Search for diboson resonances with boson-tagged jets in pp collisions at √s = 13 TeV with the ATLAS detector

Texte intégral




Search for diboson resonances with boson-tagged jets in pp collisions at √s = 13 TeV with the ATLAS detector

ATLAS Collaboration AKILLI, Ece (Collab.), et al.


Narrow resonances decaying into WW , WZ or ZZ boson pairs are searched for in 36.7 fb −1 of proton–proton collision data at a centre-of-mass energy of s=13 TeV recorded with the ATLAS detector at the Large Hadron Collider in 2015 and 2016. The diboson system is reconstructed using pairs of large-radius jets with high transverse momentum and tagged as compatible with the hadronic decay of high-momentum W or Z bosons, using jet mass and substructure properties. The search is sensitive to diboson resonances with masses in the range 1.2–5.0 TeV. No significant excess is observed in any signal region. Exclusion limits are set at the 95% confidence level on the production cross section times branching ratio to dibosons for a range of theories beyond the Standard Model. Model-dependent lower limits on the mass of new gauge bosons are set, with the highest limit set at 3.5 TeV in the context of mass-degenerate resonances that couple predominantly to bosons.

ATLAS Collaboration, AKILLI, Ece (Collab.), et al . Search for diboson resonances with

boson-tagged jets in pp collisions at √s = 13 TeV with the ATLAS detector. Physics Letters. B , 2018, vol. 777, p. 91-113

DOI : 10.1016/j.physletb.2017.12.011

Available at:

Disclaimer: layout of this document may differ from the published version.

1 / 1


Contents lists available atScienceDirect

Physics Letters B

Search for diboson resonances with boson-tagged jets in pp collisions at √

s = 13 TeV with the ATLAS detector


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



Receivedinrevisedform4December2017 Accepted5December2017

Availableonline7December2017 Editor:M.Doser

NarrowresonancesdecayingintoW W,W Z orZ Zbosonpairsaresearchedforin36.7 fb1ofproton–

proton collision data at a centre-of-mass energy of

s=13 TeV recorded with the ATLAS detector at the LargeHadron Collider in 2015and 2016. The diboson systemis reconstructed using pairs of large-radiusjetswith high transverse momentumand taggedas compatiblewith the hadronicdecay ofhigh-momentumW orZ bosons,usingjetmassandsubstructureproperties.Thesearchissensitive todiboson resonanceswith massesinthe range1.2–5.0 TeV.Nosignificantexcessisobservedinany signalregion.Exclusionlimitsaresetatthe95%confidencelevelontheproductioncrosssectiontimes branchingratiotodibosonsforarangeoftheoriesbeyondtheStandardModel.Model-dependentlower limitsonthemassofnewgaugebosonsareset,withthehighestlimitsetat3.5 TeVinthecontextof mass-degenerateresonancesthatcouplepredominantlytobosons.

©2017TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (

1. Introduction

A major goal of the physics programme at the Large Hadron Collider(LHC)isthesearchfornewphenomenathatmaybecome visibleinhigh-energyproton–proton(pp)collisions. Onepossible signature of such new phenomena is the production of a heavy resonance with the subsequent decay into a final state consist- ing ofa pairofvector bosons (W W, W Z, Z Z). Many modelsof physics beyond the Standard Model (SM) predict such a signa- ture. These include extensions to the SM scalar sector as in the two-Higgs-doubletmodel(2HDM)[1]thatpredictnewspin-0res- onances,composite-Higgsmodels [2–4] andmodels motivatedby Grand Unified Theories [5–7] that predict new W spin-1 reso- nances,andwarpedextradimensionsRandall–Sundrum(RS)mod- els[8–10]thatpredictspin-2Kaluza–Klein(KK)excitationsofthe graviton,GKK.Theheavyvectortriplet (HVT)[11,12]phenomeno- logical Lagrangian approach provides a more model-independent frameworkforinterpretationofspin-1dibosonresonances.

ThesearchpresentedherefocusesonTeV-scaleresonancesthat decayintopairsofhigh-momentumvectorbosonswhich,inturn, decay hadronically. The decay products of each of those vector bosonsarecollimated dueto thehighLorentz boostandaretyp- icallycontainedinasinglejetwithradius R=1.0.Whiletheuse ofhadronicdecaysofthevector bosons benefitsfromthelargest branching ratio(67% for W and70% for Z bosons) amongst the possiblefinal states,it suffers froma large background contami-

nation fromtheproductionofmultijetevents.However,thiscon- taminationcanbemitigatedwithjetsubstructuretechniquesthat exploitthetwo-bodynatureofVqqdecays(withV =W or Z).

Previous searches fordibosonresonances were carriedout by the ATLAS and CMS collaborations with pp collisions at

s=7, 8 and13 TeV. Theseinclude fullyleptonic (νν, ν) [13–16], semileptonic(ννqq,νqq,qq)[17–19]andfullyhadronic(qqqq) V V [17,19]finalstates.Bycombiningtheresultsofsearchesinthe

ννqq,νqq,qqandqqqqchannels, theATLASCollaboration[17]

setalowerboundof2.60 TeVonthemassofaspin-1resonanceat the95%confidencelevel,inthecontextoftheHVTmodelBwith gV=3 (describedin Section2). Wheninterpreted inthecontext ofthebulkRSmodelwithaspin-2KKgravitonandk/MPl=1,this lower massbound is1.10 TeV.The resultspresentedherebenefit froman integratedluminosity of 36.7 fb1,which isan order of magnitudelargerthanwasavailablefortheprevioussearchinthe fullyhadronicfinalstateat

s=13 TeV[17].

2. Signalmodels

Theanalysisresultsareinterpretedintermsofdifferentmodels thatpredicttheproductionofheavyresonanceswitheitherspin0, spin 1orspin 2. Inthecaseofthespin-0 interpretation,aheavy scalaris producedvia gluon–gluon fusionwithsubsequent decay intoapairofvectorbosons.Forthisempiricalmodel,thewidthof thesignalinthedibosonmassdistributionisassumedtobedom- inated by the experimental resolution. The width of a Gaussian distributioncharacterising themassresolutionafterfulleventse- lectionrangesfromapproximately3%to2%astheresonancemass

0370-2693/©2017TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense( SCOAP3.


increasesfrom1.2to 5.0 TeV. The spin-0model isreferred to as theheavyscalarmodelintherestofthisLetter.

Inthe HVTphenomenologicalLagrangian model,a newheavy vector triplet (W, Z) is introduced,withthe newgauge bosons degenerate in mass (also denoted by V in the following). The couplings between those bosons and SM particles are described inageneralmanner, therebyallowing abroadclass ofmodelsto be encompassedby thisapproach. Thenew tripletfield interacts with the Higgs field and thus with the longitudinally polarised W and Z bosons by virtue of the equivalence theorem [20–22].

ThestrengthofthecouplingtotheHiggsfield,andthusSMgauge bosons,is controlled bythe parametercombination gVcH, where cH isamultiplicativeconstantusedtoparameterisepotentialdevi- ationsfromthetypicalstrengthoftripletinteractionstoSMvector bosons,takentobe gV.CouplingofthetripletfieldtoSMfermions issetbytheexpressiong2cF/gV,where gistheSMSU(2)Lgauge couplingand,likeforthecouplingtotheHiggsfield,cF isamul- tiplicative factorthat modifiesthe typical couplingof the triplet field to fermions. The HVT model A with gV =1, cH g2/g2V andcF 1 [11] is usedasa benchmark. Inthis model,the new tripletfieldcouplesweaklytoSMparticlesandarisesfromanex- tensionoftheSMgaugegroup.BranchingratiosforWW Z and ZW W are approximately2.0%each.The intrinsicwidthof thenewbosons isapproximately2.5% ofthemass,whichresults inobservable mass peakswitha widthdominated bythe exper- imental resolution.Inthismodel,the dominantdecaymodesare intofermionpairsandsearchesinthe andν finalstates[23, 24] provide the best sensitivity. The calculated production cross section timesbranchingratio(σ×B)valuesfor WW Z with W and Z bosons decaying hadronically are 8.3 and 0.75 fb for W massesof2and3 TeV, respectively.Correspondingvaluesfor ZW W are3.8and0.34 fb.

The HVTmodel B with gV=3 and cH cF 1 [11] isused as another benchmark. This model describes scenarios in which strongdynamicsgiverisetotheSMHiggsbosonandnaturallyin- cludea newheavy vectortriplet field withelectroweakquantum numbers. The constants cH andcF are approximately unity, and couplingstofermionsaresuppressed,givingrisetolargerbranch- ingratios(50%)foreitherWW Z orZW W decaysthan in model A. Resonance widths and experimental signatures are similar to those obtained for model A and the predicted σ×B

valuesforWW Z withhadronicW and Z decaysare 13and 1.3 fbfor W massesof2and3 TeV, respectively. Corresponding valuesfor ZW W are6.0and0.55 fb.

The RS modelwith one warped extra dimension predicts the existence ofspin-2 Kaluza–Klein excitationsofthe graviton, with thelowestmodebeingconsidered inthissearch.Whiletheorigi- nalRSmodel[8](oftenreferredtoasRS1)isconstructedwithall SMfieldsconfinedto afour-dimensionalbrane (the“TeV brane”), thebulkRSmodel[8,9]employedhereallowsthosefieldstoprop- agate in the extra-dimensional bulk between the TeVbrane and thePlanckbrane.Althoughruledoutbyprecisionelectroweakand flavour measurements, the RS1 model is used as a benchmark modeltointerpretdiphoton anddileptonresonancesearches due tothe sizeable GKK couplings tolight fermions inthat model.In the bulk RS model, those couplings are suppressed and decays intofinal states involvingheavy fermions,gauge bosons orHiggs bosons are favoured. The strength of the coupling depends on k/MPl,wherekcorresponds tothecurvatureofthe warpedextra dimension,andthe effectivefour-dimensionalPlanckscale MPl= 2.4×1018GeV.Thecrosssectionandintrinsic width scaleasthe squareofk/MPl.Forthechoicek/MPl=1 usedinthissearch,the σ×B valuesfor GKKW W withW decayinghadronically are 0.54 and 0.026 fb for GKK masses of 2 and 3 TeV, respectively.

CorrespondingvaluesforGKKZ Z are0.32and0.015 fb.Inthe

rangeofGKKmassesconsidered,thebranchingratiotoW W (Z Z) variesfrom24%to20%(12%to10%)asthemassincreases.Decays intothet¯tfinalstatedominatewithabranchingratiovaryingfrom 54%to60%.TheGKKresonancehasavaluethatisapproximately 6%ofitsmass.

3. ATLASdetector

The ATLAS experiment [25,26] at the LHC is a multi-purpose particle detector with a forward–backward symmetric cylindrical geometry and a near 4π coverage in solid angle.1 It consists of aninnerdetectorfortrackingsurroundedbyathinsuperconduct- ing solenoidprovidinga2 Taxial magneticfield,electromagnetic and hadronic calorimeters, and a muon spectrometer. The inner detector covers the pseudorapidity range |η|<2.5.It consistsof silicon pixel, silicon microstrip, and transition radiation tracking detectors.Anewinnermostpixellayer[26]insertedataradiusof 3.3 cmhasbeenusedsince2015.Lead/liquid-argon(LAr)sampling calorimeters provide electromagnetic (EM) energy measurements withhighgranularity.Ahadronic(steel/scintillator-tile)calorimeter covers the central pseudorapidity range (|η|<1.7). The end-cap and forward regions are instrumented with LAr calorimeters for both theEM andhadronicenergymeasurements upto |η|=4.9.

The muon spectrometer surrounds the calorimeters and features threelargeair-coretoroidalsuperconductingmagnetsystemswith eight coils each. The field integral ofthe toroidsranges between 2.0and6.0 Tmacrossmostofthedetector.The muonspectrom- eter includes a system of precision tracking chambers and fast detectors for triggering. A two-level trigger system [27] is used to select events. The first-level trigger is implemented in hard- ware anduses asubsetofthedetectorinformationtoreduce the acceptedratetoatmost100 kHz.Thisisfollowedbyasoftware- based triggerlevelthat reducestheacceptedeventrateto1 kHz onaverage.

4. Dataandsimulation 4.1. Data

ThedataforthisanalysiswerecollectedduringtheLHC ppcol- lisionrunningat

s=13 TeV in2015and2016.Eventsmustpass a trigger-level requirementofhaving atleastone large-radius jet withtransverse energy ET>360 GeV in 2015and ET>420 GeV in 2016, where the jet is reconstructed using the anti-kt algo- rithm [28] witha radius parameter of 1.0. Those thresholds cor- respondtothelowest-ET,unprescaledlarge-radiusjettriggersfor each ofthetwodata-taking periods.Afterrequiringthat thedata were collected during stable beam conditions and the detector components relevant to this analysis were functional, the inte- gratedluminosityofthesampleamountsto3.2 fb1and33.5 fb1 of ppcollisionsin2015and2016,respectively.

4.2. Simulation

The search presented here uses simulated Monte Carlo (MC) event samples to optimise the selection criteria, to estimate the

1 ATLASusesaright-handedcoordinatesystemwithitsoriginatthenominalin- teractionpoint(IP)inthecentreofthedetectorandthez-axisalongthebeampipe.

Thex-axispointsfromtheIPtothecentreoftheLHCring,andthe y-axispoints upwards.Cylindricalcoordinates(r,φ)areusedinthetransverseplane,φbeingthe azimuthalanglearoundthez-axis.Thepseudorapidityisdefinedintermsofthepo- larangleθasη= −ln tan(θ/2).Therapidityisdefinedrelativetothebeamaxisas y=12lnEE+ppz




acceptancefordifferentsignalprocesses,andtovalidatetheexper- imental proceduredescribed below. However, itdoes notrely on MC eventsamples toestimate thebackground contributionfrom SMprocesses.

Signaleventsfortheheavyscalarmodel[29]wereproducedat next-to-leading-orderviathegluon–gluonfusionmechanismwith Powheg-Boxv1 [30,31] using theCT10 partondistribution func- tion(PDF)set[32].EventswereinterfacedwithPythiav8.186[33]

for parton showering and hadronisation using the CTEQ6L1 PDF set[34]andtheAZNLO setoftuned parameters(laterreferredto astune)[35].Thewidthoftheheavyscalarisnegligiblecompared totheexperimentalresolution.

Inthe caseofthe HVTandRSmodels, eventswere produced atleading order(LO) withtheMadGraph5_aMC@NLO v2.2.2 [36]

event generator using the NNPDF23LO PDF set [37]. To study thesensitivityofthespin-2 resonancesearch toproductionfrom quark–antiquarkorgluon–gluon initial statesaswell asto differ- ent vector-bosonpolarisation states, events were generated with JHUGen v5.6.3 [38] and the NNPDF23LO PDF set. For these sig- nalmodels,theeventgeneratorwasinterfacedwithPythiav8.186 forparton showeringand hadronisation with the A14 tune [39].

The GKK samples are normalised according to calculations from Ref. [40]. In all signal samples, the W and Z bosons are longi- tudinallypolarised.

MultijetbackgroundeventsweregeneratedwithPythiav8.186 withtheNNPDF23LO PDFsetandtheA14tune. Samplesof W+ jets andZ+jets events weregeneratedwithHerwig++ v2.7.1[41]


Forall MC samples, charm-hadron andbottom-hadron decays werehandledbyEvtGenv1.2.0[43].Minimum-biaseventsgener- atedusingPythia 8were addedtothehard-scatterinteractionin such a wayas to reproduce the effects ofadditional pp interac- tions in each bunch crossing during data collection (pile-up). An averageof23pile-upinteractions areobservedinthedatainad- ditionto the hard-scatter interaction. The detectorresponse was simulated with Geant 4 [44,45] and the events were processed withthesamereconstructionsoftwareasforthedata.

5. Eventreconstructionandselection 5.1.Reconstruction

Theselection of eventsrelieson the identificationandrecon- structionofelectrons,muons,jets,andmissingtransversemomen- tum.Althoughtheanalysis primarilyrelies onjets, other particle candidatesare neededtoreject eventsthat are includedincom- plementarysearchesfordibosonresonances.

The trajectories of charged particles are reconstructed using measurements in the inner detector. Of the multiple pp colli- sion vertices reconstructed from the available tracks in a given event, a primary vertex is selected as the one with the largest p2T,wherethesumisoveralltrackswithtransversemomentum pT>0.4 GeV thatare associatedwiththe vertex. Tracksthat are consistentwith theprimary vertex maybe identified aselectron or muon candidates. Electron identification is based on match- ing tracks to energy clusters in the electromagnetic calorimeter andrelyingonthelongitudinalandtransverseshapesoftheelec- tromagnetic shower. Electron candidates are required to satisfy the“medium”identificationcriterion [46]andtopassthe“loose”

track-basedisolation [46]. Muonidentification reliesonmatching tracksintheinnerdetectortomuonspectrometer tracksortrack segments.Muoncandidatesmustalsosatisfythe“medium”selec- tioncriterion[47]andthe“loose”trackisolation[47].

Large-radius jets (hereafter denoted large-R jets) are recon- structed from locally calibrated clusters of energy deposits in

calorimetercells [48] with theanti-kt clustering algorithm using a radius parameter R=1.0. Jets are trimmed [49] to minimise the impactofpile-up byreclusteringthe constituentsof eachjet with the kt algorithm [50] into smaller R=0.2 subjets and re- moving those subjets with psubjetT /pjetT <0.05, where psubjetT and pjetT arethetransversemomentaofthesubjetandoriginaljet,re- spectively.Theclustering andtrimming algorithmsusetheFastJet package [51].Calibration ofthe trimmed jet pT and massis de- scribedinRef.[52].

Thelarge-Rjetmassiscomputedusingmeasurementsfromthe calorimeterandtrackingsystems[53]accordingto

mJ=wcalmcal+wtrk pT ptrkT mtrk,

where ptrkT isthe transverse momentum ofthe jet evaluated us- ing only charged-particle tracksassociated withthe jet,mcal and mtrk arethemassescomputedusingcalorimeterandtrackermea- surements, and wcal and wtrk are weights inverselyproportional tothesquareoftheresolutionofeach ofthecorrespondingmass terms. Ghost association [54] is performed to associate tracks to thejetsbeforethetrimmingprocedureisapplied.Inthismethod, tracksareaddedwithaninfinitesimallysmallmomentumasaddi- tionalconstituentsinthejetreconstruction.Tracksassociatedwith thejetsarerequiredtohavepT>0.4 GeV andsatisfyanumberof qualitycriteriabasedonthenumberofmeasurementsinthesili- conpixelandmicrostripdetectors;tracksmustalsobeconsistent withoriginatingfromtheprimary vertex[53].Includinginforma- tionfromthetrackingsystemprovidesimprovedmassresolution, especiallyathighjet pT,duetotherelativelycoarseangularreso- lutionofthecalorimeter.

The magnitude of the event’s missing transverse momentum (EmissT ) is computed from the vectorial sum of calibrated elec- trons, muons, and jets in the event [55]. For this computation andtherejectionofnon-collisionbackgrounddiscussedbelow,jets arereconstructed fromtopologicalclustersusingtheanti-kt algo- rithmwitharadiusparameterR=0.4 and arerequiredtosatisfy pT>20 GeV and |η|<4.9. Calibration of those jetsis described in Ref. [56]. The ETmiss value is corrected usingtracks associated withtheprimary vertexbutnot associatedwithelectrons,muons orjets.

5.2. Selection

Eventsusedincomplementarysearchesfordibosonresonances in different final states are removed, in anticipation of a future combination. Accordingly, events are rejectedif they contain any electron or muon with pT>25 GeV and |η|<2.5. Furthermore, eventswithEmissT >250 GeV arerejected.

Events with jets that are likely to be due to non-collision sources, including calorimeternoise, beamhalo andcosmic rays, are removed [57]. Events are required to contain at least two large-R jetswith|η|<2.0 (toguaranteeagood overlapwiththe tracking acceptance) and mass mJ>50 GeV. The leading (high- est pT) large-R jet must have pT>450 GeV and the subleading (secondhighest pT) large-R jetmusthave pT>200 GeV.The in- variant mass ofthe dijet system formed by these two jetsmust bemJJ>1.1 TeV toavoidinefficienciesduetotheminimumjet-pT requirementsandtoguaranteethatthetriggerrequirementisfully efficient.Onlyjetsinthissystemareconsideredintherestofthis Letter.Eventspassingtheaboverequirementsaresaidtopassthe event“preselection”.

Furtherkinematicrequirementsareimposedtosuppressback- groundfrommultijetproduction.Therapidityseparationbetween theleadingandsubleadingjets(identifiedwithsubscripts1and2



Sujets connexes :