Probing Higgs bosons via the type III seesaw mechanism at the LHC

Priyotosh Bandyopadhyay,1,3 Suyong Choi,2 Eung Jin Chun,1 and Kyungnam Min2

1Korea Institute for Advanced Study, Seoul 130-722, Korea2Department of Physics, Korea University, Seoul 136-713, Korea

3Department of Physics and Helsinki Institute of Physics, University of Helsinki, FIN-00014, Finland(Received 30 December 2011; published 20 April 2012)

We show that the type III seesaw mechanism opens up a promising possibility of searching the Higgs

boson in the b �b channel through the Higgs production associated with a charged lepton coming from the

decay of the triplet seesaw particle. In particular we look for the 2b signals with trileptons or same-sign

dileptons to construct the Higgs and the triplet fermion mass and calculate the reach with the integrated

luminosity of 10 fb�1 at the 14 TeV LHC.

DOI: 10.1103/PhysRevD.85.073013 PACS numbers: 14.80.Bn, 13.35.Hb, 13.85.Qk, 14.60.St

Finding the Higgs boson and thus verifying the elec-troweak symmetry breaking mechanism is a primarygoal of the LHC. The recent LHC data constrain theHiggs mass in a narrow range of 115–130 GeV [1]. Forsuch a low mass Higgs boson, its discovery relies on thecombination of several channels based on the gluonfusion production, in particular, gg ! h ! ��. Anotherinteresting channel is the associated Higgs boson pro-duction pp ! WðZÞh followed by the dominant Higgsdecay h ! b �b and leptonic decays of WðZÞ for which alow significance due to large backgrounds can be over-come by using subjet techniques in a boosted regime [2].Probing such a channel is important as it can provide anindependent information on the Higgs boson coupling togauge bosons and b quarks.

The purpose of this work is to explore another possibil-ity for the Higgs discovery which arises in the type IIIseesaw mechanism, where the origin of the observed neu-trino masses and mixing is attributed to SUð2Þ tripletfermions with hypercharge zero [3]. Such new particlescan be produced through the electroweak interaction andsubsequently decay to a lepton plus W, Z or h. Thesesignatures can be traced successfully to reconstruct thenew triplets and thus confirm the type III seesaw mecha-nism [4–9]. Among these type III seesaw signatures, wefocus on the Higgs production associated with a chargedlepton followed by the Higgs decay h ! b �b to show thatthis channel provides a promising search channel for thelow mass Higgs boson.

The type III seesaw mechanism introduces SUð2ÞL trip-let fermions with Y ¼ 0, � ¼ ð�þ;�0;��Þ, which can bewritten in a matrix form:

� ¼ �0ffiffiffi

2p

�þffiffiffi

2p

�� ��0

!

: (1)

Then the gauge invariant Yukawa terms are

L ¼ ½yiH" ��PLli þ H:c:� þ 14m� Tr½ ����; (2)

where li is the lepton doublet and H is the Higgsdoublet: li ¼ ð�i; eiÞL and H ¼ ðHþ; H0Þ. In the unitary

gauge, Hþ ¼ 0 and H0 ¼ vþ h=ffiffiffi

2p

with v ¼ 174 GeV,we get

LYukawa ¼ yi½ffiffiffi

2p

���PLei þ ��0PL�i þ H:c:� hffiffiffi

2p : (3)

Here we take only one generation of the triplet for ourillustration. Note that the neutrinos get a seesaw massm�

ij ¼ yiyjv2=m� which becomes of the order 0.1 eV for

yi � 10�6 andm� � 1 TeV. The neutrino Dirac mass, yiv,induces mixing between l and�. The mixing angles for theneutral and charged part are

��i� yiv

m�

and �li �ffiffiffi

2p yiv

m�

; (4)

respectively. Because of the l-� mixing (4), we get themixed gauge interaction as follows:

Lgauge ¼ �g��iWþ�

�

1ffiffiffi

2p ��0��PLei þ ��i�

�PR���

� g��iW��

�

1ffiffiffi

2p �ei�

�PL�0 þ �����PR�i

�

þ g��i2cW

Z�½ffiffiffi

2p

�����PLei þffiffiffi

2p

�ei��PL�

�

� ��0���5�i�: (5)

The electroweak production of these triplets happenthrough the W� and Z exchange at the LHC. The totalpartonic level production cross-section is given by [4]

� ¼ �ð3� �2Þ48�

NcsðV2L þ V2

RÞ; (6)

where, Nc ¼ 3, s ¼ parton level center of mass energy,

� ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

1� 4M2=sp

, and

PHYSICAL REVIEW D 85, 073013 (2012)

1550-7998=2012=85(7)=073013(5) 073013-1 � 2012 American Physical Society

VA ¼ 0 for q �q ! ���0;

VA ¼ Qqe2

sþ gqAg

2

s�M2Z

for q �q ! �þ��;

VA ¼ g2

s�M2W

�ALffiffiffi

2p for u �d ! �þ�0;

(7)

where gqA ¼ T3 � s2WQq is the Z coupling to quark q for

A ¼ fL; Rg.Thus, the electroweak production of the triplets at the

LHC, pp ! ���0;����, will leave bunch of multilep-ton final states followed by the triplet decays:

�� ! l�h; l�Z0; �W�; �0��

�0 ! �h; �Z0; l�W�:(8)

Among them, trilepton (3l) and same-sign dilepton(SSD) signals were shown to provide the most promisingchannels for the triplet search [9]. This is also true forprobing the Higgs boson in the type III seesaw. That is, themain channels for the Higgs search studied in this paperwill be 3l and SSD final states coming from ���0 asfollows:

l�hlWðl�Þ; l�h�ZðllÞ; l�ZðllÞ�h; l�hl�W�ðl�;jjÞ:

The current experimental limit on the heavy chargedlepton mass is rather week: mL� * 100 GeV [10], whichwe apply to our triplet mass. For the collider analysis, wehave chosen two benchmark points, BP1 and BP2, withm� ¼ 250 and 400 GeV, respectively, taking the Higgsmass of 120 GeV. The production cross-sections of thetriplet pairs at the 14 TeV LHC corresponding to BP1 andBP2 are listed in Table I. The branching ratios of thetriplet decay are calculated in Table II. The decay rate of�� ! �0�� is suppressed by a small mass splittingbetween �� and �0 arising from one-loop correction,while the other decay rates are proportional to y2i comingfrom the mixing (4) [4]. The size of the neutrino Yukawacoupling y can be quantified by the effective neutrino mass~m� � jyj2v2=m� where we take, in our analysis, y ¼ y1 ory2 denoting the electron or muon neutrino Yukawa cou-pling, respectively. When ~m� is sufficiently small, thetriplet decays will occur at large displaced vertices and

thus enable us to trace a displaced Higgs production freefrom backgrounds [11]. In this work, we do not look for thesignatures with displaced vertices as our analysis can beapplied to any value of ~m�. For our presentation, we set~m� ¼ 10 meV corresponding to the solar neutrino massscale.In this study, MADGRAPH [12] has been used for

generating parton-level events for the relevant processes.The Les Houches Accord event file interface [13] was thenused to pass the MADGRAPH-generated events to PYTHIA

[14]. We use CTEQ6L parton distribution function (PDF)[15,16]. In MADGRAPH we opted for the lowest order s

evaluation, which is appropriate for a lowest order PDFlike CTEQ6L. The renormalization/factorization scale in

MADGRAPH is set atffiffiffi

sp

. This choice of scale results in a

somewhat conservative estimate for the event rates. Initialstate radiation/final state radiation were switched on inPYTHIA for a realistic simulation. For this analysis we

have assumed a b-jet tagging efficiency of � 50% [17].For hadronic level simulation we have used PYCELL, the

TABLE I. Two benchmark production cross-sections.

Production cross-sections (fb)

m� 250 GeV 400 GeV

�þ�0 439.1 73.8

�þ�� 320.0 50.0

���0 221.8 32.3

TABLE II. Triplet branching ratios for ~m� ¼ 10 meV.

Decay modes Branching ratios

m� 250 GeV 400 GeV

�0 ! h� 0.17 0.22

�0 ! Z� 0.27 0.26

�0 ! W�l� 0.56 0.52

�� ! hl� 0.17 0.22

�� ! Zl� 0.27 0.26

�� ! W�� 0.55 0.52

�� ! �0�� 0.009 0.003

TABLE III. Number of events for 2bþ 3l; for mb-b within120� 25 GeV; and for mb-b-l within 250ð400Þ � 50 GeV.

2bþ 3lSignal Backgrounds

BP1 BP2 t�t t�tb �b t�tZ t�th VV t�tW116.89 40.32 5.0 1.77 31.53 9.86 0.0 8.67

mb-bSignal Backgrounds

BP1 BP2 t�t t�tb �b t�tZ t�th VV t�tW28.44 4.46 1.0 0.50 8.3 2.2 0.0 2.5

mb-b-lSignal Backgrounds

t�t t�tb �b t�tZ t�th VV t�tWBP1 61.89 2.0 0.1 5.9 1.5 0.0 0.9

BP2 5.19 0.0 0.0 1.5 0.06 0.0 0.5

BANDYOPADHYAY, et al. PHYSICAL REVIEW D 85, 073013 (2012)

073013-2

toy calorimeter simulation provided in PYTHIA, with thefollowing criteria:

(i) the calorimeter coverage is jj< 4:5 and the seg-mentation is given by ��� ¼ 0:09 0:09which resembles a generic LHC detector;

(ii) a cone algorithm with �R ¼ ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi

�2 þ��2p ¼ 0:5

has been used for jet finding;

(iii) pjetT;min ¼ 20 GeV and jets are ordered in pT;

(iv) leptons (‘ ¼ e, �) are selected with pT � 20 GeVand jj 2:5;

(v) and no jet should match with a hard lepton in theevent.

Higgs search with 2bþ 3l—For the Higgs event selec-tion, we first study the final state topology with at leasttwo tagged b-jets and at least three isolated leptons.Dominant standard model (SM) backgrounds are denotedin Table III. Here VV þ n� jets do not contribute as

potential backgrounds. The number of the signal and back-ground events for the two benchmark points are listed inTable III for an integrated luminosity of 10 fb�1 at the14 TeV LHC. As we can see from the Table III thesignificance for BP1 is �9� and that of BP2 is 4� at10 fb�1 of integrated luminosity.The b-jet pair invariant mass distribution, after the

event selection, is presented in Fig. 1 which shows thesmeared distribution for lower invariant mass due to the Zpeak contribution. Thus reconstructing Higgs mass doesnot increase the signal significance. Selecting events withinthe window of 95 GeV mb-b 145 GeV as in Table III,we get the required luminosity for 5� signal significance13:3 fb�1 for BP1, and 238 fb�1 for BP2. The correspond-ing decay length of the triplet is 0.06 (0.02) cm and 0.03(0.01) cm for the neutral (charged) triplet fermion for themasses 250 and 400 GeV, respectively.

0

2

4

6

8

10

12

100 150 200 250 300 350 400 450 500

Num

ber

of e

vent

s

mb-b-jets in GeV

BP1, mΣ=250 GeV

TotalBackground

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

100 150 200 250 300 350 400 450 500

Num

ber

of e

vent

s

mb-b-jets in GeV

BP2, mΣ=400 GeV

TotalBackground

FIG. 1 (color online). The b-b invariant mass from � 2bþ 3lfinal states.

0

2

4

6

8

10

12

14

100 150 200 250 300 350 400 450 500

Num

ber

of e

vent

s

mb-b-l in GeV

BP1, mΣ=250 GeV

TotalBackground

0

0.2

0.4

0.6

0.8

1

1.2

300 350 400 450 500 550 600 650 700

Num

ber

of e

vent

s

mb-b-l in GeV

BP2, mΣ=400 GeV

TotalBackground

FIG. 2 (color online). The b-b-l invariant mass from � 2bþ3l final states.

PROBING HIGGS . . .. III SEESAW MECHANISM . . . PHYSICAL REVIEW D 85, 073013 (2012)

073013-3

To check if the reconstructedHiggses are indeed from thetriplet decay, invariant mass distribution of the two b-jetswith one of the lepton among the tree isolated leptons isconstructed. In principle among the three leptons the rightone will peak at the triplet mass and the others will contrib-ute in the combinatorial background. For this, we select theb-jets within 60–150GeVof the invariant mass distribution.Figure 2 describes the invariant of distribution of b-b-l. Wethen select for 200 GeV mb-b-l 300 GeV for BP1 and350 GeV mb-b-l 450 GeV for BP2. From Fig. 2 wecan see that the distribution has a edge at the triplet mass.This is because one lowpT lepton coming from the decay ofgauge boson contributes in the lower end of the massspectrum. This makes the peak asymmetric in nature.From the signal numbers listed in Table III, one finds thesignificances of �7:3� for BP1 and�2� for BP2.Higgs search with 2bþ SSD—The final states with

same-sign dileptons are also free from severe standard

TABLE IV. Number of events for 2bþ SSD; for mb-b within120� 25 GeV; and for mb-b-l within 250ð400Þ � 50 GeV.

� 2b� jetþ SSDSignal Backgrounds

BP1 BP2 t�t t�tb �b t�tZ t�th VV t�tW127.38 29.09 24.0 7.5 41.6 29.0 0.0 41.4

mb-bSignal Backgrounds

BP1 BP2 t�t t�tb �b t�tZ t�th VV t�tW60.61 10.39 8.0 3.0 10.6 6.5 0.0 9.6

mb-b-lSignal Backgrounds

t�t t�tb �b t�tZ t�th VV t�tWBP1 117.37 4.0 0.35 7.00 2.67 0.0 2.50

BP2 11.74 0.0 0.0 1.71 0.12 0.0 1.05

0

5

10

15

20

25

100 150 200 250 300 350 400 450 500

Num

ber

of e

vent

s

mb-b-jets in GeV

BP1, mΣ=250 GeV

TotalBackground

0

2

4

6

8

10

12

100 150 200 250 300 350 400 450 500

Num

ber

of e

vent

s

mb-b-jets in GeV

BP2, mΣ=400 GeV

TotalBackground

FIG. 3 (color online). The b-b invariant mass from � 2bþSSD final states.

0

5

10

15

20

25

100 150 200 250 300 350 400 450 500

Num

ber

of e

vent

s

mb-b-l in GeV

BP1, mΣ=250 GeV

TotalBackground

0

0.5

1

1.5

2

2.5

300 350 400 450 500 550 600 650 700

Num

ber

of e

vent

s

mb-b-l in GeV

BP2, mΣ=400 GeV

TotalBackground

FIG. 4 (color online). The b-b-l invariant mass from � 2bþSSD final states.

BANDYOPADHYAY, et al. PHYSICAL REVIEW D 85, 073013 (2012)

073013-4

model backgrounds. The final decay modes from � thatcontribute to the process are hlWl, hlZl, ZlZl, ZlWl,respectively. Basically, the 3l events always carry same-sign dileptons and will contribute to this final state. Inaddition, signal acceptance is gained by requiring two ormore leptons. If one of the leptons in a trilepton event doesnot pass the acceptance, there is a chance that it will still beaccepted in the SSD selection.

We select the events with at least two b-tagged jets andat least two isolated same-sign leptons in the final states. InTable IV we present the corresponding signal numbers forthe two benchmark points and the SM backgrounds at10 fb�1 of integrated luminosity. The signal significancesare �8� for BP1 and 2:2� for BP2.

After analyzing the final state with SSD and 2b-jetswe construct the Higgs mass peak. We plot the invariantmass distribution of these to b-jets which peak aroundthe Z and Higgs mass in Fig. 3. Selecting events 95 GeV mb-b 145 GeV we get the signal numbers for the Higgsmass peak as listed in Table IV. The corresponding signalsignificance over SM backgrounds are 6:11� and 1:5� forBP1 and BP2, respectively. This shows the SSD events arebetter than the 3l events in probing the Higgs boson.

Similar to the 3l case, we reconstruct the triplet massfrom the selected Higgs events and the leptons in thefinal state. For that we take again the b-jets within themass window of 60–150 GeV and plot the invariant mass

distribution with the lepton in the final state in Fig. 4.Selecting the events within the windows of 200 GeV mb-b-l 300 GeV for BP1 and 350 GeV mb-b-l 450 GeV for BP2, we get the result in Table IV. In thiscase the significance really gets enhanced: ’ 10� for BP1and 3� for BP2.In conclusion, we examined b-jet pair signals from

Higgs in association with trileptons or same-sign dileptonsin the type III seesaw mechanism. These channels enoughsignificances over the standard model backgrounds, inparticular, in the high pT regime, and thus provide viablechannels for the light Higgs boson search. Early data of the14 TeV LHC will enable us to reconstruct the Higgs bosoncoming from the triplet decay through b-b and b-b-linvariant mass distributions for relatively lower tripletmass. On the other hand the reach goes down rapidly forhigher triplet mass due to the strong depletion of the tripletproduction cross-section which will be overcome by fur-ther accumulation of the luminosity.

E. J. C. was supported by Korea Neutrino ResearchCenter through National Research Foundation GrantNo. 2009-0083526. S. C. was supported by NationalResearch Foundation Grant No. 2009-0069251. We alsoacknowledge Roberto Francheschini for providing us withthe MADGRAPH model files implementing the type III see-saw model.

[1] ATLAS Collaboration, Report No. ATLAS-CONF-2011-163; CMS Collabortion, Report No. CMS-PAS-HIG-11-032.

[2] J.M. Butterworth, A. R. Davison, M. Rubin, and G. P.Salam, Phys. Rev. Lett. 100, 242001 (2008).

[3] R. Foot, H. Lew, X.G. He, and G. C. Joshi, Z. Phys. C 44,441 (1989).

[4] R. Franceschini, T. Hambye, and A. Strumia, Phys. Rev. D78, 033002 (2008).

[5] O. J. P. Eboli, J. Gonzalez-Fraile, and M. C. Gonzalez-Garcia, J. High Energy Phys. 12 (2011) 009.

[6] A. Arhrib, B. Bajc, D.K. Ghosh, T. Han, G.-Y. Huang, I.Puljak, and G. Senjanovic, Phys. Rev. D 82, 053004(2010).

[7] P. Bandyopadhyay, S. Choubey, and M. Mitra, J. HighEnergy Phys. 10 (2009) 012.

[8] T. Li and X.G. He, Phys. Rev. D 80, 093003 (2009).[9] F. del Aguila and J. A. Aguilar-Saavedra, Nucl. Phys.

B813, 22 (2009).

[10] K. Nakamura et al. (Particle Data Group), J. Phys. G 37,075021 (2010).

[11] P. Bandyopadhyay and E. J. Chun, J. High Energy Phys.11 (2010) 006.

[12] J. Alwall, P. Demin, S. de Visscher, R. Frederix, M.Herquet, F. Maltoni, T. Plehn, D. L. Rainwater, and T.Stelzer, J. High Energy Phys. 09 (2007) 028.

[13] J. Alwall et al., Comput. Phys. Commun. 176, 300(2007).

[14] T. Sjostrand, , L. Lonnblad, and S. Mrenna, arXiv:hep-ph/0108264.

[15] H. L. Lai, J. Huston, S. Kuhlmann, J. Morfin, F. Olness,J. F. Owens, J. Pumplin, and W.K. Tung, Eur. Phys. J. C12, 375 (2000).

[16] J. Pumplin, D. R. Stump, J. Huston, H.-L. Lai, P.Nadolsky, and W.-K. Tung, J. High Energy Phys. 07(2002) 012.

[17] H. Baer, V. Barger, G. Shaughnessy, H. Summy, and L.-T.Wang, Phys. Rev. D 75, 095010 (2007).

PROBING HIGGS . . .. III SEESAW MECHANISM . . . PHYSICAL REVIEW D 85, 073013 (2012)

073013-5