Search for W' -> tb in the lepton plus jets final state in proton-proton collisions at a centre-of-mass energy of root √s=8 TeV with the ATLAS detector

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Search for W

→ ^t b in ¯ the lepton plus jets final state in proton–proton collisions at a centre-of-mass energy of √

s = ^{8 TeV with the ATLAS} detector

.ATLASCollaboration^

a r t i c l e i n f o a b s t ra c t

Articlehistory:

Received15October2014

Receivedinrevisedform17February2015 Accepted23February2015

Availableonline25February2015 Editor:W.-D.Schlatter

Asearchfornewchargedmassivegaugebosons,calledW^,isperformedwiththeATLASdetectoratthe LHC,inproton–protoncollisionsatacentre-of-massenergyof√

s=^{8 TeV,usingadataset}corresponding toan integratedluminosity of20.3 fb⁻¹.Thisanalysissearches for W^ bosonsinthe W^→^tb decay¯ channelinfinalstateswithelectronsormuons,usingamultivariatemethodbasedonboosteddecision trees.Thesearchcoversmassesbetween0.5and3.0 TeV,forright-handedorleft-handedW^bosons.No significantdeviationfromtheStandardModelexpectationisobservedandlimitsaresetontheW^→ tb cross-section¯ times branchingratio and on the W^-boson effective couplingsas a functionof the W^-bosonmassusingtheCLs procedure.Foraleft-handed(right-handed)W^ boson,massesbelow1.70 (1.92) TeV areexcludedat95%confidencelevel.

PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

1. Introduction

ManyapproachestotheoriesbeyondtheStandardModel(SM) introducenew chargedvector currents mediated by heavy gauge bosons, usually called W^. For example, the W^ boson can ap- pearintheorieswithuniversalextradimensions,such asKaluza–

Klein excitations of the SM W boson [1–3], or in theories that extend fundamental symmetries of the SM and propose a mas- siveright-handed counterpartto the W boson[4–6]. LittleHiggs theories [7] also predict a W^ boson. The search for a W^ bo- son decaying to a top quark anda b-quark exploresmodels po- tentially inaccessible to searches for a W^ boson decaying into leptons [8–11]. For instance, in the right-handed sector, the W^ bosoncannot decaytoachargedlepton andaright-handedneu- trino if the latter has a mass greater than the W^-boson mass.

Also,inseveraltheoriesbeyondtheSMthe W^bosonisexpected to be coupled more strongly to the third generation of quarks than to the first and second generations [12,13]. Searches for a W^ boson decaying to the tb final¯ state¹ have been performed

 E-mailaddress:[email protected].

1 Forsimplicity,the notation“tb”¯ isusedtodescribeboththe W^{ +}→tb and¯ W^{ −}→ ¯tb processes.

at the Tevatron [14,15] in the leptonic top-quark decay chan- nel and at the Large Hadron Collider (LHC) in both the leptonic [16–18]andfullyhadronic[19]finalstates,excludingright-handed W^ bosons with massesup to 2.05 TeV at 95% confidence level (CL).

This Letter presents a search for W^ bosons using data col- lected in 2012 by the ATLAS detector [20] at the LHC, corre- sponding to an integrated luminosity of 20.3 fb⁻¹ from proton–

proton (pp) collisions at a centre-of-mass energy of 8 TeV. The searchisperformedinthe W^→^tb¯→ νbb decay¯ channel,where the lepton, , is either an electron or a muon, using a mul- tivariate method based on boosted decision trees. Right-handed and left-handed W^ bosons, denoted W_R^ and W_L^, respectively, are searched for in the mass range of 0.5 to 3.0 TeV. A gen- eralLorentz-invariantLagrangianisusedtodescribethecouplings of the W^ boson to fermions for various W^-boson masses [21, 22]. The mass of the right-handed neutrino is assumed to be larger than the mass of the W^ boson [23], thus allowing only hadronic decays of the W_R^ boson. In the case of a W^_L boson, leptonic decays are allowed and, since the signal has the same event signature asSM s-channel single top-quark production,an interference termbetween thesetwo processesis taken into ac- count[24].

http://dx.doi.org/10.1016/j.physletb.2015.02.051

0370-2693/PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

2. ATLASdetector

Charged particles in the pseudorapidity² range |η_|<2.5 are reconstructed with the inner detector, which consists of several layers of semiconductor detectors (pixel and microstrip), and a straw-tube transition–radiation tracker, the latter covering |η_|<

2.0.Theinnertrackingdetectorsystemisimmersedinahomoge- neous2 Tmagneticfieldprovidedby asuperconductingsolenoid.

Thesolenoidissurroundedby ahermetic calorimeterthatcovers

|η_|<4.9 and provides three-dimensional reconstruction ofparti- cle showers. The lead/liquid-argon electromagnetic compartment isfinelysegmentedfor|η|<2.5,whereitplays animportantrole in electron identification. Hadronic calorimetry is provided by a steel/scintillator-tilescalorimeterfor|η_|<1.7 andbyliquid-argon withcopper ortungsten absorbers end-capcalorimeters that ex- tend the coverage to |η|=⁴.9. Outside the calorimeter, air-core toroids provide the magnetic field for the muon spectrometer.

Threestationsofprecision drifttubes andcathode-stripchambers providean accuratemeasurementofthemuon trackcurvaturein theregion|η_|<2.7.Resistive-plateandthin-gapchambersprovide muontriggeringcapabilityupto|η_|=2.4.

3. Dataandsimulationsamples

Thedatausedforthisanalysiswererecordedusingunprescaled single-electron and single-muon triggers. After stringent data- quality requirements, the amount of data corresponds to an in- tegratedluminosityof20.3±⁰.6 fb⁻¹ [25].

The W_R^ and W_L^ signals are generated with MadGraph5[26]

using FeynRules [27,28] and the CTEQ6L1 [29] parton distribu- tionfunction(PDF)set. Pythia8[30]isusedforpartonshowering andhadronisation. Simulated samplesare normalised to next-to- leading order (NLO) QCD calculations [22] using K -factors rang- ing from1.15to 1.35depending on the massandhandedness of the W^ boson. The model assumes that the W^-boson coupling strength to quarks, g^, is the same as forthe W boson: g^_R=⁰ and g^_L=^{g (g}R=^{g and g}L=^{0) for} left-handed (right-handed) W^ bosons, where g is the SM SU(2)L coupling. The total width oftheleft-handed(right-handed) W^ bosonincreasesfrom17to 104 GeV (12 to 78 GeV) for masses between 0.5 and 3.0 TeV, where the decay to leptons is (is not) allowed [21]. In order to accountforthe effectoftheinterferencebetween W_L^-bosonand s-channelsingle top-quarkproductiondedicated pp→^W_L^/W→ tb¯→ νbb samples¯ are simulated,using MadGraph5,andassum- ing a destructive interferenceterm[24].In addition, samplesare generatedforvaluesof g^/g upto5.0,forseveralW^-boson(left- andright-handed)masshypotheses.

Top-quarkpair(tt)¯ andsingletop-quarks-channelandW t pro- cesses are simulated with the Powheg [31,32] generator, which uses a NLO QCD matrix element with the CT10 PDFs [29]. The parton shower and the underlying event are simulated using Pythia v6.4[33].The t-channel single-top-quark process is mod- elled using the AcerMC v3.8 [34] generator with the CTEQ6L1 PDFsand Pythia v6.4. The t¯t cross-section is calculatedat next- to-next-to-leadingorder(NNLO)inQCDincludingresummationof next-to-next-to-leadinglogarithmicsoftgluontermswith top++2.0 [35–41]. The single top-quark cross-sections are obtained from

2 ATLASusesaright-handed coordinatesystemwith itsoriginat thenominal interactionpointinthecentreofthedetectorandthez-axisalongthebeampipe.

Thex-axispointsfromtheinteractionpointtothecentreoftheLHCring,andthe y-axispointsupward.Cylindricalcoordinates(r,φ)areusedinthetransverseplane, φbeingtheazimuthalanglearoundthebeampipe.Thepseudorapidityisdefined intermsofthepolarangleθasη= −^{ln tan}(θ/2).Observableslabelled“transverse”

areprojectedintothex– y plane.

approximate NNLO calculations [42–44]. A top-quark mass of 172.5 GeV is assumed for the production of all simulated pro- cessesthatincludeatopquark.

The Alpgen leading-order multileg generator [45] with the CTEQ6L1 PDFsand Pythia v6.4isusedto generatevector bosons inassociationwithjets:W+jets (includingthecontributionsfrom W bb¯+^{jets, W cc}¯+^{jets and W c}+^{jets) and Z}+jets events. Di- boson samples (W W , Z Z , and W Z ), where at least one of the bosons decaysleptonically,aremodelledusing Herwig v6.52[46]

withtheCTEQ6L1PDFs.Thesingle-bosonanddibosonsimulation samplesarenormalisedtotheproductioncross-sectionscalculated atNNLO[47,48]andNLO[49]inQCD,respectively.

All generatedsamples are passed through afull simulation of theATLAS detector[50]basedonGEANT4 [51]andreconstructed using the sameprocedure as forcollision data.Simulated events include the effect of multiple pp collisions from the same and previousbunch-crossings(in-timeandout-of-timepileup)andare re-weighted tomatchtheconditionsofthedatasample (20.7in- teractionsperbunchcrossingonaverage).

4. Objectandeventselections

The search for W^→^tb events¯ relies on the measurement of the following objects: electrons, muons, jets, and the missing transverse momentum.Electrons areidentified asenergyclusters intheelectromagneticcalorimetermatchedtoreconstructedtracks intheinnerdetector[52,53].Electroncandidatesarerequiredtobe isolated,usingafixedcone-sizeisolationcriterion[54],fromother objectsintheeventandfromhadronicactivity,toreducethecon- taminationfrommis-reconstructedhadrons,electronsfromheavy- flavour decaysand photon conversions. Electrons are required to have transverse momentum, pT, above 30 GeV and |η| <2.47 with a veto on the barrel-endcap transition region in the range 1.37<|η_{| <}1.52.

Muonsareidentifiedusingthemuonspectrometerandthein- ner detector [55]. A variable cone-size isolation criterion [54,56]

isappliedtoreducethecontributionofmuonsfromheavy-flavour decays. Muon candidates are required to have pT>30 GeV and

|η| <².5.

Jets are reconstructed usingthe anti-kt algorithm [57] with a radius parameter R=⁰.4,usingtopologicalenergyclustersasin- puts [58,59]. Jets are calibrated using energy- and η-dependent correction factors derived fromsimulation andwithresidualcor- rectionsfrominsitumeasurements [60].Jetsarerequiredtohave pT>25 GeV and |η_|<2.5.Tosuppressjetsfromin-time pileup, at least 50% of the scalar p_T sum of the tracks associated with a jet is required to be from tracks associated with the primary vertex [61]. This requirement, called the “jet vertex fraction” re- quirement,isappliedonlyforjetswithp_T<50 GeV and|η|<2.4.

The identification of jets originating from the hadronisation of b-quarks(“b-tagging”)isbasedonpropertiesspecifictob-hadrons, such aslonglifetimeandlargemass.Thisanalysisusesa neural- network-basedcombinationofseveralhigh-performanceb-tagging algorithms[62].Thealgorithmhasanefficiencyof70%(20%,0.7%) forjetsoriginatingfromb-quarks(c-quarks,light-quark/gluon)as obtainedfromsimulatedtt events.¯

The missing transverse momentum, E^miss_T , is the modulus of the vector sumof thetransverse momentum incalorimetercells associated with topological clusters, and is further refined with object-levelcorrectionsfromidentifiedelectrons, muons,andjets [63,64].ThisanalysisrequireseventstohaveE^miss_T >35 GeV tore- ducethemultijetbackground.

Candidate events are requiredto have exactlyone lepton and twoorthreejetswithexactlytwoofthemidentifiedasoriginating from a b-quark (denoted 2-tag events). The multijet background

contribution is further reduced by imposing a requirement on the sum of the W -boson transverse mass,³ mT(W), and E^miss_T : mT(W)+^Emiss_T >60 GeV.Eventswithexactlytwo(three)jetspass- ingalltheaboveselectionsdefinethe2-jet(3-jet)channel.

Thesignal regionis definedbyselecting eventswherethere- constructed invariant mass of the t_{b system,}¯ _{mtb¯ (see definition} below),islargerthan330 GeV.Theacceptancetimesefficiencyfor theW^→^tb process¯ intheleptonplusjetsfinalstateis5.5%,2.2%

and2.1%(4.9%,2.2%,2.3%)fora W^_R(W_L^)bosonwithamassof1, 2or3 TeV,respectively.ThedropinacceptanceforhighW^-boson massesisduetotheleptonfailingtheisolationcriterionandtothe decreaseoftheb-taggingefficiency.Acontrolregionisdefinedby invertingtherequirementonthetb invariant¯ mass,m_tb¯<330 GeV, and is used to derive the normalisation of the W +^{jets back-} ground.

The method used to reconstruct the invariant mass of the tb system¯ in the selected sample proceeds as follows.The four- momentumofthetopquark isreconstructedby addingthefour- momentaofthe W bosonandofthe b-taggedjet that givesthe reconstructed invariant top-quarkmass closest to the value used forgeneration (172.5 GeV). Thereafter, thisb-tagged jet iscalled the“top-jet”and isassumedto be theb-jet fromthe top quark.

In this calculation the transverse momentum of the neutrino is givenbythe x- and y-componentsofthe E^miss_T vector, whilethe unmeasuredz-component of theneutrino momentumis inferred by imposing a W -boson mass constrainton the lepton–neutrino system[65].Thefour-momentumofthetb system¯ isthenrecon- structedby addingthefour-momenta ofthetop quark tothat of theremainingb-taggedjet.

5. Backgroundestimation

Thett,¯ single-top-quark,dibosonandZ+jets backgroundsare modelledusingthesimulationandarescaledtothetheorypredic- tionsoftheinclusivecross-sections.

Thebackgroundoriginatingfrommultijetevents,whereajetis misidentifiedasaleptonoranon-promptleptonappearsisolated (bothreferredtoasa“fake”lepton),isestimateddirectlyfromdata usingthematrixmethod[54].Theshapeandnormalisationofthe multijetbackgroundaredeterminedinboththeelectronandmuon channelsusingthismethod.

TheW+jets backgroundisalsomodelledusingthesimulation, butinthecaseofthe2-jetchanneltheeventyieldforthisprocess isderivedfromdatatoimprovethemodellinginthischannel.The numberof W+jets events is estimatedin the 2-jet control re- gionasthenumberofdataeventsobservedaftersubtractionofall non-W+jets background sources described above. This estimate isthenextrapolatedtothe2-jet signalregion usingtheW+^jets simulation.Forthe3-jetchanneltheW+jets backgroundisscaled tothetheoryprediction.

6. Analysis

The analysisstrategy relies on a multivariate approach, based ontheboosteddecisiontree(BDT)methodusingtheframeworkof TMVA[66],toenhancetheseparationbetweenthesignalandthe background.ForeachjetmultiplicityandW^-bosonhandedness,a separateBDTistrainedinthesignalregion.Forthebackground,a mixtureoftop-quark, W/Z+^{jets,dibosonandmultijetssamples,} all weighted according to their relative abundances, is used. The W^-bosonsample usedas signal in the BDT training andtesting

3 Defined as mT(W)=

(pT()+^EmissT )²− (^px()+^Emissx )²− (^py()+^Emissy )², whereE^miss_x andE^miss_y arethex- andy-componentsoftheE^miss_T vector.

phasesischosenatamassof1.75 TeV sincethisgivesthebestex- pected exclusionlimit onthe W^-bosonmass,compared toBDTs trained withother W^-boson mass samples.Thischoice also en- suresverygoodseparationbetweentheBDTshapesofsignaland backgroundforW^-bosonmassesof1 TeV andabove.Thisanaly- sis isthussensitive tothepresenceofa signalover awide mass range.

Ten (eleven) variables with significant separation power are identified in the 2-jet (3-jet) samples for the W^-boson search.

TheseareusedasinputstotheBDTs.Thelistofvariableschanges slightlydependingonthechiralityofthesignal.

A set offive variables is common to all fourBDTs. Two vari- ables,m_t¯bandthetransversemomentumofthereconstructedtop quark, pT(t), provide the best separation poweramong all those considered and are shown in Fig. 1. The other three common variables are: the angularseparation⁴ betweenthe jet associated withtheb-jetoriginatingfromthe W^bosonandthetop-jet(de- noted bt), R(b,bt);the transverseenergyof thetop-jet, ET(bt), andtheaplanarity.⁵

In addition, for the 2-jet channel, the following variables are used: theangularseparationbetweenthe top-jetandthe W bo- son, R(b_t,W),andtheηbetweenthe leptonandthetop-jet,

η(,bt). Forthe caseofthe right-handed W^-boson search the followingvariablesare alsoused:thesphericity;theangularsep- arationbetweentheleptonandtheb-jetoriginatingfromtheW^ boson, R(,b); the transverse momentum of thelepton, pT(). Fortheleft-handedW^-bosonsearch,threedifferentvariablesare chosen:theanglebetweenthetop-jetandthemissingtransverse momentum, φ (bt,E^miss_T ); the ratio of the transverse momenta of the top-jet and of the b-jet originating from the W^ boson, pT(bt)/pT(b),andmT(W).

Foreventswiththree jets, thefollowing variablesare used in addition to the commonset ofvariables: R(,bt); the spheric- ity; p_T(b); the invariant mass of the three jets m(b,b_t,j). Two more variables are used, for the right-handed case only: pT() and R(b,W), and for the left-handed case: φ (bt,E^miss_T ) and p_T(b_t)/p_T(b).

Fig. 2showstheexpectedBDToutputdistributions,normalised tounity, inthesignal regionfortheelectron andmuon channels combined, forseveral simulated right-handed W^-boson samples andfortheexpectedbackground.

7. Systematicuncertainties

Systematicuncertaintiescanaffecttheshapeandnormalisation oftheBDT outputdistributions.Theyare splitintothecategories describedbelow.

Objectmodelling: Themain uncertainty inthis category isdue to uncertaintiesonb-taggingefficiencyandmistaggingrates[68,69].

The resulting uncertainty on the event yield is 6% for the total backgroundcontributionand8–30%forthesignal.Thelargeuncer- taintiesonthe signalratesareduetoadditionalb-tagginguncer- taintiesforjetswith pT above300 GeV.Theseuncertaintiesrange from3% forb-jetswith p_T of50 GeV upto15% at500 GeV and 35%above.Theimpactissizeableforthesignalwherehigh-pTjets stemfromtheW^-bosondecay,inparticularwhenthe W^-boson massisabove1 TeV.Thejetenergyscaleuncertaintydependson thepTand ηofthereconstructedjetandincludestheuncertainty on the b-jet energy scale. It results in an uncertainty on event yieldsof1–6%forthesignaland1–4%forthebackground,depend- ingonthechannel.Thesystematicuncertaintyassociatedwiththe

4 DefinedasR=

(φ)²+ (η)².

5 Aplanarityandsphericityareeventshapevariablescalculatedfromthespheric- itytensoroftheleptonandjetmomenta[67].