Dirac or Inverse Seesaw Neutrino Masses from Gauged B − L

Symmetry

Ernest Ma1, ∗ and Rahul Srivastava2, †

1 Department of Physics and Astronomy, University of California,
Riverside, California 92521, USA
2 The Institute of Mathematical Sciences, Chennai 600 113, India
arXiv:1504.00111v1 [hep-ph] 1 Apr 2015

Abstract

The gauged B − L symmetry is one of the simplest and well studied extension
of standard model. In the conventional case, addition of three singlet right handed
neutrinos each transforming as −1 under the B − L symmetry renders it anomaly
free. It is usually assumed that the B − L symmetry is spontaneously broken by a
singlet scalar having two units of B − L charge, resulting in a natural implementa-
tion of Majorana seesaw mechanism for neutrinos. However, as we discuss in this
proceeding, there is another simple anomaly free solution which leads to Dirac or
inverse seesaw masses for neutrinos. These new possibilities are explored along with
an application to neutrino mixing with S3 flavour symmetry.

I. INTRODUCTION

The nature of neutrinos i.e. whether they are Majorana or Dirac particles is one of the
most important open questions in neutrino physics. Answering this question is essential to
finding the underlying theory of neutrino masses and mixing. This issue can be potentially
resolved by neutrinoless double beta decay experiments (0νββ). Currently several ongoing
experiments are looking for signals of 0νββ but no such signal has been observed so far [1–3].
At present there is no compelling evidence from experiments or cosmological observations

∗
Electronic address: [email protected]
†
Electronic address: [email protected]
 2

in favor of either Dirac or Majorana nature of neutrinos. With our current understanding,
Dirac neutrinos are as plausible as Majorana ones.
However, theoretically Majorana neutrinos have received considerably more attention
than Dirac neutrinos. There are several mechanisms (e.g. seesaw mechanisms) which satis-
factorily explain smallness of neutrino masses if neutrinos are Majorana particles. On the
other hand Dirac neutrinos are not so well studied. There are only few models capable of
providing a natural explanation for smallness of Dirac neutrino masses. In this proceed-
ing we present one simple model for Dirac neutrinos with naturally small masses based on
gauged B − L symmetry. This proceeding is based on our work [4] and interested reader is
referred to it for further details.
The plan of this proceeding is as follows. In Section II we will briefly review the familiar
scenario of gauged B − L symmetry leading to the Majorana neutrinos and seesaw mech-
anism. The possibility of Dirac neutrinos in such a scenario will also be briefly discussed.
In Section III we will work out the new anomaly free solutions for gauged B − L symmetry
and show how it leads to Dirac neutrinos with naturally small masses. In Section IV we will
expand on our results and using S3 flavour symmetry will construct realistic neutrino mass
matrices consistent with present oscillation data. In Section V we will discuss the possibility
of inverse seesaw Majorana neutrinos masses arising from the new anomaly free solutions.
We will conclude in Section VI.

II. MAJORANA NEUTRINOS FROM B − L GAUGE SYMMETRY

Historically Baryon number (B) and Lepton numbers (Li ) were introduced to explain the
stability of proton and absence of lepton flavour changing processes. In Standard Model (SM)
the baryon and lepton numbers turn out to be accidentally conserved classical symmetries.
The B and L currents are anomalous and only the combination B −L is anomaly free. These
accidental symmetries need not be conserved by beyond standard model (BSM) physics. For
example if one adds a Majorana mass term for neutrinos then the B − L symmetry gets
broken by 2 units. The addition of right handed neutrinos provides the possibility to promote
this global B − L symmetry to a anomaly free gauged B − L symmetry.
Now addition of any new U(1)x gauge symmetry implies that one needs to cancel anoma-
lies [5–7]. The gauged B − L symmetry can potentially induce both gauge as well as gauge-
 3

gravitational anomalies. The triangular gauge anomalies arising from gauged U(1)B−L are:

• Tr U(1)B−L [SU(2)L ]2

• Tr U(1)B−L [U(1)Y ]2

• Tr (U(1)B−L )3

With the particle content of SM the first two anomalies are automatically canceled.
Moreover, if the right handed neutrinos νiR ; i = 1, 2, 3 transform as νiR ∼ −1 under U(1)B−L ,
U(1)3B−L = 0. The gauge - gravitational anomalies also vanish in this case as
P
then
−3(−1) = 3.
Lets briefly discuss the possibility of Dirac neutrinos in this scenario before discussing
the conventional case of Majorana neutrinos. The addition of right handed neutrinos with
−1 charge under U(1)B−L allows one to have gauge invariant Yukawa coupling for neutrinos.
The SU(3)C ⊗ SU(2)L ⊗ U(1)Y ⊗ U(1)B−L invariant Yukawa coupling for neutrinos is then
given by

X
− LνY = yij L̄iL Φ̂∗ νjR + h.c. (1)
i,j

where Φ̂∗ = iτ2 Φ∗ and Φ = (φ+ , φ0 )T is the SM Higgs doublet. Since the right as well
as left handed neutrinos transform non-trivially under the gauged U(1)B−L symmetry, this
implies that the Majorana mass term for νiR is forbidden and neutrinos are Dirac particles.
In this case the U(1)B−L symmetry remains unbroken 1 . However, in such a scenario the
smallness of neutrino masses requires unnaturally small Yukawa couplings and the model
does not provide any explanation for their smallness. For more details and variants of this
scenario we refer to [10–12].
A relatively better understanding of smallness of neutrino masses can be obtained if in
addition to the right handed neutrinos one also adds a singlet scalar χ transforming as χ ∼ 2
under U(1)B−L . In this case the SU(3)C ⊗ SU(2)L ⊗ U(1)Y ⊗ U(1)B−L invariant Yukawa
coupling for neutrinos is given by

1
The Z ′ gauge boson associated with U (1)B−L can get mass via Stuckelberg mechanism without breaking
the B − L symmetry [8–10].
 4

X 1X
− LνY = yij L̄iL Φ̂∗ νjR + c
fij ν̄iR χνjR + h.c. (2)
i,j
2 i,j

The spontaneous symmetry breaking (SSB) of χ then leads to breaking of B−L symmetry.
√
If hχi = u the right handed neutrinos get a Majorana mass Mij = 2fij u. If u >> v i.e. the
B − L symmetry breaking scale is far greater than the electroweak scale, then MR >> mD
leading to a natural implementation of Type I seesaw mechanism.
Before ending this section we like to remark that the gauged B − L symmetry can be
imbedded in other Beyond Standard Model scenarios e.g. it is an essential ingredient of
Left-Right symmetric models. It can also be embedded in GUT groups e.g. SO(10).

III. DIRAC NEUTRINOS FROM GAUGED B − L SYMMETRY

In this section we look at the possibility of another simple choice of B − L charges for
right handed neutrinos which leads to anomaly free U(1)B−L gauge symmetry. Unlike the
previous case, let the 3 right handed neutrinos transform as νiR = (+5, −4, −4) under B − L
symmetry [4, 13–15]. Since νiR ∼ (+5, −4, −4), therefore

− (+5)3 − (−4)3 − (−4)3 = +3 (3)

−(5) − (−4) − (−4) = +3 (4)

Thus in this case also the model is free from gauge as well as gauge-gravitational anoma-
lies. Now the standard-model Higgs doublet (φ+ , φ0 )T does not connect νL with νR . There-
fore the neutrinos do not get mass from the standard electroweak symmetry breaking. To
generate neutrino masses let us add three heavy Dirac singlet fermions NL,R transforming
as −1 under B − L symmetry. They will not change the anomaly cancellation conditions
and the model will remain anomaly free. Also let us add a singlet scalar χ3 transforming as
+3 under B − L.
Now for νR2 and νR3 , (ν̄L , N̄L ) is linked to (νR , NR ) through the 2 × 2 mass matrix as
follows
 
0 m0
Mν,N =   (5)
m3 M
 5

where m0 comes from hφ0 i. Moreover, m3 comes from hχ3 i, due to the Yukawa coupling
N̄L νR χ3 . The invariant mass M is naturally large, so the Dirac seesaw [16] yields a small
neutrino mass m3 m0 /M.
In the conventional U(1)B−L model, χ2 ∼ +2 under B − L is chosen to break the gauge
symmetry, so that νR gets a Majorana mass and lepton number L is broken to (−1)L .
Here, χ3 ∼ +3 means that it is impossible to construct an operator of any dimension for a
Majorana mass term and L remains a conserved global symmetry, with νL,R and NL,R all
having L = 1.
Since νR1 ∼ +5 does not connect with νL or NL directly, there is one massless neutrino
in this case. The dimension-five operator N̄L νR3 χ∗3 χ∗3 /Λ is allowed by U(1)B−L and would
give it a small Dirac mass. Alternatively, one can add a second scalar χ6 ∼ 6 to the model
to account for mass of νR1 .

IV. THE S3 FLAVOUR SYMMETRY

The discussion in previous section was aimed primarily at mass generation for Dirac neu-
trinos. The U(1)B−L symmetry alone does not provide any explanation for the currently
observed PMNS mixing pattern. In this section we will generalize our discussion and will
construct phenomenologically viable lepton mass matrices. In order to understand the lep-
tonic family structure consistent with present neutrino oscillation data, we will make use of
the non-Abelian discreet symmetry group S3 .
The S3 group is the smallest non-Abelian discreet symmetry group and is the group of
the permutation of three objects. It consists of six elements and is also isomorphic to the
symmetry group of the equilateral triangle. It admits three irreducible representations 1, 1′
and 2 with the tensor product rules.

1 ⊗ 1′ = 1′ , 1′ ⊗ 1′ = 1, 2 ⊗ 1 = 2,

2 ⊗ 1′ = 2, 2 ⊗ 2 = 1 ⊕ 1′ ⊕ 2 (6)

In this proceeding we will use the complex representation of the S3 group [17, 18]. In the
complex representation, if
 6

       
φ1 ψ1 φ†2 ψ2†
 ,   ∈2 ⇒  ,   ∈2 (7)
φ2 ψ2 φ†1 ψ1†

then

φ1 ψ2 + φ2 ψ1 , φ†2 ψ2 + φ†1 ψ1 ∈1
φ1 ψ2 − φ2 ψ1 , φ†2 ψ2 − φ†1 ψ1 ∈ 1′
   
†
φψ φ ψ
 2 2,  1 2 ∈2 (8)
φ1 ψ1 φ†2 ψ1

With this brief summary of S3 group and its irreducible representations, we now move
on to constructing an S3 invariant lepton sector. The B − L charge and S3 assignment of
the fields for the lepton sector is as shown in Table I.

Fields B−L S3 Fields B−L S3

Le −1 1′ eR −1 1′
Lµ −1 1′ µR −1 1′
Lτ −1 1 τR −1 1
NL1 −1 1′ NR1 −1 1′
NL2 −1 1′ NR2 −1 1′
NL3 −1 1 NR3 −1 1
Φ 0 1 e
νR 5 1′
   
µ
νR χ2
  −4 2   3 2
τ
νR χ3

TABLE I: The B − L and S3 charge assignment for the fields.

where we denote the left handed lepton doublets by Lα = (νLα , lLα )T where α = e, µ, τ ; the
right handed charged leptons are denoted as eR , µR , τR and the right handed neutrinos as
νRe , νRµ , νRτ . Also, let us denote the heavy singlet fermions as NL,R
i
; i = 1, 2, 3. The “Standard
T
Model like” scalar doublet is denoted by Φ = (φ+ , φ0 ) and the singlet scalars are denoted
by χ2,3 .
 7

The S3 and B − L invariant Yukawa interaction LY can then be written as

LY = LLα lR + LLα NR + LNL NR + LNL νR (9)

where

LLα lR = ye′ L̄e Φ eR + y12

′
L̄e Φ µR + y21
′
L̄µ Φ eR + yµ′ L̄µ Φ µR + yτ L̄τ Φ τR
′
LLα NR = g11 L̄e Φ̂∗ NR1 + g12
′
L̄e Φ̂∗ NR2 + g21
′
L̄µ Φ̂∗ NR1 + g22
′
L̄µ Φ̂∗ NR2
+ g33 L̄τ Φ̂∗ NR3
LNL NR = M11 ′
N̄L1 NR1 + M12 ′
N̄ 1 N 2 + M ′ N̄ 2 N 1 + M22′
N̄L2 NR2 + M33 N̄L3 NR3
 L R  21 L R    
∗ ∗ ∗ ∗
f ′ χ 3  χ 3  f ′ χ 3  χ 3 
LNL νR = 11 N̄L1 νRe ⊗  + 21 N̄L2 νRe ⊗ 



⊗ ⊗
Λ χ∗2 χ∗2 Λ χ∗2 χ∗2
1 1
       
νRµ χ2 νRµ χ
′ 1
+ f12 N̄L ⊗    ⊗    ′ 2
+ f22 N̄L ⊗    ⊗  2 
νRτ χ3 νRτ χ3
1′ 1′
   
νRµ χ
3
+ f33 N̄L ⊗    ⊗  2  (10)
νRτ χ3
1

In writing (10), we have used the notation Φ̂∗ = iτ2 Φ∗ = (φ0 , φ− )T . Here, yα are the
Yukawa couplings of the charged leptons whereas fij , gij and Mij denote the dimensionless
coupling constants between the leptons and the heavy fermions.
At this point we like to remark that in LY there is still a freedom to redefine a few fields
(i.e. the pairs NL1 − NL2 , NR1 − NR2 and eR − µR ) in a way that certain couplings can be made
equal to zero. For sake of later convenience we choose to use this freedom of field redefinition
′ ′ ′
to make f11 = M12 = y21 = 0. Moreover, we relabel the remaining non-zero couplings of
these redefined fields as fij′ → fij , gij′ → gij , Mij′ → Mij .
After symmetry breaking the scalar fields get VEVs hφ0 i = v, hχi i = ui ; i = 2, 3. Then
the mass matrix relevant to charged leptons is given by

 
y y 0
 e 12 
Ml = v  0 yµ 0  (11)
 
 
0 0 yτ
 8

This mass matrix can be readily diagonalized by bi-unitary transformation. In the limit
of ye << yµ we get

 
l −y12−1 l
θ12≈ tan ; me ≈ v ye cos θ12
yµ
l l


mµ ≈ v yµ cos θ12 − y12 sin θ12 ; mτ ≈ vyτ (12)
√
l
If y12 = yµ then maximal mixing is achieved i.e. θ12 = − π4 , with mµ = 2vyµ .
Also, the 6 × 6 mass matrix spanning (ν̄Le , ν̄Lµ , ν̄Lτ , N̄L1 , N̄L2 , N̄L3 ) and
(νRe , νRµ , νRτ , NR1 , NR2 , NR3 )T of neutrinos and the heavy fermions is given by

 
∗ ∗
0 0 0 g11 v g12 v 0
 

 0 0 0 g21 v ∗ g22 v ∗ 0 

 
 0 0 0 0 0 g33 v ∗ 
Mν,N =  (13)
 


 0 f12 u3 −f12 u2 M11 0 0  
 f21 ∗ ∗


 uu
Λ 2 3
f22 u3 −f22 u2 M21 M22 0  
0 f33 u3 f33 u2 0 0 M3

As remarked earlier, the mass terms Mij between the heavy fermions can be naturally
large, so we can block diagonalize the mass matrix assuming that fij , gij << Mij . The block
diagonalized mass matrix of light neutrinos is given by

−1
Mν = mNL νR MNL NR mLα NR
 
(g21 M11 −g11 M21 )f12 u2 (g22 M11 −g12 M21 )f12 u2 −f12 g33 u3
 M11 M22 M11 M22 M33 
(g21 M11 −g11 M21 )f22 u2 + f21 g11 M22 u6 (g22 M11 −g12 M21 )f22 u2 + f21 g12 M22 u6 −f22 g33 u3
= v∗ 
 
M11 M22 M11 M22 M33

 
(g21 M11 −g11 M21 )f33 u2 (g22 M11 −g12 M21 )f33 u2 f33 g33 u3
M11 M22 M11 M22 M33
(14)

u∗2 u∗3
where we have written u6 = Λ
. Also, the 3×3 mass matrices mLα NR , MNL NR and mNL νR
are obtained from the terms LLα NR , LNL NR and LNL νR respectively. This light neutrino mass
matrix can be further diagonalized by the bi-unitary transformation.
The neutrino masses and the mixing angles so obtained will be dependent on the specific
values of the coupling constants fij , gij , Mij as well as the VEVs v, ui ; i = 2, 3. In the
simplifying case of gij = g and Mij = M we get
 9

 
0 0 −f12 u3
gv ∗  
Mν =  f21 u6 f21 u6 −f22 u3  (15)
 
M  
0 0 f33 u3

Diagonalizing the mass matrix we have

  !
ν ν f12 f22
θ12 ≈ 0; θ13 ≈ tan−1 ; ν
θ23 ≈ tan−1 p
f33 2
f12 2
+ f33
p
2 2
2(f12 + f33 )f21 g|v|
mν1 ≈ 0; mν2 ≈ p
2 2 2
|u6|;
M f12 + f22 + f33
p
2 2 2
f12 + f22 + f33 g|v|
mν3 ≈ |u3 | (16)
M

Since, u6 << u3 , we have a normal hierarchy pattern with two nearly massless neutrinos
and one relatively heavy neutrino. Moreover, the massless neutrino will also gain small
mass, if any of the Mij ’s or gij ’s are not equal to M or g respectively. Also, if they deviate
significantly from these values then one can possibly recover degenerate or inverted hierarchy
patterns also. Now, if Ul and Uν are the mixing matrices of the charged leptons and neutrinos
respectively, then the PMNS mixing matrix is given by

UPMNS = Ul† Uν (17)

p
Taking y12 = yµ , f12 = − f233 and f22 = 2
f12 2
+ f33 ν
in (12), (16) we get θ23 l
= −θ12 = π
4
ν
and θ13 = tan−1 (− 12 ) which gives PMNS mixing angles consistent with present 3 − σ limits
of global fits obtained from experiments [19].
In our minimal model with only one doublet scalar, the quark sector can be accommo-
dated in a simple way if both the left handed quark doublets QiL = (uiL , diL )T , i = 1, 2, 3 and
the right handed quark singlets uiR , diR ; i = 1, 2, 3 transform as 1 of S3 . A better under-
standing of the quark sector can be obtained if, to our minimal model, we add more doublet
scalars transforming non-trivially under S3 . One such example for quark sector, albeit in
context of a different model for lepton sector, has already been worked out in [17, 18]. We
are currently working on a similar extension of our minimal model.
 10

V. INVERSE SEESAW

Apart from the Dirac neutrinos, other possibilities can also arise depending on the particle
content [4, 15]. For example instead of adding the singlet scalar χ3 ∼ 3 under U(1)B−L , one
can add two complex scalar fields χ2 ∼ 2 and χ6 ∼ 6 under U(1)B−L . In this case, νL is
not connected to νR1,R2 ∼ −4. It is connected however to NL,R through the mass matrix
spanning (ν̄L , N̄Rc , N̄L ) as follows:
 
0 m0 0
 
Mν,N =  m0 m′2 M  (18)
 
 
0 M m2
where m2 and m′2 come from the Yukawa couplings with χ2 . This leads to an inverse
seesaw [20–22], i.e. mν ≃ m20 m2 /M 2 . In the case of νR3 ∼ +5, the corresponding mass
matrix spanning (ν̄L , N̄Rc , N̄L , ν̄R3
c
) is given by

 
0 m0 0 0
 
 m m′ M 0 
 0 2
Mν,N = (19)


 0 M m2 m6 
 
0 0 m6 0
where m6 comes from the Yukawa coupling with χ6 . Thus νR3 also gets an inverse seesaw
mass ≃ m26 m′2 /M 2 which is the 4 × 4 analog of the 3 × 3 lopsided seesaw discussed in [23]. In
this scheme, νL and νR3 get small masses via inverse seesaw mechanism. Also, N1,2,3 become
heavy pseudo-Dirac fermions. However, νR1,R2 remain massless. They can be given mass by
adding extra scalars e.g. by adding a third scalar χ8 ∼ 8 under B − L.

VI. CONCLUSION AND FUTURE WORK

The idea that B − L should be a gauge symmetry has been around for some time.
However most of the work on gauged B − L symmetry has been done for the case of the
three right handed neutrinos transforming as - 1 under U(1)B−L . In this work we looked
at another possible anomaly free solution for gauged B − L interaction with the three right
handed neutrinos transforming as (+ 5, - 4, - 4) under U(1)B−L . We showed how these
assignments can be used to obtain seesaw Dirac neutrino masses, as well as inverse seesaw
 11

Majorana neutrino masses. We then showed that imposition of S3 flavour symmetry to the
first case can lead to realistic neutrino and charged-lepton mass matrices with a mixing
pattern consistent with experiments. In our model the B − L symmetry breaking scale can
be as low as in TeV range. This raises the possibility of testing it in future runs of LHC
or in other future colliders. We are planning to look for the phenomenological consequences
of our model. The phenomenology of the Z ′ boson would be of particular interest. The
cosmological implications of our model will also be interesting.

Acknowledgments

RS will like to thank the organizers for inviting and giving opportunity to present this
work at International Conference on Massive Neutrinos - 2015, Singapore. EM is supported
in part by the U. S. Department of Energy under Grant No. de-sc0008541.

[1] M. Auger et al. [EXO Collaboration], Phys. Rev. Lett. 109, 032505 (2012) [arXiv:1205.5608
[hep-ex]].
[2] M. Agostini et al. [GERDA Collaboration], Phys. Rev. Lett. 111, 122503 (2013)
[arXiv:1307.4720 [nucl-ex]].
[3] A. Gando et al. [KamLAND-Zen Collaboration], Phys. Rev. Lett. 110, no. 6, 062502 (2013),
arXiv:1211.3863.
[4] E. Ma and R. Srivastava, Phys. Lett. B 741, 217 (2015) [arXiv:1411.5042 [hep-ph]].
[5] E. Ma, Mod. Phys. Lett. A 17, 535 (2002) [hep-ph/0112232].
[6] E. Ma, Phys. Rev. Lett. 89, 041801 (2002) [hep-ph/0201083].
[7] M. C. Chen, A. de Gouvea and B. A. Dobrescu, Phys. Rev. D 75, 055009 (2007)
[hep-ph/0612017].
[8] E. C. G. Stueckelberg, Helv. Phys. Acta 11, 225 (1938).
[9] D. Feldman, P. Fileviez Perez and P. Nath, JHEP 1201, 038 (2012) [arXiv:1109.2901 [hep-ph]].
[10] J. Heeck, Phys. Lett. B 739, 256 (2014) [arXiv:1408.6845 [hep-ph]].
[11] J. Heeck and W. Rodejohann, Europhys. Lett. 103, 32001 (2013) [arXiv:1306.0580 [hep-ph]].
[12] J. Heeck, Phys. Rev. D 88, 076004 (2013) [arXiv:1307.2241 [hep-ph]].
 12

[13] J. C. Montero and V. Pleitez, Phys. Lett. B 675, 64 (2009) [arXiv:0706.0473 [hep-ph]].
[14] A. C. B. Machado and V. Pleitez, Phys. Lett. B 698, 128 (2011) [arXiv:1008.4572 [hep-ph]].
[15] A. C. B. Machado and V. Pleitez, J. Phys. G 40, 035002 (2013) [arXiv:1105.6064 [hep-ph]].
[16] P. Roy and O. U. Shanker, Phys. Rev. Lett. 52, 713 (1984) [Erratum-ibid. 52, 2190 (1984)].
[17] S. L. Chen, M. Frigerio and E. Ma, Phys. Rev. D 70, 073008 (2004) [Erratum-ibid. D 70,
079905 (2004)] [hep-ph/0404084].
[18] E. Ma and B. Melic, Phys. Lett. B 725, 402 (2013) [arXiv:1303.6928 [hep-ph]].
[19] F. Capozzi, G. L. Fogli, E. Lisi, A. Marrone, D. Montanino and A. Palazzo, Phys. Rev. D 89,
093018 (2014) [arXiv:1312.2878 [hep-ph]].
[20] D. Wyler and L. Wolfenstein, Nucl. Phys. B 218, 205 (1983).
[21] R. N. Mohapatra and J. W. F. Valle, Phys. Rev. D 34, 1642 (1986).
[22] E. Ma, Phys. Lett. B 191, 287 (1987).
[23] E. Ma, Mod. Phys. Lett. A 24, 2161 (2009) [arXiv:0904.1580 [hep-ph]].