass matrix Mfor the LH


avor neutrino elds
lL
:
M
l
/
l
2h
l
/
l
(
0
)

2
_
h
l
/
l
v

; l
/
; l e; ; : (3)
The requisite v

0 arises from the minimization of the


most general SU(2) U(1)
Y
invariant Higgs potential
[5,6]:
V m
2
(
|
) +
1
(
|
)
2
+M
2

Tr(
|
)
+
2
jTr(
|
)|
2
+
3
Det(
|
)
+
4
(
|
)Tr(
|
) +
5
(
|

i
)Tr(
|

i
)
+

2
_ (
T
i
2

|
) +H:c:

; (4)

T
(
+

0
)
T
being the SM Higgs doublet eld. In
Eq. (4), M
2

>0 is the common mass of the triplet scalars.


The choice m
2
<0 ensures that (
0
) v=

2
_
0, which
breaks spontaneously SU(2) U(1)
Y
to U(1)
Q
. In the
model of Gelmini-Roncadelli [4], the term
(
T
i
2

|
) is absent, which leads for M
2

<0 to a
spontaneous breaking of the global U(1)
L
symmetry asso-
ciated with the conservation of the total lepton number.
The resulting Higgs spectrum contains a massless
Goldstone bosonthe triplet scalar Majoron, J, and an-
other light scalar, H
0
. The decay Z
0
-H
0
J would give too
large a contribution to the invisible decay width of the Z
0
boson and this model was excluded by the LEP data. The
inclusion of the term (
T
i
2

|
) explicitly breaks the
lepton number conservation when is assigned two units
of the total lepton charge L, and therefore avoids the
presence of a Goldstone bosonthe Majoron, in the model
[2,3]. Thus, the scalar potential in Eq. (4) together with the
triplet Yukawa interaction of Eq. (2) lead to a phenomeno-
logically viable model of neutrino mass generation.
The expression for v

resulting from the minimization


of the potential V, Eq. (4), reads
v

v
2
2M
2

+(
4
+
5
)v
2
; for v

<v: (5)
In the scenario of relatively light triplet scalars within the
discovery reach of the LHC we will be interested in, one
has M

- v and Eq. (5) leads to v

- . In extensions of
the HTM, the term (
T
i
2

|
) can arise in various
ways: (i) through the vev of a Higgs singlet eld [16,17];
(ii) generated at higher orders in perturbation theory [6]; or
(iii) appearing in the context of theories with extra dimen-
sions [5,18].
An upper limit on v

can be obtained from considering


its effect on the parameter M
2
W
=M
2
Z
cos
2

W
. In the
SM, 1 at tree level, while in the HTM one has
1 +
1 +2x
2
1 +4x
2
; x v

=v: (6)
The measurement - 1 leads to the bound v

=v & 0:03,
or v

<8 GeV. At the 1-loop level v

must be renormal-
ized and explicit analyses lead to bounds on its magnitude
similar to those derived from the tree-level analysis [19].
The HTM has seven physical Higgs scalar particles
(H
++
; H

; H
+
; H

; H
0
; A
0
; h
0
). The doubly charged
Higgs eld H
++
coincides with the triplet scalar eld

++
. The remaining Higgs mass-eigenstate elds are, in
S. T. PETCOV, H. SUGIYAMA, AND Y. TAKANISHI PHYSICAL REVIEW D 80, 015005 (2009)
015005-2
general, mixtures of the doublet and triplet elds. The
corresponding mixing parameter is proportional to the ratio
of triplet and doublet vevs, v

=v, and hence is small even if


v

assumes its largest value of a few GeV. Therefore H


+
,
H
0
, A
0
are predominantly composed of the triplet elds,
while h
0
is predominantly composed of the doublet eld
and plays the role of the SM Higgs boson. The masses of
H

, H

, H
0
, A
0
are of order M

with splittings of order

5
v. For M

<1 TeV of interest for direct searches for the


Higgs bosons at the LHC, the couplings h
l
/
l
are constrained
to be O(0:1) or less by a variety of processes such as -
eee, -lll, etc. These constraints are reviewed in
[20,21].
In this article we will be interested, in particular, in the
decays of H

into a pair of same-sign charged leptons,


H

-l
/
l

, l
/
, l e, , , which give clear signals
even in hadron colliders like the LHC. These decays are
important not only for the searches of H

, but also
because of the very interesting possibility that their lepton
avor dependence can be directly related to the Majorana
mass matrix of the LH avor neutrinos.
In our analysis, we will assume that M
H
M
H
and
v

& 1 MeV. Under these conditions the decay H

-
H

is forbidden, while the decay H

-W

is
sufciently strongly suppressed. In this case, the branching
ratios of the decays H

-l
/
l

are given by the follow-


ing simple expressions (see, e.g. [79]):
BR
l
/
l
BR(H

-l
/
l

)
2
1 +
l
/
l
}h
l
/
l
}
2
P
l
/
l
}h
l
/
l
}
2
(7)

2
1 +
l
/
l
}M
l
/
l
}
2
P
i
m
2
i
; (8)
where
l
/
l
is the Kronecker delta. Note that the branching
ratios depend only on the parameters of neutrino mass
matrix. The measurement of BR
l
/
l
can give signicant
information on the elements of the neutrino mass matrix
1
}M
l
/
l
}, and therefore, e.g. on the absolute neutrino mass
scale (i.e. lightest neutrino mass), type of neutrino mass
spectrum, Majorana CP violating phases in the neutrino
mixing matrix, etc. This simple relation does not hold in a
type of the left-right symmetric models [23] with Higgs
triplets [24], in which lepton number is spontaneously
broken and Majoron is eaten by new gauge bosons, be-
cause of contaminations from the right-handed neutrino
sector.
Among six channels of H

-l
/
l

we use BR
ee
,
BR
e
, and BR

in our analysis. For measurements of


those three branching ratios in pp -H
++
H

(whose
cross section is about 20 fb at M
H
300 GeV for

s
_

14 TeV), SM backgrounds come from t

t, t

tZ, and ZZ.


Those backgrounds can be eliminated easily by requiring
high transverse momentum, exclusion of invariant masses
of l
/+
l

around m
Z
, and coincidence of two invariant
masses of l
/+
l
+
and l
/
l

. The background from W for


pp -H

(whose cross section is about twice of that


of pp -H
++
H

for M
H
M
H
[25]) can be cut by
taking high transverse invariant mass given by transverse
energy-momentums of l and missing one. Some details on
cuts of SM backgrounds can be seen in [15] for also some
channels involving s.
III. THE NEUTRINO MASSES, MIXING, AND THE
()
0
DECAY
We work in the avor basis in which the mass matrix of
the charged leptons is diagonal. As we have shown, in the
Higgs triplet model of interest, the LH avor neutrino
elds
lL
acquire a Majorana mass term. The correspond-
ing Majorana mass matrix M is diagonalized with the help
of the PMNS neutrino mixing matrix [12]:
M
ll
/ jU
PMNS
diag(m
1
; m
2
; m
3
)U
T
PMNS
|
ll
/ ; (9)
where m
j
, j 1, 2, 3, are the real positive eigenvalues of
M
ll
/ the masses of the Majorana neutrinos
j
with de-
nite mass. In what follows, we will use the standard pa-
rametrization of the PMNS matrix (see, e.g. [26,27]):
U
PMNS

c
12
c
13
s
12
c
13
s
13
e
i
s
12
c
23
c
12
s
23
s
13
e
i
c
12
c
23
s
12
s
23
s
13
e
i
s
23
c
13
s
12
s
23
c
12
c
23
s
13
e
i
c
12
s
23
s
12
c
23
s
13
e
i
c
23
c
13
0
B
@
1
C
Adiag(1; e
i(
21
=2)
; e
i(
31
=2)
); (10)
where c
ij
cos
ij
, s
ij
sin
ij
, the angles
ij
j0; =2|
(i <j 1, 2, 3), j0; 2| is the Dirac CP-violating
phase, and
21
and
31
are two Majorana CP-violation
phases [13,28]. The phases
21
and
31
can vary in the
interval j0; 2|. It proves useful for our further discussion
to dene also the difference of the two Majorana phases:

32

31

21
. Let us add that at present we do not have
experimental information on ,
21
, and
31
.
The existing neutrino oscillation data [2933] allow us
to determine with a rather good precision the mixing angles
and neutrino mass squared differences which drive the
solar neutrino and the dominant atmospheric neutrino os-
cillations, sin
2
2
12
, m
2
21
and sin
2
2
23
, }m
2
31
}(
}m
2
32
}), and to obtain a rather stringent limit on the
CHOOZ angle
13
. In our analysis, we will use the follow-
ing best t values of sin
2
2
12
, m
2
21
, sin
2
2
23
, and }m
2
31
}
[3436]:
1
For the current experimental information on }M
ll
/ }, see [22].
NEUTRINOLESS DOUBLE BETA DECAY AND . . . PHYSICAL REVIEW D 80, 015005 (2009)
015005-3
m
2
21
7:6 10
5
eV
2
; sin
2
2
12
0:87; (11)
}m
2
31
} 2:4 10
3
eV
2
; sin
2
2
23
1: (12)
The upper limit on sin
2
2
13
obtained in CHOOZ reactor
antineutrino experiment [32] reads
sin
2
2
13
<0:14: (13)
From the global analyses of the neutrino oscillation data
one nds (see, e.g. [36])
sin
2

13
<0:056; 99:73%C:L: (14)
The next generation of experiments with reactor
e
, which
are under preparation, Double CHOOZ [37], Daya Bay
[38], RENO [39], can improve the currently reached sen-
sitivity to the value of sin
2
2
13
by a factor of (5)(10) (see,
e.g. [40]), while future long baseline experiments aim at
measuring values of sin
2
2
13
as small as 10
4
10
3
(see,
e.g. [41]).
Let us note that the uncertainty in the experimental
determination of sin
2
2
23
corresponds to a rather large
interval of allowed values of s
2
23
[30]: 0:38 s
2
23
0:62.
We will take into account this uncertainty in our numerical
analysis.
2
It should be added that the accuracy on sin
2
2
23
is planned to be improved considerably in future long
baseline experiments. The uncertainty in sin
2
2
23
is fore-
seen to be reduced in the T2K experiment [42], for in-
stance, to sin
2
2
23
>0:99 (0:45 <s
2
23
<0:55), if the true
value of sin
2
2
23
1. As we will see, the correlations
between the branching ratios of the decays H

-
l

l
/
, BR
ll
/ , l, l
/
e, , , and the effective Majorana
mass in neutrinoless double beta [()
0
] decay, }(m)}
M
ee
, which is the main subject of our study, depend not
only on the elements of the neutrino mixing matrix, but
also on the type of neutrino mass spectrum and on the
absolute scale of neutrino masses.
As is well known, owing to the fact that the sign of m
2
31
cannot be determined from the existing data, there are two
possible types of neutrino mass spectrum compatible with
the datawith normal ordering and with inverted ordering.
In the standard convention we are going to employ, the two
spectra correspond to
(i) m
1
<m
2
<m
3
, m
2
31
>0, normal ordering (NO),
(ii) m
3
<m
1
<m
2
, m
2
31
<0, inverted ordering (IO).
The -mass spectrum can be (i) normal hierarchical,
m
1
<m
2
<m
3
, with m
2

m
2
21
q
8:8 10
3
eV,
m
3

m
2
31
q
4:9 10
2
eV; (ii) inverted hierarchical,
m
3
<m
1
<m
2
, with m
2

m
2
23
q
4:9 10
2
eV,
m
1

m
2
23
m
2
21
q
4:8 10
2
eV; and
(iii) quasidegenerate), m
1
m
2
m
3
, m
2
1;2;3
>}m
2
31
}.
In the latter case one has m
j
* 0:10 eV.
The type of neutrino mass spectrum, i.e. sgn(m
2
31
), can
be determined by studying oscillations of neutrinos and
antineutrinos, say,

-
e
and

-
e
, in which matter
effects are sufciently large. This can be done in long
baseline -oscillation experiments (see, e.g. [41]). If
sin
2
2
13
* 0:05 and sin
2

23
* 0:50, information on
sgn(m
2
31
) might be obtained in atmospheric neutrino ex-
periments by investigating the effects of the subdominant
transitions
(e)
-
e()
and
(e)
-
e()
of atmospheric
neutrinos which traverse the Earth [43]. For
(e)
[or
(e)
]
crossing the Earth core, new type of resonancelike en-
hancement of the indicated transitions takes place due to
the (Earth) mantle-core constructive interference effect
[neutrino oscillation length resonance (NOLR)] [44].
3
For m
2
31
>0, the neutrino transitions
(e)
-
e()
are
enhanced, while for m
2
31
<0 the enhancement of anti-
neutrino transitions
(e)
-
e()
takes place, which might
allow to determine sgn(m
2
31
). If sin
2

13
is sufciently
large, the sign of m
2
31
can also be determined by studying
the oscillations of reactor
e
on distances of (2040) km
[48]. An experiment with reactor
e
, which, in particular,
might have the capabilities to measure sgn(m
2
31
), was
proposed recently in [49] (see also [50]). Information on
the type of neutrino mass spectrum can also be obtained in
-decay experiments having a sensitivity to neutrino
masses

}m
2
A
}
q
5 10
2
eV [51] (i.e. by a factor
of 4 better than the planned sensitivity of the KATRIN
experiment [52]; see below).
Direct information on the absolute neutrino mass scale
can be derived in
3
H-decay experiments [5254]. The
most stringent upper bounds on the
e
mass were obtained
in the Troitzk [54] and Mainz [52] experiments:
m

e
<2:3 eV; 95%C:L: (15)
We have m

e
m
1;2;3
in the case of the QD -mass spec-
trum. The KATRIN experiment [52], which is under prepa-
ration, is planned to reach a sensitivity of m

e
0:20 eV,
i.e. it will probe the region of the QD spectrum.
The CMB data of the WMAP experiment [55], com-
bined with data from large scale structure surveys
(2dFGRS, SDSS), lead to the following upper limit on
the sum of neutrino masses (see, e.g. [56]):
2
Varying sin
2

12
in the 3 interval of allowed values of
sin
2

12
[34,36], 0:25 & sin
2

12
& 0:37, has an essentially neg-
ligible effect on the results of our analysis.
3
As a consequence of this effect, the corresponding
(e)
[or

(e)
] transition probabilities can be maximal [45] (for the
precise conditions of the mantle-core (NOLR) enhancement,
see [44,45]). Let us note that the Earth mantle-core (NOLR)
enhancement of neutrino transitions differs [44] from the
Mikheev-Smirnov-Wolfenstein one. The conditions of the
Earth mantle-core enhancement [44,45] also differ [46] from
the conditions of the parametric resonance enhancement of the
neutrino transitions discussed in the articles [47].
S. T. PETCOV, H. SUGIYAMA, AND Y. TAKANISHI PHYSICAL REVIEW D 80, 015005 (2009)
015005-4
X
j
m
j
<(0:41:7) eV; 95%C:L: (16)
Data on weak lensing of galaxies, combined with data from
the WMAP and PLANCK experiments, may allow to be
determined with an uncertainty of 0:04 eV [56,57].
In our analysis, we will consider both types of neutrino
mass spectrumwith normal and with inverted ordering,
as well as the specic cases of NH, IH, and QD spectra.
Correspondingly, the lightest neutrino mass min(m
j
)
m
0
, which determines the absolute neutrino mass scale,
will be varied in the interval
0 m
0
0:3 eV; m
0
min(m
j
); j 1; 2; 3:
(17)
As we will show, the results we obtain essentially do not
depend on the maximal value m
0
as long as the latter is not
smaller than 0.3 eV. The reason is that for m
0
* 0:3 eV
(i.e. in the QD region), the branching ratios BR
l
/
l
we are
interested in practically do not depend on the neutrino
masses:
BR
l
/
l

2
3(1 +
l
/
l
)

X
j
U
l
/
j
U
lj

2
; m
0
* 0:3 eV:
(18)
A. Neutrinoless Double Beta decay
In the Higgs triplet model the massive neutrinos are
Majorana particles. Determining the nature of massive
neutrinos is of fundamental importance for understanding
the origin of neutrino masses and, more generally, for
understanding the symmetries governing the particle inter-
actions. The existence of massive Majorana neutrinos is
associated with nonconservation of the total lepton charge.
In this case, the neutrinoless double beta decay (A; Z) -
(A; Z +2) +2e

is allowed (see, e.g. [14,58,59]).


Assuming that the dominant mechanism for the decay is
the exchange of light Majorana neutrinos (Fig. 1(a)), the
half-life T
0
1=2
for the decay is given by
T
0
1=2
(G
0
}M
0
}
2
}M
ee
}
2
)
1
; (19)
where G
0
is a phase space factor and M
0
is the nuclear
matrix element (NME) of the process. All the dependence
of T
0
1=2
on the neutrino masses and mixing parameters
factorizes into the ()
0
-decay effective Majorana mass
M
ee
:
}M
ee
} }c
2
12
c
2
13
m
1
+s
2
12
c
2
13
m
2
e
i
21
+s
2
13
m
3
e
i(
31
2)
}:
(20)
The most stringent upper bound on }M
ee
}, }M
ee
} <
(0:351:05) eV was obtained by using the lower limit
T
0
1=2
>1:9 10
25
yr (90% C.L.) found
4
in the
Heidelberg-Moscow
76
Ge experiment [61]. The IGEX col-
laboration has obtained the result T
0
1=2
>1:57 10
25
yr
(90% C.L.), from which the limit }M
ee
} <
(0:331:35) eV was derived [62]. A positive
()
0
-decay signal at >3, corresponding to T
0
1=2

(0:694:18) 10
25
yr (99.73% C.L.) and implying
}M
ee
} (0:10:9) eV, is claimed to be observed in [63],
while a recent analysis reports evidence at 6 of ()
0
decay with }M
ee
} 0:32 0:03 eV [64]. Two experi-
ments, NEMO3 (with
100
Mo,
82
Se, etc.) [65] and
CUORICINO (with
130
Te) [66], designed to reach a sensi-
tivity to }M
ee
} (0:20:3) eV, set the limits, }M
ee
} <
(0:611:26) eV [65] and }M
ee
} <(0:190:68) eV [66]
(90% C.L.), where estimated uncertainties in the NME
are accounted for. The two upper limits were derived
from the experimental lower limits on the half-lives of
100
Mo and
130
Te, T
0
1=2
>5:8 10
23
yr (90% C.L.) [65]
and T
0
1=2
>3:0 10
24
yr (90% C.L.) [66]. Most impor-
tantly, a large number of projects aim at a sensitivity to
}M
ee
} (0:010:05) eV [67]: CUORE (
130
Te), GERDA
(
76
Ge), SuperNEMO, EXO (
136
Xe), MAJORANA (
76
Ge),
MOON (
100
Mo), COBRA (
116
Cd), XMASS (
136
Xe),
CANDLES (
48
Ca), etc. These experiments, in particular,
will test the positive result claimed in [63].
The predicted value of }M
ee
} depends strongly on the
type of -mass spectrum [26,68], more precisely, on the
FIG. 1. Diagrams for the neutrinoless double beta decay. The
diagram (a) is the standard and dominant one, and (b)-(f) are
possible but negligible in the HTM.
4
In the quoted upper bound for }M
ee
} a factor of 3 uncertainty
in the relevant NME (see, e.g. [60]) is taken into account.
NEUTRINOLESS DOUBLE BETA DECAY AND . . . PHYSICAL REVIEW D 80, 015005 (2009)
015005-5
type of hierarchy neutrino masses obey. The existence of
signicant and robust lower bounds on }M
ee
} in the cases of
IH and QD spectra [68] (see also [69]), given, respec-
tively,
5
by }M
ee
} * 0:01 eV and }M
ee
} * 0:03 eV, which
lie either partially (IH spectrum) or completely (QD spec-
trum) within the range of sensitivity of the next generation
of ()
0
-decay experiments, is one of the most important
features of the predictions of }M
ee
}. At the same time we
have }M
ee
} & 5 10
3
eV in the case of NH spectrum
[70]. The fact that max(}M
ee
}) in the case of NH spectrum
is considerably smaller than min(}M
ee
}) for the IH and QD
spectrum opens the possibility of obtaining information
about the type of -mass spectrum from a measurement
of }M
ee
} 0 [68]. More specically, a positive result in the
future generation of ()
0
-decay experiments with
}M
ee
} >0:01 eV would imply that the NH spectrum is
strongly disfavored (if not excluded). For m
2
31
>0, such
a result would mean that the neutrino mass spectrum is
with normal ordering, but is not hierarchical. If m
2
31
<0,
the neutrino mass spectrum should be either IH or QD.
IV. PREDICTION FOR }M
ee
} FROM
MEASUREMENTS OF BR(H

-l
/
l

)
In this section, we investigate within the HTM the
predictions one can obtain for the ()
0
-decay effective
Majorana mass }M
ee
} by using data on BR
l
/
l
. The dominant
mechanism of ()
0
decaythe light Majorana neutrino
exchange corresponds to the diagram in Fig. 1(a), the
contributions from the diagrams in Figs. 1(b)1(f) being
negligible [71,72].
We use three branching ratios, BR
ee
, BR
e
, and BR

,
in our analysis. The expressions for these branching ratios
in terms of neutrino masses, neutrino mixing angles, and
CP violating phases read (see [79])

X
i
m
2
i

BR
ee
}M
ee
}
2
}c
2
12
c
2
13
m
1
+s
2
12
c
2
13
m
2
e
i
21
+s
2
13
m
3
e
i(
31
2)
}
2
; (21)

X
i
m
2
i

BR
e
2}c
12
c
13
(s
12
c
23
c
12
s
23
s
13
e
i
)m
1
+s
12
c
13
(c
12
c
23
s
12
s
23
s
13
e
i
)m
2
e
i
21
+s
23
c
13
s
13
m
3
e
i(
31
)
}
2
; (22)

X
i
m
2
i

BR

}(s
12
c
23
c
12
s
23
s
13
e
i
)
2
m
1
+(c
12
c
23
s
12
s
23
s
13
e
i
)
2
m
2
e
i
21
+s
2
23
c
2
13
m
3
e
i
31
}
2
: (23)
Given the solar and atmospheric neutrino oscillation
parameters and the CHOOZ angle, sin
2

12
, m
2
21
,
sin
2

23
, }m
2
31
}, and
13
, }M
ee
} depends on m
0

min(m
j
),
21
,
31
, and on the type of neutrino mass
spectrum (NO or IO). In the case of spectrum with IO or of
QD type, the dependence of }M
ee
} on
13
is relatively weak
and can be neglected, as long as
6
cos2
12
>sin
2

13
. In
this case }M
ee
} does not depend on the Majorana phase
31
and on the Dirac phase . From the measurement of the
three observables, BR
ee
, BR
e
, and BR

, three parame-
ters, say, m
0
,
21
, and
31
, can, in principle, be determined
and information on the type of neutrino mass spectrum
with NH, IH or QDcan be obtained [79]. This would
allow us to tightly constrain }M
ee
}. Let us review briey the
predictions for BR
ee
, BR
e
, and BR

in the cases of NH,


IH, and QD spectrum (see also [79]).
A. NH spectrum, m
1
<m
2
< m
3
Using the best t values of the neutrino oscillation
parameters one nds
7
that in this case BR
ee
, BR
e
, and
BR

can take values in the following intervals: 0 &


BR
NH
ee
& 10
2
, 0 & BR
NH
e
& 0:08, 0:16 & BR
NH

& 0:31.
We get BR
NH
ee
0 for (
32
2) and s
2
13

s
2
12

m
2
21
=m
2
31
q
0:05, while BR
NH
e
0 for (
32

) and s
2
13
s
2
12
c
2
12
cot
2

23
(m
2
21
=m
2
31
)
6:9 10
3
. The minimal and maximal values of BR
NH

depend weakly on s
2
13
. Neglecting this dependence, one
obtains a simple expression for the Majorana phase (dif-
ference)
32
in terms of BR
NH

:
cos
32

1
2

m
2
31
m
2
21

1=2
BR
NH

s
4
23
c
2
12
c
2
23
s
2
23
: (24)
B. IH spectrum, m
3
<m
1
< m
2
We nd very different results in this case:
0:5c
4
13
cos
2
2
12
& BR
IH
ee
& 0:5c
4
13
or 0:06 & BR
IH
ee
& 0:5,
0 & BR
IH
e
& 0:48, 0 & BR
IH

& 0:14. One has BR


IH

0 for 0,
21
, and s
2
13
0:036. Now BR
IH
ee
exhib-
its a very weak dependence on s
2
13
. For the Majorana phase

21
we obtain in terms of BR
IH
ee
:
5
Up to small corrections we have in the cases of two spectra
[68]: }M
ee
} * }m
2
32
cos2
12
} (IH) and }M
ee
} * m
0
cos2
12
(QD). The possibility of cos2
12
0 is ruled out at 6 by
the existing data which also imply that cos2
12
* 0:30 (0.26) at
2 (3) [34,36]. We also have 2:07 10
3
eV
2
& }m
2
32
} &
2:75 10
3
eV
2
at 3.
6
The inequality cos2
12
>sin
2

13
is fullled for the 2
experimentally allowed ranges of values of cos2
12
and
sin
2

13
; see, e.g. [34,36]. If one uses the 3 ranges, one obtains
sin
2

13
= cos2
12
& 0:22.
7
The limiting values quoted in this paragraph are obtained for
the best t values of the neutrinos oscillation parameters and for
sin
2
2
13
0:14.
S. T. PETCOV, H. SUGIYAMA, AND Y. TAKANISHI PHYSICAL REVIEW D 80, 015005 (2009)
015005-6
cos
21
1
c
4
13
2BR
IH
ee
2c
4
13
c
2
21
s
2
21
: (25)
C. QD spectrum, m
1;2;3
* 0:1 eV
It is not difcult to convince oneself that the branching
ratios of interest for the QD spectrum to a good approxi-
mation can take values in the following intervals:
cos
2
2
12
=3 & BR
QD
ee
& 1=3 or 0:03 & BR
QD
ee
& 0:33, 0 &
BR
QD
e
& 0:46, cos
2
2
23
=3 & BR
QD

& 0:33. Actually, we


have up to small corrections BR
QD
ee
(2=3)BR
IH
ee
. For the
Majorana phase
21
in this case we get
cos
21
1
c
4
13
3BR
QD
ee
2c
4
13
c
2
21
s
2
21
: (26)
Given
21
and a sufciently large s
13
, information about
the Dirac phase and the Majorana phase
31
can be
obtained from the knowledge of BR
QD
e
and BR
QD

. If,
however, a stringent limit on s
13
will be obtained,
31
can be determined using BR
QD

and the knowledge of


21
.
It is clear from the above simple analysis that the mea-
surement of the three branching ratios BR
ee
, BR
e
, and
BR

would provide information about the type of neu-


trino mass spectrum and the Majorana phases. If, for
instance, it is experimentally established that BR
ee
>
0:01, the neutrino mass spectrum of NH type will be
excluded. The spectrum can either be of IH or QD type.
If in addition BR

is determined to be BR

>0:14, the
IH spectrum will be ruled out. If, however, the neutrino
mass spectrum will turn out to be QD, it will be very
difcult (if not practically impossible) to distinguish be-
tween the spectrum with NO and that with IO, i.e. to get
information about the sign of m
2
31
.
Consider next the more general case of m
0
having an
arbitrary value. First, let us set
13
0 for simplicity. We
will consider the case of
13
0 later. For
13
0, the
main uncertainty in the prediction of }M
ee
} comes from the
lack of knowledge of m
0
and
21
. Note that in this case
BR
ee
and BR
e
are independent of
31
, similar to M
ee
.
8
Knowing these two branching ratios allows us to determine
m
0
and cos
21
. In the case of spectrum with NO we have
for the lightest neutrino mass
m
2
1

(m
2
21
+m
2
31
)(2c
2
23
BR
ee
+BR
e
) 2s
2
12
c
2
23
m
2
21
2c
2
23
6c
2
23
BR
ee
3BR
e
;
(27)
cos
21

m
1

m
2
1
+m
2
21
q
+
m
2
21
(2c
2
12
c
2
23
BR
ee
s
2
12
BR
e
) m
2
1
BR
e
2c
2
12
s
2
12
m
1

m
2
1
+m
2
21
q
(BR
e
+2BR
ee
c
2
23
)
:
(28)
Equation (27) is valid also for the second to lightest neu-
trino mass m
1
in the case of spectrum with IO. The ex-
pression for cos
21
obviously cannot be used to determine
cos
21
for m
1
0: for
13
0 and m
1
0, neither BR
ee
nor BR
e
depend on
21
. In the case of spectrum with IO
we obtain
m
2
3

(2m
2
23
m
2
21
)(2c
2
23
BR
ee
+BR
e
) 2c
2
23
(m
2
23
c
2
12
m
2
21
)
2c
2
23
6c
2
23
BR
ee
3BR
e
; (29)
cos
21
1
BR
e
2c
2
12
s
2
12
(BR
e
+2BR
ee
c
2
23
)
+O

m
2
21
m
2
23

:
(30)
As can be expected, for m
2
1
>m
2
21
, the expression for
cos
21
in the case of spectrum with NO coincides with that
for spectrum with IO. Using Eq. (27), we get a universal
expression for }M
ee
} valid for both types of spectrum
with NO and IO and any hierarchy between neutrino
masses:
}M
ee
}
2

X
i
m
2
i

BR
ee

2c
2
23
m
2
31
2c
2
23
(3s
2
12
1)m
2
21
2c
2
23
6c
2
23
BR
ee
3BR
e
BR
ee
(31)

sgn(m
2
31
) BR
ee
1 3BR
ee
3BR
e
2:4 10
3
eV
2
; (32)
where in the last equation we have used the best t value of

23
and have neglected the term (3s
2
12
1)m
2
21
=m
2
31
.
Note that the denominator in the expression for }M
ee
},
Eq. (31), does not go through zero since we have
2c
2
23
BR
ee
+BR
e

2c
2
23
(m
2
1
+s
2
12
m
2
21
)
3m
2
1
+m
2
21
+m
2
31
;
NO spectrum;
(33)
2c
2
23
BR
ee
+BR
e

2c
2
23
(m
2
3
+m
2
23
)
3m
2
3
+2m
2
23
;
IO spectrum;
(34)
where we have neglected terms (m
2
21
=m
2
23
) in the
second equation. We see that in the QD region, where
8
BR
e
is independent of
31
as well, but this mode is more
difcult to measure than the two modes we are discussing.
NEUTRINOLESS DOUBLE BETA DECAY AND . . . PHYSICAL REVIEW D 80, 015005 (2009)
015005-7
}M
ee
} has a relatively large value, one has (2c
2
23
BR
ee
+
BR
e
) (2=3)c
2
23
(1 +m
2
23
=(3m
2
0
)).
It follows from Eq. (32) that }M
ee
} >0:05 eV

}m
2
31
}
q
can be predicted without the knowledge of
sgn(m
2
31
), if the collider experiments show that the
branching ratios BR
ee
and BR
e
satisfy

4c
2
23
3
BR
ee
+
2c
2
23
3
*BR
e
*
8c
2
23
3
BR
ee
+
2c
2
23
3
: (35)
If indeed sin
2

13
is negligibly small and these conditions
are satised by the measured BR
ee
and BR
e
, a positive
result can be expected in the next generation of
()
0
-decay experiments having a sensitivity to }M
ee
}
0:05 eV. Note that the magnitude of the left and right sides
of the inequality is very sensitive to the value of c
2
23
. Note
also that these conditions do not depend explicitly on
BR

. For this reason we will rst obtain constraints on


}M
ee
} using only the branching ratios BR
ee
and BR
e
.
Next, we analyze the case of
13
0 numerically. We
calculated BR
ee
, BR
e
, and M
ee
by using }m
2
31
}, m
2
21
,
and sin
2
2
12
given in Eq. (11) and (12). We allow m
0
to
vary in the interval in Eq. (17), while the other parameters
are varied in the following ranges reecting the uncertain-
ties in their knowledge or lack of any constraints:
sin
2
2
23
>0:94; sin
2
2
13
<0:14;
;
21
;
31
0 2:
(36)
Later we will present results for the prospective smaller
uncertainties in sin
2
2
23
and sin
2
2
13
, corresponding to
sin
2
2
23
>0:99 and sin
2
2
13
<0:04.
In Fig. 2, we show the regions in the BR
ee
BR
e
plane
where we denitely have }M
ee
} 0:05 eV or }M
ee
} <
0:05 eV. More specically, the solid (red) line determines
the complete allowed region in the HTM, corresponding to
m
2
21
, }m
2
31
}, and sin
2
2
12
given in Eqs. (11) and (12),
and values of m
0
, sin
2
2
23
, sin
2
2
13
, ,
21
, and
31
,
which were allowed to vary in the ranges specied in
Eqs. (17) and (36). The dashed blue (dash-dotted green)
lines determine the black (grey) regions where }M
ee
} is
denitely larger (smaller) than 0.05 eV in the HTM when
m
0
, sin
2
2
23
, sin
2
2
13
, ,
21
, and
31
, are varied within
the indicated intervals [i.e. Eqs. (17) and (36)]. For values
of BR
ee
and BR
e
from the region depicted in white (and
located between those shown in black and in grey), the
determination of }M
ee
} is not unambiguous: both values of
}M
ee
} 0:05 eV and }M
ee
} <0:05 eV are possible. This
degeneracy can be lifted to certain extent, but not com-
pletely, by using additional information on BR

(see
further). The dotted black line in Fig. 2 corresponds to
BR
ee
+BR
e
1. We show it only to indicate the bound-
ary of the region of possible values of BR
ee
and BR
e
: the
area above this line is unphysical. The results in Fig. 2 are
obtained without using the possible additional data on
9
BR

. The left and middle panels correspond to NO and


IO spectrum, respectively, while the results shown in the
right panel were obtained assuming that the sgn(m
2
31
) (i.e.
the type of the neutrino mass spectrum) is unknown. The
black area where }M
ee
} is, e.g. denitely larger than 0.05 eV
in the right panel corresponds to the intersection of the
black areas in left and middle panels. Note that we can
have }M
ee
} * 0:05 eV also in the region shown in white
and located between the grey areas in the right panel of
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
=1
FIG. 2 (color online). Values of BR
ee
and BR
e
, for which }M
ee
} >0:05 eV (black areas limited by the dashed blue lines) or }M
ee
} <
0:05 eV (grey areas limited by the dash-dotted green lines) in the HTM. The solid (red) line shows the entire region of allowed values
of BR
ee
and BR
e
in the HTM (black and grey areas and the white area between the colored one). The results shown are obtained by
varying m
0
, sin
2
2
23
, sin
2
2
13
, ,
21
, and
31
, in the ranges given in Eqs. (17) and (36). The left and middle panels correspond to NO
and IO spectrum, respectively, while the right panel was obtained assuming that sgn(m
2
31
) is unknown. The dotted line corresponds to
BR
ee
+BR
e
1. The region above this line is unphysical. See text for further details.
9
We recall that in this analysis we do not use possible data on
BR

, BR

, and BR
e
.
S. T. PETCOV, H. SUGIYAMA, AND Y. TAKANISHI PHYSICAL REVIEW D 80, 015005 (2009)
015005-8
Fig. 2. This cannot be unambiguously predicted, however,
knowing only the values of BR
ee
and BR
e
which lie in the
white area.
Next we show how the results discussed above change
when we add information on BR

. The

decay
mode of H

is relatively easy to measure at LHC by


virtue of the two same-sign muons in the nal state. We
present results for BR

0, 0.1, 0.2, 0.3 in Figs. 36,


respectively, where BR

0 in practice corresponds to
BR

<0:01. When we quote a specic value of BR

x, we include an uncertainty of 0:01 in x, i.e. we use


BR

x 0:01 in the numerical calculations. The dot-


ted lines in Figs. 36 correspond to BR
ee
+BR
e
+
BR

1. We do not use this constraint: the line repre-


sents the boundary of the physical region of values of BR
ee
,
BR
e
, and BR

.
It is clear from Fig. 3 that the measurement of BR

can
improve the predictability of }M
ee
}: the relative magnitude
of the white degeneracy region, in general, is smaller
than in the case when no information on BR

is available.
This is not the case, however, for IOspectrumand values of
BR

0:1, 0.2 (Figs. 4 and 5, middle panels).


Figures 26 show also the regions (dash-dotted line)
where one denitely has }M
ee
} <0:05 eV. If the measured
values of BR
ee
, BR
e
, and BR

lie in one of these


regions, the observation of ()
0
decay in the next gen-
eration of experiments can be extremely challenging. Even
in such a case, however, searches for the ()
0
decay are
important and necessary also as a test of the HTM itself. If
the ()
0
decay is observed while the measured values of
the H

leptonic decay branching ratios imply, e.g. a


negative result of the searches for ()
0
decay, we will
be led to conclude that M
l
/
l
and h
l
/
l
are not directly related:
M
l
/
l

2
_
h
l
/
l
v

. Such a situation can arise, for instance, if


v

0, or in models with H
++
which is not an SU(2)
L
triplet, but, e.g. is a Y 4 singlet [73] with couplings to the
charged leptons given by h
l
/
l
(l
/
R
)
C
l
R
H
++
.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO, BR

=0
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO, BR

=0
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1 BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
, BR

=0
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
FIG. 3 (color online). The same as in Fig. 2, but assuming that the experimentally determined BR

0. The dotted line


corresponds to BR
ee
+BR
e
+BR

1; the region above the line is unphysical. See text for further details.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO, BR

=0.1
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO, BR

=0.1
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
, BR

=0.1
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
FIG. 4 (color online). The same as in Fig. 3, but for BR

0:1.
NEUTRINOLESS DOUBLE BETA DECAY AND . . . PHYSICAL REVIEW D 80, 015005 (2009)
015005-9
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO, BR

=0.3
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO, BR

=0.3
possible in HTM
M
ee
> 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
, BR

=0.3
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
FIG. 6 (color online). The same as in Fig. 3, but for BR

0:3.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO, BR

=0.2
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO, BR

=0.2
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
, BR

=0.2
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
FIG. 5 (color online). The same as in Fig. 3, but for BR

0:2.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
=1
FIG. 7 (color online). The same as in Fig. 2, but allowing sin
2
2
23
and sin
2
2
13
to vary in the following more narrow intervals:
sin
2
2
23
>0:99 and sin
2
2
13
<0:04. No information on BR

was used in deriving the results shown in the gure. See text for further
details.
S. T. PETCOV, H. SUGIYAMA, AND Y. TAKANISHI PHYSICAL REVIEW D 80, 015005 (2009)
015005-10
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO, BR

=0
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO, BR

=0
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
, BR

=0
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
FIG. 8 (color online). The same as in Fig. 7, but assuming that the experimentally determined BR

0. The dotted line


corresponds to BR
ee
+BR
e
+BR

1; the region above the line is unphysical.


0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO, BR

=0.2
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO, BR

=0.2
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
, BR

=0.2
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
FIG. 9 (color online). The same as in Fig. 8, but for BR

0:2.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
NO, BR

=0.3
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
IO, BR

=0.3
possible in HTM
M
ee
> 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
B
R
e

BR
ee
unknown sign of m
2
31
, BR

=0.3
possible in HTM
M
ee
> 0.05eV is predicted
M
ee
< 0.05eV is predicted
BR
ee
+BR
e
+BR

=1
FIG. 10 (color online). The same as in Fig. 8, but for BR

0:3.
NEUTRINOLESS DOUBLE BETA DECAY AND . . . PHYSICAL REVIEW D 80, 015005 (2009)
015005-11
We have performed the same analysis, but with reduced
uncertainties in sin
2
2
23
and sin
2
2
13
: sin
2
2
23
>0:99 and
sin
2
2
13
<0:04. The indicated precisions (or better ones)
in the determination of sin
2
2
23
and sin
2
2
13
are expected
to be achieved in the upcoming T2K [42] and reactor
antineutrino experiments Double CHOOZ [37], Daya
Bay [38], and RENO [39], respectively. The results of
the analysis are shown graphically in Figs. 710. The
notations in Figs. 710 are the same as in Figs. 26. We
see from Figs. 710 that improving the precision on
sin
2
2
23
and sin
2
2
13
leads to a noticeable reduction of
the regions of values of BR
ee
and BR
e
, for which it is
impossible to determine whether }M
ee
} 0:05 eV or
}M
ee
} <0:05 eV. As Fig. 8 demonstrates, the reduction
will be particularly signicant if the measured BR

<
10
2
(which we remind the reader corresponds to the case
denoted by us as BR

0).
V. CONCLUSIONS
We have investigated the connection between the
()
0
-decay effective Majorana mass }M
ee
}, and the
branching ratios of the decays H

-l

l
/
, l, l
/
e,
, of the doubly charged Higgs boson H

within the
HTM of neutrino mass generation. Our analysis was per-
formed within the version of the model with explicit break-
ing of the total lepton charge conservation, in which
H

-l

l
/
, l, l
/
e, , , are the dominant decay
modes of H

. In this model, the couplings of the doubly


charged Higgs eld H
++
to the avor neutrinos and
charged leptons are proportional to the elements of the
Majorana mass matrix of the (avor) neutrinos M
l
/
l
and
the branching ratios BR(H

-l

l
/
) are entirely deter-
mined by the elements of the PMNS matrix and neutrino
masses. The latter possibility is realized if the mass of the
doubly charged Higgs scalar does not exceed the mass of
the singly charged one, M
H
M
H
, and if the vacuum
expectation value of the neutral component of the Higgs
triplet eld satises v

& 1 MeV. The model under dis-


cussion was shown [79] to have a rich and physically
interesting phenomenology owing to the fact that the
physical doubly charged and singly charged Higgs elds,
H

and H

, can have masses in the range from


100 GeV to 1 TeV and thus can, in principle, be
produced and observed at LHC. More importantly, by
studying the decays H

-l

l
/
, l, l
/
e, , , it might
be possible to obtain information on the absolute neutrino
mass scale, on the type of neutrino mass spectrum and on
the Majorana CP violating phases present in the neutrino
mixing matrix [79].
In the present article we have investigated the possibility
to use the information on the neutrino mass spectrum and
the Majorana CP violating phases from the measurements
of the H

-l

l
/
decay branching ratios, BR(H

-
l

l
/
), l, l
/
e, , in order to obtain predictions for the
effective Majorana mass in neutrinoless double beta
[()
0
] decay, }M
ee
}. Among the different decay channels
H

-l

l
/
, l, l
/
e, , , the easier to observe and
measure the corresponding branching ratios with high
precision are those with two electrons (positrons), two
muons (antimuons), or an electron (positron) and a muon
(antimuon), e

, e

, and

, in the nal state. If


the mass of H

does not exceed approximately 400 GeV,


the branching ratios of the H

decays into e

, e

,
and

can be measured at LHC with a few percent


error [15].
Taking into account the current and prospective uncer-
tainties in the values of the neutrino mixing parameters
most relevant for the problem studiedthe atmospheric
neutrino mixing angle
23
and the CHOOZ angle
13
and
allowing the lightest neutrino mass and the CP violating
Dirac and Majorana phases to vary in the intervals [0,
0.3 eV] and j0; 2|, respectively, we have derived the
regions of values of BR(H

-e

) and BR(H

-
e

) for which we denitely have }M


ee
} 0:05 eV, or
}M
ee
} <0:05 eV. This is done for neutrino mass spectrum
with NO, IO, and in the case when the type of the spectrum
is not known. In what concerns the branching ratio
BR(H

), we have considered two cases:


(i) the possible data on BR(H

) is not used
as an additional constraint in the analysis, (ii) the possible
data on BR(H

) is included in the analysis. In


the latter case, results for several values of BR(H

) have been obtained.


Our results are presented graphically in Figs. 210. They
show that if the doubly charged Higgs bosons H

will be
discovered at LHC and at least the two branching ratios
BR(H

-e

) and BR(H

-e

) will be mea-
sured with a sufcient accuracy, one can obtain important
information on the ()
0
-decay effective Majorana mass
}M
ee
}. In the various cases considered, we have identied
the regions values of BR(H

-e

) and BR(H

-
e

), for which }M
ee
} is denitely bigger or smaller than
0.05 eV (Fig. 2). We have shown also that due to (i) the
uncertainties in the determination of sin
2
2
23
and sin
2
2
13
,
(ii) the absence of data on the CP violating phases in the
neutrino mixing matrix, and (iii) the existing rather loose
upper bound on the absolute neutrino mass scale, there
exist also noticeable regions of values of BR(H

-
e

) and BR(H

-e

) for which it is impossible


to determine unambiguously whether }M
ee
} 0:05 eV or
}M
ee
} <0:05 eV (Fig. 2). This degeneracy can be partially
lifted by using the additional information from a measure-
ment of BR

(Figs. 36).
The same analysis was performed with reduced uncer-
tainties in sin
2
2
23
and sin
2
2
13
corresponding to
sin
2
2
23
>0:99 and sin
2
2
13
<0:04. The results are pre-
sented graphically in Figs. 710. They show that improv-
ing the precision on sin
2
2
23
and sin
2
2
13
leads to a
noticeable reduction of the regions of values of BR
ee
and
BR
e
for which it is impossible to determine whether
S. T. PETCOV, H. SUGIYAMA, AND Y. TAKANISHI PHYSICAL REVIEW D 80, 015005 (2009)
015005-12
}M
ee
} 0:05 eV or }M
ee
} <0:05 eV. The reduction will
be particularly signicant if the measured BR

<10
2
.
ACKNOWLEDGMENTS
This work was supported in part by the INFN under the
program Fisica Astroparticellare, by the European
Network of Theoretical Astroparticle Physics ILIAS/N6
(Contract No. RII3-CT-2004-506222) and by World
Premier International Research Center Initiative (WPI
Initiative), MEXT, Japan. S. T. P. acknowledges with grate-
fulness the hospitality and support of IPMU, University of
Tokyo, where part of the work on the present article was
done.
[1] W. Konetschny and W. Kummer, Phys. Lett. 70B, 433
(1977).
[2] T. P. Cheng and L. F. Li, Phys. Rev. D 22, 2860 (1980).
[3] J. Schechter and J. W. F. Valle, Phys. Rev. D 22, 2227
(1980).
[4] G. B. Gelmini and M. Roncadelli, Phys. Lett. 99B, 411
(1981).
[5] E. Ma, M. Raidal, and U. Sarkar, Phys. Rev. Lett. 85, 3769
(2000); Nucl. Phys. B615, 313 (2001).
[6] E. J. Chun, K. Y. Lee, and S. C. Park, Phys. Lett. B 566,
142 (2003).
[7] A. G. Akeroyd, M. Aoki, and H. Sugiyama, Phys. Rev. D
77, 075010 (2008).
[8] J. Garayoa and T. Schwetz, J. High Energy Phys. 03
(2008) 009.
[9] M. Kadastik, M. Raidal, and L. Rebane, Phys. Rev. D 77,
115023 (2008).
[10] F. del Aguila and J. A. Aguilar-Saavedra, Nucl. Phys.
B813, 22 (2009).
[11] A. Aranda, J. Hernandez-Sanchez, and P. Q. Hung, J. High
Energy Phys. 11 (2008) 092.
[12] B. Pontecorvo, Sov. Phys. JETP 6, 429 (1957) [Zh. Eksp.
Teor. Fiz. 33, 549 (1957)]; Sov. Phys. JETP 7, 172 (1958)
[Zh. Eksp. Teor. Fiz. 34, 247 (1957)]; Sov. Phys. JETP 26,
984 (1968) [Zh. Eksp. Teor. Fiz. 53, 1717 (1967)]; Z.
Maki, M. Nakagawa, and S. Sakata, Prog. Theor. Phys. 28,
870 (1962).
[13] S. M. Bilenky, J. Hosek, and S. T. Petcov, Phys. Lett. 94B,
495 (1980).
[14] S. M. Bilenky and S. T. Petcov, Rev. Mod. Phys. 59, 671
(1987).
[15] A. Hektor, M. Kadastik, M. Muntel, M. Raidal, and L.
Rebane, Nucl. Phys. B787, 198 (2007); P. Fileviez Perez,
T. Han, G. y. Huang, T. Li, and K. Wang, Phys. Rev. D 78,
015018 (2008).
[16] J. Schechter and J. W. F. Valle, Phys. Rev. D 25, 774
(1982).
[17] M. A. Diaz, M. A. Garcia-Jareno, D. A. Restrepo, and
J. W. F. Valle, Nucl. Phys. B527, 44 (1998).
[18] M. C. Chen, Phys. Rev. D 71, 113010 (2005).
[19] T. Blank and W. Hollik, Nucl. Phys. B514, 113 (1998); M.
Czakon, M. Zralek, and J. Gluza, Nucl. Phys. B573, 57
(2000); M. Czakon, J. Gluza, F. Jegerlehner, and M.
Zralek, Eur. Phys. J. C 13, 275 (2000); J. R. Forshaw,
D. A. Ross, and B. E. White, J. High Energy Phys. 10
(2001) 007; M. C. Chen and S. Dawson, Phys. Rev. D 70,
015003 (2004);M. C. Chen, S. Dawson, and T. Krupov-
nickas, Int. J. Mod. Phys. A 21, 4045 (2006); Phys. Rev. D
74, 035001 (2006).
[20] J. F. Gunion, J. Grifols, A. Mendez, B. Kayser, and F. I.
Olness, Phys. Rev. D 40, 1546 (1989).
[21] F. Cuypers and S. Davidson, Eur. Phys. J. C 2, 503 (1998).
[22] A. Merle and W. Rodejohann, Phys. Rev. D 73, 073012
(2006).
[23] J. C. Pati and A. Salam, Phys. Rev. D 10, 275 (1974); 11,
703(E) (1975); R. N. Mohapatra and J. C. Pati, Phys. Rev.
D 11, 2558 (1975); G. Senjanovic and R. N. Mohapatra,
Phys. Rev. D 12, 1502 (1975).
[24] R. N. Mohapatra and G. Senjanovic, Phys. Rev. Lett. 44,
912 (1980).
[25] A. G. Akeroyd and M. Aoki, Phys. Rev. D 72, 035011
(2005).
[26] S. M. Bilenky, S. Pascoli, and S. T. Petcov, Phys. Rev.
D64, 053010 (2001); 64, 113003 (2001).
[27] S. T. Petcov, Nucl. Phys. B, Proc. Suppl. 143, 159 (2005).
[28] J. Schechter and J. W. F. Valle, Phys. Rev. D 22, 2227
(1980); M. Doi et al., Phys. Lett. 102B, 323 (1981).
[29] B. T. Cleveland et al., Astrophys. J. 496, 505 (1998); W.
Hampel et al. (GALLEX Collaboration), Phys. Lett. B
447, 127 (1999); J. N. Abdurashitov et al. (SAGE
Collaboration), Zh. Eksp. Teor. Fiz. 122, 211 (2002) [J.
Exp. Theor. Phys. 95, 181 (2002)]; J. Hosaka et al. (Super-
Kamkiokande Collaboration), Phys. Rev. D 73, 112001
(2006); B. Aharmim et al. (SNO Collaboration), Phys.
Rev. Lett. 101, 111301 (2008); C. Arpesella et al.
(Borexino Collaboration), Phys. Rev. Lett. 101, 091302
(2008).
[30] Y. Ashie et al. (Super-Kamiokande Collaboration), Phys.
Rev. D 71, 112005 (2005); J. L. Raaf (Super-Kamiokande
Collaboration), J. Phys. Conf. Ser. 136, 022013 (2008).
[31] M. H. Ahn et al. (K2K Collaboration), Phys. Rev. D 74,
072003 (2006); P. Adamson et al. (MINOS
Collaboration), Phys. Rev. Lett. 101, 131802 (2008).
[32] M. Apollonio et al. (CHOOZ Collaboration), Eur. Phys. J.
C 27, 331 (2003).
[33] S. Abe et al. (KamLAND Collaboration), Phys. Rev. Lett.
100, 221803 (2008).
[34] A. Bandyopadhyay et al., Phys. Lett. B 608, 115 (2005).
[35] G. L. Fogli et al., Phys. Rev. D 78, 033010 (2008).
[36] T. Schwetz, M. Tortola, and J. W. F. Valle, New J. Phys.
10, 113011 (2008).
[37] F. Ardellier et al. (Double Chooz Collaboration), arXiv:
hep-ex/0606025.
[38] See, e.g., K. M. Heeger, 22nd International Conference on
NEUTRINOLESS DOUBLE BETA DECAY AND . . . PHYSICAL REVIEW D 80, 015005 (2009)
015005-13
Neutrino Physics and Astrophysics (Neutrino 2006), Santa
Fe, New Mexico, 2006 (unpublished), and the Daya Bay
homepage http://dayawane.ihep.ac.cn/.
[39] S.-B. Kim et al. (RENO Collaboration), J. Phys. Conf. Ser.
120, 052025 (2008).
[40] K. Anderson et al., arXiv:hep-ex/0402041, and the refer-
ences quoted therein.
[41] A. Bandyopadhyay et al. (ISS Physics Working Group),
arXiv:0710.4947, and the references quoted therein.
[42] Y. Itow et al. (T2K Collaboration), arXiv:hep-ex/0106019.
For an updated version, see http://neutrino.kek.jp/jhfnu/
loi/loi.v2.030528.pdf.
[43] M. V. Chizhov, M. Maris, and S. T. Petcov, arXiv:hep-ph/
9810501; J. Bernabeu, S. Palomares-Ruiz, and S. T.
Petcov, Nucl. Phys. B669, 255 (2003); S. Palomares-
Ruiz and S. T. Petcov, Nucl. Phys. B712, 392 (2005).
[44] S. T. Petcov, Phys. Lett. B 434, 321 (1998); 444, 584(E)
(1998).
[45] M. V. Chizhov and S. T. Petcov, Phys. Rev. Lett. 83, 1096
(1999); Phys. Rev. D 63, 073003 (2001).
[46] M. V. Chizhov and S. T. Petcov, Phys. Rev. Lett. 85, 3979
(2000).
[47] V. K. Ermilova et al., Short Notices of the Lebedev
Institute 5, 26 (1986); E. K. Akhmedov, Sov. J. Nucl.
Phys. 47, 301 (1988) [Yad. Fiz. 47, 475 (1988)]; P. I.
Krastev and A. Yu. Smirnov, Phys. Lett. B 226, 341
(1989).
[48] S. T. Petcov and M. Piai, Phys. Lett. B 533, 94 (2002); S.
Choubey, S. T. Petcov, and M. Piai, Phys. Rev. D 68,
113006 (2003).
[49] J. Learned et al., Phys. Rev. D 78, 071302 (2008); M.
Batygov et al., arXiv:0810.2580.
[50] L. Zhan et al., Phys. Rev. D 78, 111103 (2008); 79,
073007 (2009).
[51] S. M. Bilenky, M. D. Mateev, and S. T. Petcov, Phys. Lett.
B 639, 312 (2006).
[52] K. Eitel et al., Nucl. Phys. B, Proc. Suppl. 143, 197
(2005).
[53] F. Perrin, Comptes Rendus 197, 868 (1933); E. Fermi,
Nuovo Cimento 11, 1 (1934).
[54] V. Lobashev et al., Nucl. Phys. A719, C153 (2003).
[55] D. N. Spergel et al. (WMAP Collaboration), Astrophys. J.
Suppl. Ser.. 148, 175 (2003).
[56] M. Tegmark, Phys. Scr. T121, 153 (2005); S. Hannestad,
H. Tu, and Y. Y. Y. Wong, J. Cosmol. Astropart. Phys. 06
(2006) 025.
[57] W. Hu and M. Tegmark, Astrophys. J. Lett. 514, L65
(1999).
[58] C. Aalseth et al., arXiv:hep-ph/0412300.
[59] S. T. Petcov, New J. Phys. 6, 109 (2004); Phys. Scr. T121,
94 (2005); S. Pascoli and S. T. Petcov, arXiv:hep-ph/
0308034.
[60] V. A. Rodin et al., Nucl. Phys. A766, 107 (2006); A793,
213(E) (2007); A. Poves, NDM06 International
Symposium, Paris, 2006 (unpublished); E. Caurier et al.,
Phys. Rev. Lett. 100, 052503 (2008); Eur. Phys. J. A 36,
195 (2008).
[61] H. V. Klapdor-Kleingrothaus et al., Nucl. Phys. B, Proc.
Suppl. 100, 309 (2001).
[62] C. E. Aalseth et al., Phys. At. Nucl. 63, 1225 (2000).
[63] H. V. Klapdor-Kleingrothaus et al., Phys. Lett. B 586, 198
(2004).
[64] H. V. Klapdor-Kleingrothaus et al., Mod. Phys. Lett. A 16,
2409 (2001).
[65] A. S. Barabash (NEMO Collaboration), arXiv:0807.2336.
[66] C. Arnaboldi et al. (CUORICINO Collaboration), Phys.
Rev. C 78, 035502 (2008).
[67] F. Avignone, Nucl. Phys. B, Proc. Suppl. 143, 233 (2005).
[68] S. Pascoli and S. T. Petcov, Phys. Lett. B 544, 239 (2002);
580, 280 (2004).
[69] S. Pascoli, S. T. Petcov, and L. Wolfenstein, Phys. Lett. B
524, 319 (2002); S. Pascoli and S. T. Petcov, Yad. Fiz. 66,
472 (2003) [Phys. At. Nucl. 66, 444 (2003)].
[70] S. Pascoli, S. T. Petcov, and T. Schwetz, Nucl. Phys. B734,
24 (2006); S. Pascoli and S. T. Petcov, Phys. Rev. D 77,
113003 (2008).
[71] J. Schechter and J. W. F. Valle, Phys. Rev. D 25, 2951
(1982).
[72] R. N. Mohapatra and J. D. Vergados, Phys. Rev. Lett. 47,
1713 (1981); W. C. Haxton, S. P. Rosen, and G. J.
Stephenson, Phys. Rev. D 26, 1805 (1982); L.
Wolfenstein, Phys. Rev. D 26, 2507 (1982).
[73] A. Zee, Nucl. Phys. B264, 99 (1986); K. S. Babu, Phys.
Lett. B 203, 132 (1988).
S. T. PETCOV, H. SUGIYAMA, AND Y. TAKANISHI PHYSICAL REVIEW D 80, 015005 (2009)
015005-14

You might also like