Search for W' decaying to tau lepton and neutrino in proton-proton collisions at √s=8 TeV

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Search for W

decaying to tau lepton and neutrino in proton–proton collisions at √

s = ^{8 TeV}

.CMSCollaboration^

CERN,Switzerland

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

Articlehistory:

Received17September2015

Receivedinrevisedform20December2015 Accepted1February2016

Availableonline3February2016 Editor:M.Doser

Keywords:

CMS Physics W^decays

The firstsearchforaheavy chargedvectorbosoninthe finalstatewith atau leptonand aneutrino isreported, using19.7 fb⁻¹ofLHCdataat√

s=^{8 TeV.A signalwouldappearasan excessofevents} withhightransversemass,wherethestandardmodelbackgroundislow.Noexcessisobserved.Limits are set on amodel inwhichthe W^ decayspreferentially tofermions ofthe third generation.These resultssubstantiallyextendpreviousconstraintsonthismodel.Massesbelow2.0to2.7 TeVareexcluded, dependingonthe modelparameters.Inaddition,the existenceofaW^ bosonwithuniversal fermion couplingsisexcludedat95%confidencelevel,forW^massesbelow2.7 TeV.Forfurtherreinterpretation amodel-independentlimitonpotentialsignalsforvarioustransversemassthresholdsisalsopresented.

©^{2016CERNforthebenefitoftheCMS}Collaboration.PublishedbyElsevierB.V.Thisisanopenaccess articleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

1. Introduction

Newheavygaugebosonsarepredictedbyvariousextensionsof thestandardmodel(SM).Charged heavygaugebosons aregener- ally referred to as W^ [1]. Non-universal gauge interaction mod- els (NUGIM) [2–5] predict a larger W^-boson branching fraction tothe third generationof fermions.Searches fora W^ bosonde- cayingto a tau lepton andneutrino have never been performed before,while theelectron andmuonchannels havebeenstudied extensivelyat the Tevatron[6,7] and by the ATLAS andCMS ex- periments at the LHC [8,9]. This Letter describes a search for a W^ boson decayingto a tauleptonanda neutrinowiththe CMS detector[10]attheCERNLHC,usingproton–protoncollisionscol- lected in2012at acenter-of-massenergy of8 TeV.The dataset corresponds to an integrated luminosity of 19.7±⁰.5 fb⁻¹. The results are interpreted in the context of the sequential standard model (SSM) W^ boson [1]as well asan extended gauge group NUGIM[2,11,12].Thesignature ofaW^-bosoneventissimilarto that ofa W-boson eventinwhich theW boson is produced“off shell”withahighmass.Eventsofinterestare thoseinwhichthe only detectable products ofthe W^ decay forma single hadron- ically decaying tau (τh). The hadronic decays of the tau lepton are experimentallydistinctive because theyresultin low charged hadronmultiplicity,unlikeQCDjets,whichhavehighhadronmul- tiplicity,orotherleptonicW^decays,whichhavenone.Incontrast, thedecaysW^→τ ντ→^eνeντντ andW^→τ ντ→μνμντντ can-

 E-mailaddress:[email protected].

not be distinguished from W^→^eνe and W^→μν_μ, thus they suffer fromlow significanceand are notselected in thisanalysis butratherinthecorrespondingleptonic(e, μ)W^searches.

2. Physicsmodels

IntheSSM,theW^bosonisaheavyanalogueoftheWboson.It isanarrowresonancewithfermionicdecaymodesandbranching fractionssimilartothoseoftheSMWboson,withtheadditionof thedecayW^→^{tb, whichbecomesrelevantforW-bosonmasses} largerthan180 GeV.IftheW^bosonisheavy enoughtodecayto top andbottomquarks,theSSMbranching fractionforthedecay W^→τ ν is 8.5%.Under these assumptions, thetotal width of a 1 TeVW^bosonisabout33 GeV.DecaysoftheW^ bosonintoWZ bosonsdependonthespecificmodelassumptionsandareusually considered to be suppressedin theSSM,asassumedby the cur- rentsearchandbyprevioussearchesinotherfinalstates[9,13].If theW^interactswithleft-handedparticlesandright-handedanti- particles (V−A coupling), interferencewiththe SM W boson is expected[14–16].

Models with non-universal couplings predict an enhanced branchingfractiontothethirdgenerationoffermionsandexplain the large mass of the top quark. In the other model studied in this analysis, NUGIM [2,11,12], the weak SM SU(2)W group is a low-energylimit oftwogauge groups,a lightSU(2)l andaheavy SU(2)h, which coupleonly to the light fermions of the first two generationsandtotheheavyfermionsofthethirdgeneration,re- spectively. These two groups mix such that an SM-like SU(2)W andanextendedgroupSU(2)E exist.ThesecondSU(2)E extended http://dx.doi.org/10.1016/j.physletb.2016.02.002

0370-2693/©^{2016CERNforthebenefitoftheCMS}Collaboration.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

Fig. 1. Branchingfractions(left-handscaleandsolidlines)andtotalwidth(right- handscaleanddottedlines)forW^decaysintheNUGIM,ascalculatedinRefs.[2, 11,12].ForcotθE=^{1 thevaluesarethesameasthoseintheSSM,rescaledtoac-} commodatetheWH decaychannel.

gauge group gives riseto additional gauge bosons such asa W^. Themixingofthetwogaugegroupsisdescribedbyamixingan- gleoftheextendedgroup θE,whichmodifiesthecouplingtothe heavybosons.Hencethemixingchangestheproductioncrosssec- tion and, as illustrated in Fig. 1, the branching fractions of the W^.For cotθE^{3 theW boson decays to fermions ofthe third} generationonly,whereasatcotθE=^{1 the branchingfractionsare} identicaltothose oftheSSM,andthe W^ couples democratically toall fermions.For cotθE<1 thedecays into light fermions are dominant. Inthe NUGIM, thedecayinto WZ bosons isnegligible byconstruction.IneithertheSSMortheNUGIM,thepresenceofa W^-bosonsignalovertheW-bosonbackgroundcould beobserved inthedistributionofthetransversemass(MT)ofthe τhandthe missingtransverseenergy(E^miss_T ):

M_T=

2 p^τ_TE_T^miss[¹−^cosφ (τ,p^miss_T )], ⁽¹⁾

wherepτ_TdenotesthepTofthe τhandE^miss_T = |^pmiss_T |^,wherep^miss_T isdefinedas−pTofallreconstructedparticles.Theangleinthe transverseplane between^pmiss_T ^{andthedirectionof} τhisdenoted

φ (τ,^pmiss_T ).

3. Generationofbackgroundandsignalsamples

The major SM backgrounds are dominated by W and Z+jets productionand are generated using MadGraph 5.1 [17] (for on- shellW andZ+jetsbackgrounds), pythia 6.426 [18] (foroff-shell W,WW,WZ,andZZ backgrounds)and powheg 1.0[19–23](fort¯t andsingle t+jets).The tau decayis simulatedby tauola[24] for all samples.Forthe hadronization ofthe MadGraph background, pythiais used. The response of these events in the CMS detec- torissimulatedusing Geant4[25].Thebackgroundsareproduced at leading-order (LO), but reweighted to higher order cross sec- tions. For the main W+jets background, a differential cross sec- tion as a function of the mass of the W-boson decay products isreweighted,incorporatingnext-to-next-to-leading-order(NNLO) QCDandnext-to-leading-order(NLO)electroweakcorrections.The effectwithrespecttotheLOcalculationcorrespondstoaK-factor of 1.3 at a mass of 0.3 TeV and drops for higher masses to 1.1

for a mass of 1 TeV. The calculation uses Monte Carlo genera- tors mcsanc 1.01 [26] and fewz 3.1 [27], following the recom- mended combination from Ref. [28]. For the Z+jets background, theinclusiveNNLOQCDcrosssectioniscalculatedusing fewz.For t¯t events, the inclusive NNLO calculation from [29] is used. For the diboson (VV) backgrounds, inclusiveNLO QCD cross sections are calculated using mcfm 6.6[30].The background contribution from multijet events is estimated from control samples in data.

Thesignal eventsfortheSSMW^ aregeneratedwith pythia with NNLOcrosssectionsfrom fewz. TheNUGIMsignalsaregenerated with MadGraph 4.5.1[17] andhadronized with pythia.The par- ton distributionfunctions(PDFs)usedareCTEQ6L1 [31] forlead- ingordersimulationandCTEQ10[32] for(N)NLO simulation.The electroweak NLO calculation NNPDF 2.3 at NNLO QCD with and withoutQEDcontributions[33]areused.

4. TheCMSdetector

The central feature of the CMS apparatus is a superconduct- ing solenoidof6 m internal diameter,providinga magneticfield of3.8 T.Withinthesolenoidvolume area siliconpixelandstrip tracker,aleadtungstatecrystalelectromagneticcalorimeter(ECAL), andabrass andscintillatorhadroncalorimeter(HCAL),eachcom- posed of a barrel and two endcap sections. Muons are mea- suredingas-ionizationdetectorsembeddedinthesteelflux-return yokeoutsidethesolenoid. Extensive forwardcalorimetrycomple- ments thecoverage provided bythe barrelandendcapdetectors.

A particle-flow(PF)eventalgorithm[34]isusedtoreconstructthe events,identifythetaucandidatesanddeterminethemissingE_T. The algorithmreconstructs andidentifiessingle particles withan optimizedcombinationofall subdetectorinformation.The events are triggered by theCMS trigger system, which is split into two levels,afirst level(L1)composed ofcustomhardwareprocessors, anda high-level trigger(HLT) processor farm.For thisanalysisa

“jetplus E^miss_T ”triggerisused,withthresholdsof pT>80 GeV for the jet and E^miss_T >105 GeV, wherethe latteris seededatL1 in thecalorimeterwith E^miss_T above 40 GeV. Bothobjectsare recon- structedattheHLTlevelusingthePFeventreconstruction.A more detaileddescriptionoftheCMSdetector,togetherwithadefinition ofthecoordinatesystemusedandtherelevantkinematicvariables, canbefoundinRef.[10].

5. Reconstructionandidentificationofphysicsobjects

TaureconstructioninCMS[35]isappliedtojetsclusteredfrom PFobjects,usingtheanti-kT algorithmwithaparameter R=⁰.5.

Tau candidates must be distinguished from quark or gluon jets (QCD jetsin the following). The hadronic taudecays, τh, are re- constructedusingthe“hadron-plus-strips”(HPS)algorithm,which isbasedondecaymodesproceedingviaspecificintermediateres- onances,withacombinedbranchingfractionof 65%.Theyinclude modeswith eitherone orthree chargedhadrons, and up totwo neutralpions.Neutralpionsarereconstructedviatheirdecayinto pairs ofphotonsdetected intheECAL.The patternofenergyde- positionintheECALtypicallyoccursin“strips”,elongatedintheφ directionasaresultofinteractionsinthetrackermaterialandthe effectoftheaxialmagneticfield.The τhcandidateisreconstructed from stripsand chargedhadrons, which are combined using the mass ranges expectedfromthe intermediate resonances. A more detailed discussion of the HPS algorithm can be found in [35].

Thereconstruction ofhadronictaudecayshasbeenoptimizedfor tauleptonswithlargep_T wheredifferenttrackspotentiallymerge.

Thisoccursbecauseeitherthetrackreconstructionseedcannotbe resolvedorthetrackssharesomanyhitsthatonetrackcannot be reconstructed.Thisleads toreconstructeddecaymodeswithonly

twochargedhadrons(instead ofthree) beingacceptedtoaccom- modate the boostedtopology. The energy measurement of these high-pTobjectsisdominatedbythecalorimeterandthereforehas agood p_Tresolution.Theallowedmassrangefortheintermediate statereconstructionisbroadenedforhigh-pT tauleptons,tocom- pensate for the mass resolution. With these adaptations the tau reconstructionefficiencyisconstantat60%±^{6% for p}T>80 GeV, ashasbeencheckedinsimulationsuptop_T=^{3 TeV.Hadronictau} decaysidentified bythe HPSalgorithm arerequired tobe within thetrackingacceptance,|η_|<2.3,andthetaupTisrequiredtobe largerthan50 GeVtoreducethecontaminationfromQCDjets.Ad- ditionally the p_T oftheleading chargedhadron isrequired tobe larger than 20 GeV.Subsequently, τh isdistinguished fromother objectsthat could mimica taucandidate,such asQCD jets,elec- trons, or muons. The discriminator against QCD jets is the most important,sincetherateofQCDjetsattheLHCisseveralorders ofmagnitude larger than thetau productionrate. Discrimination is based on isolation criteria: no additional PF charged hadrons orphotonswith |pT|^{above 2 GeV are allowed inan isolation} cone of R=

(φ)²+ (η)²=⁰.3 (where φ is the azimuthal angleinradians and η isthe pseudorapidity)around the τhcan- didatedirection. Particle-flow objects are correctedfor additional collisions in the same bunch crossing (pileup). Charged hadrons areidentifiedaspileupobjectsbyvertexassociation.Neutralpar- ticlecandidatesare corrected by usingan average p_T subtraction fromthechargedhadronsidentifiedaspileupinaR=⁰.6 cone.

Details can be found in Ref. [35]. Discrimination against elec- tronsisobtainedusingamultivariatetechnique, basedonvarious tau, photon, trackandelectron properties.The muon discrimina- torsearchesforhitsinthemuonsystemassociatedwiththetrack ofthe τh candidate.Both discriminatorssuppresslightleptons by threeorders ofmagnitude, without a significantreduction ofthe tauefficiency.Eventsofinterest forthisanalysisarerequirednot tocontainidentifiedelectrons ormuons.Electronsarerequiredto satisfy shape and isolation criteria aswell as pT>20 GeV, and

|η_|<1.44 or1.56<|η_|<2.50.Muonsarerequiredtobeisolated andtohave pT>20 GeV and|η_|<2.4.

6. Analysisstrategy

Thestrategy ofthisanalysisis toselecta heavyboson decay- ingalmostatrestinto τhandE^miss_T .Inthetauchannel,theimpact ofthe interference betweenW^ andW bosons is expectedto be substantiallylowerthanthatpreviouslyfoundintheelectronand muonchannels [9].This occursbecausethesignal shapeof aW^ bosonwithhadronicallydecayingtauleptonsdoesnotshowaJa- cobianpeakstructure,becauseofthepresenceoftwoneutrinosin thefinalstate.Theinterferenceeffecthasthereforenotbeencon- sideredinthisanalysis.Forthe“jet+E^miss_T ”trigger,analysisthresh- oldsofpT>100 GeV fortheleadingjetand E^miss_T >140 GeV are appliedtoaccountfordifferencesoftriggerandreconstructeden- ergydefinitions.TheseanalysisthresholdsonthetaupTandE^miss_T , alongwiththekinematicselectionontheratioof pτ_T/E^miss_T ,yield an implicitlower threshold on thetransverse mass.The eventis requiredto contain one isolatedtau lepton.Twokinematic crite- ria are applied to select signal events: the ratio ofthe τh pT to the E^miss_T isrequiredtosatisfy0.7<pτ_T/E^miss_T <1.5 andtheangle

φ (τ,p^miss_T )hastobe greaterthan2.4radians.Thiseventselec- tionmainlyreducesthe backgroundinthelow-MT region, which hasthelargestbackground,whilethesignalefficiencyathighW^ massesisonlyreducedbyabout 5%.Theefficiencyandacceptance for a W^→τ ν event depend on the mass. For MW^=².2 TeV, 21% oftheeventspassallidentificationandselectioncriteria.This reduces to 17% for M_W=^{1 TeV, 7% for M}W^=⁰.5 TeV, and, at

Fig. 2. The MT distributionafterthefinalselection.Datapointswitherror bars showLHCdata.Thehorizontalerrorbaroneachpointindicatesthewidthofthe bin,whichis25 GeVforthefirstthreebinsand50 GeVforallotherbins.Thefilled histogramshowsthebackgroundestimatediscussedinthetext,andthehatched areatheuncertaintyinthisestimate.ThesignalshapesfordifferentSSMW^boson massesareshownasopenhistograms.ThecrosssectionforSSMMW^=^{500 GeV is} scaledby0.2.Intheratioplotthebin-widthisincreasedwhereneededtohaveat leastoneexpectedbackgroundeventineachbin.

higher masses, to 16% at M_W=^{3 TeV. The reduction for lower} masses occurs becauseof the change in shape of the MT distri- bution illustrated inFig. 2, while forhigher massesthe off-shell productionbecomes dominantandshiftstheeventsto lower M_T. From thesimulationofhadronictaueventswithlarge MT values, above the kinematicturn-on,42% are acceptedonce all selection andidentificationcriteriaare takenintoaccount. Thisacceptance isindependentoftheW^mass.FortheexamplecaseofW^→τ ν

with M_W=².2 TeV, the cross section calculated in the SSM is 13.5 fb.Thisyields 54.8predictedsignal eventsinthe τh+^EmissT final state, with the 21% acceptance quoted above for this M_W

value. The variation of the predicted SSM cross section with W^ masscanbeseeninFig. 3.

7. Backgroundestimation

The transverse mass distribution withthe observed data and expected backgroundevents anduncertainties is showninFig. 2 and Table 1. The dominant background,contributing almost two thirds ofthetotal,comesfromtheoff-shelltailoftheSM Wbo- son. This background is indistinguishable fromthe signal, and is estimated from simulation.The contributionfrom W→^e/μ +ν

events, in which the electron or muon is not identified, is also taken fromsimulation. The background contribution fromevents with one QCD jet falsely identified asa τh is suppressed by the pτ_T/E^miss_T requirement. Nonetheless it is the second largest back- groundforthissearchandisestimatedfromdatausingreference regions, separated fromthe signal region using the uncorrelated quantities, pτ_T/E^miss_T and τh isolation. The shape of the QCD jet backgroundisestimatedusingdataeventswithajetidentifiedas a τh, fulfilling all kinematic criteriadescribed earlier,apart from the isolation requirement. Its normalizationisbased on theratio