Search for vector-boson resonances decaying to a top quark and bottom quark in the lepton plus jets final state in pp collisions at root s=13 TeV with the ATLAS detector

(1)

Contents lists available atScienceDirect

Physics

Letters

B

www.elsevier.com/locate/physletb

Search

for

vector-boson

resonances

decaying

to

a

top

quark

and

bottom

quark

in

the

lepton

plus

jets

ﬁnal

state

in

pp collisions

at

√

s

=

13 TeV with

the

ATLAS

detector

.TheATLASCollaboration

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

Articlehistory: Received27July2018

Receivedinrevisedform15October2018 Accepted3November2018

Availableonline22November2018 Editor: W.-D.Schlatter

A search for new charged massive gauge bosons, W, is performed with the ATLAS detector at the LHC. Datawere collectedinproton–proton collisions atacenter-of-massenergy of√s=13 TeV and correspond to an integrated luminosity of 36.1 fb−1_. _This _analysis _searches _for _W _bosons _in _the W→tb decay¯ channelinfinalstateswithanelectronormuonplusjets.Thesearchcoversresonance massesbetween0.5and 5.0 TeV andconsidersright-handedW bosons.Nosignificantdeviationfrom theStandardModel(SM)expectationisobservedandupperlimitsaresetontheW_→tb cross¯ section timesbranchingratio andthe W bosoneffective couplingsasafunctionofthe W bosonmass.For right-handed W bosonswith couplingto the SMparticlesequal tothe SMweak coupling constant, massesbelow3.15 TeV are excludedatthe95%confidencelevel.Thissearchisalsocombinedwitha previously publishedATLAS result for W→tb in¯ the fully hadronicfinal state. Usingthe combined searches,right-handedW bosonswithmassesbelow3.25 TeV areexcludedatthe95%confidencelevel. ©2018TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP3_.

1. Introduction

ManyapproachestotheoriesbeyondtheStandardModel(SM) introducenew chargedvector currents mediated by heavy gauge bosons, usually referred to as W. For example, the W boson can appear in theories with universal extra dimensions, such as Kaluza–Klein excitations of the SM W boson [1–3], or in mod-els that extend fundamental symmetries ofthe SM and propose amassiveright-handedcounterparttothe W boson [4–6]. Little-Higgs [7] and composite-Higgs [8,9] theories also predict a W boson.ThesearchforaWbosondecayingintoatopquarkanda b-quark (illustratedinFig.1) exploresmodelspotentially inacces-sibletosearchesforaWbosondecayingintoleptons [10–15].

Forinstance,intheright-handedsector, the W boson cannot decayintoachargedleptonandahypotheticalright-handed neu-trino if the latter has a mass greater than the W boson mass (mixingbetweenW andSM W bosonsisusually constrainedto besmallfromexperimentaldata [16]).Also,inseveraltheories be-yondtheSMtheWbosonisexpectedtocouplemorestronglyto thethird generationofquarksthantotheﬁrstandsecond gener-ations [17,18].Searchesfora W bosondecayingintothetb ﬁnal¯

_E-mail_address:_atlas_{.publications}_@cern_.ch_.

state1 _have_been_performed_at_the_{Tevatron [}₁₉_,₂₀_{] in}_the_leptonic

top-quark decaychannel and atthe Large Hadron Collider (LHC) inboththeleptonic [21–25] andfullyhadronic [26,27] finalstates, andthemostrecentresultsexcluderight-handedWbosonswith massesupto about3.6 TeV atthe 95%confidencelevel.A previ-ousATLASsearchintheleptonicchannel [24] usingproton–proton (pp) collisions ata center-of-mass energyof √s=8 TeV yielded a lower limit of 1.92 TeV on the mass of W boson with right-handed couplings.More recently,the CMSCollaboration reported results usinga 13 TeV pp dataset of 35.9 fb−1 [25], yielding a lowerlimit of3.6 TeV onthemassofright-handed W bosons.A search by the ATLAS Collaboration inthe fullyhadronicdecay of the tb final¯ state using 36.1 fb−1 of 13 TeV data yieldedlower limitsonthemassofright-handedW bosonsat3.0 TeV [27]. In each oftheseanalyses,the couplingstrengthofthe W boson to right-handed particleswas assumed tobe equal tothe SM weak couplingconstant.

ThisLetterpresentsasearchforWbosonsusingdatacollected during the period 2015–2016 by the ATLAS detector [28] at the LHC,correspondingtoanintegratedluminosityof36.1 fb−1 from pp collisionsat acenter-of-mass energyof13 TeV.The search is performedintheW_R →tb¯→ νbb decay¯ channel,wherethe

lep-1 _The_notation_“t_b”¯ _is_used_to_describe_both_the_W +_→_t_{b and}_¯ _W −_{→ ¯}_tb

pro-cesses.

https://doi.org/10.1016/j.physletb.2018.11.032

(2)

Table 1

Eventgeneratorsusedforthesimulationofthesignalandbackgroundprocesses.ThePS/Hadcolumn de-scribestheprogramusedforpartonshowerandhadronization.

Process Generator PS/Had MC Tune PDF

WR MadGraph5_aMC@NLO Pythia8 A14 NNPDF23LO t¯t Powheg-Box Pythia6 Perugia 2012 NLO CT10 Single-top t-channel Powheg-Box Pythia6 Perugia 2012 NLO CT10 Single-top W+t Powheg-Box Pythia6 Perugia 2012 NLO CT10 Single-top s-channel Powheg-Box Pythia6 Perugia 2012 NLO CT10

W,Z + jets Sherpa2.2.1 Sherpa2.2.1 Default NLO CT10

W W , W Z , Z Z Powheg-Box Pythia8 AZNLO LO CTEQ6L1

Fig. 1. Feynmandiagramfor Wbosonproductionfromquark–antiquark annihila-tionwiththesubsequentdecayintotb and¯ aleptonicallydecayingtopquark. ton,,iseitheran electronora muon.Right-handed W bosons, denotedW_R ,aresearchedforinthemassrangeof0.5to5.0 TeV. AgeneralLorentz-invariantLagrangianisusedtodescribethe cou-plingsoftheW_R bosontofermionsasafunction ofitsmass [29, 30]. The mass of the right-handed neutrino is supposed to be largerthanthemassoftheW_R boson,thusnon-hadronicdecaysof theW_R bosonhaveanegligiblebranchingfraction.Inthisweakly coupledmodel,theresultingbranching fractionofthe W_R tothe tb ﬁnal¯ stateincreasesasafunctionofmassfrom29.9%at0.5 TeV to33.3%at5 TeV.

2. ATLASdetector

The ATLAS detectorat the LHC covers almost the entiresolid anglearoundthecollisionpoint.2 Chargedparticlesinthe pseudo-rapidityrange|η|<2.5 arereconstructed withtheinner detector (ID), which consists of several layers of semiconductor detectors (pixelandstrip)andastraw-tubetransition–radiation tracker,the latter extending to |η|=2.0. The high-granularity silicon pixel detectorprovides four measurements per track;the closest layer to the interaction point is known as the insertable B-layer [31, 32], which was added in 2014and provides high-resolution hits at small radius to improve the tracking performance. The ID is immersed in a 2 T magnetic ﬁeld provided by a superconduct-ing solenoid. The solenoidis surrounded by electromagnetic and hadronic calorimeters, and a muon spectrometer incorporating three large superconducting air–core toroid magnet systems.The calorimetersystemcoversthepseudorapidityrange|η|<4.9. Elec-tromagnetic calorimetry is provided by barrel and endcap high-granularity lead/liquid-argon (LAr) electromagnetic calorimeters, within the region |η|<3.2. There is an additional thinLAr pre-sampler covering |η|<1.8 to correct forenergy loss inmaterial upstream of the calorimeters.For |η|<2.5, the LAr calorimeters

2 _ATLAS_uses_a_right-handed _coordinate_system_with _its_origin_at _the_nominal

interactionpointinthecenter ofthedetectorandthez-axisalongthebeampipe. Thex-axispointsfromtheinteractionpointtothecenter oftheLHCring,andthe y-axispointsupward.Cylindricalcoordinates(r,φ)areusedinthetransverseplane, φbeingtheazimuthalanglearoundthebeampipe.Thepseudorapidityisdeﬁned intermsofthepolarangleθasη= −ln tan(θ/2).Observableslabeled “transverse” areprojectedintothex– y planeandangulardistanceismeasuredinunitsofR=

(η)2_{+ (φ)}2_.

aredividedintothreelayersindepth.Hadroniccalorimetryis pro-videdby a steel/scintillator-tilecalorimeter,segmentedinto three barrel structures within |η|<1.7, and two copper/LAr hadronic endcap calorimeters,which coverthe region 1.5<|η|<3.2. The forwardsolid angleout to|η|=4.9 iscoveredbycopper/LArand tungsten/LAr calorimetermodules, which are optimized for elec-tromagnetic and hadronicmeasurements, respectively. The muon spectrometer comprisesseparate triggerandhigh-precision track-ingchambersthatmeasurethedeflectionofmuonsinamagnetic fieldgeneratedbythethreetoroidmagnetsystems.TheATLAS de-tector selects eventsusing a tiered trigger system [33]. The first levelisimplementedincustomelectronicsandreducestheevent ratefromtheLHCcrossingfrequencyof40MHztoadesignvalue of100kHz.The secondlevelisimplementedinsoftwarerunning onacommodityPCfarmwhichprocessestheeventsandreduces therateofrecordedeventsto1.0kHz.

3. Dataandsimulatedsamples

This analysis uses 36.1±0.8 fb−1 of pp collisions data at √

s=13 TeV recordedusingsingle-electronandsingle-muon trig-gers. Additional data-quality requirements are also imposed, and thesearedetailedinSection 4.During 2015thiscorrespondedto 3.2 fb−1 _with_an_average_of_13.4_interactions _per_bunch_crossing. The2016data-takingperiodcorrespondsto32.9 fb−1_with_an av-erageof25.1interactionsperbunchcrossing.

The W_R boson search isperformed inthe semileptonicdecay channel,wheretheW_R decaysintoatopquarkandab-quark,the top quark decaysintoa W boson andab-quark, andthe W bo-son decays in turn into a lepton and a neutrino. The ﬁnal-state signature thereforeconsistsoftwob-quarks,one chargedlepton3

andaneutrino, whichisundetectedandresultsinmissing trans-verse momentum, E_Tmiss.The dominant backgroundprocesses for thissignaturearethereforetheproductionofW/Z+jets(jets aris-ing fromlight and heavy partons),electroweak single top quarks (t-channel, W t ands-channel),t¯t pairsanddibosons(W W , W Z , and Z Z ).Aninstrumental backgroundduetomultijetproduction, where ahadronicjet ismisidentiﬁed asalepton, isalsopresent. MonteCarlo(MC)simulatedeventsareusedtomodeltheW_R sig-nal andall the SM background processes, withthe exception of the multijetbackground prediction, which is derived using data. TheMCgeneratorprogramsandconﬁgurationsaresummarizedin Table1,anddescribedingreaterdetailinthetextbelow.

Simulated signal events were generatedat leading order(LO) by MadGraph5_aMC@NLO v2.2.3 [34–37] usingachiralW_R boson model in which the couplings to the right-handed fermions are likethoseintheSM. MadGraph5_aMC@NLO isalsousedtomodel thedecaysofthetop quark,takingspincorrelationsintoaccount.

3 _The _analysis _selects _electrons _or _muons, _while _the _simulation _includes τ-leptons.Thustheeventyieldincludesasmallcontributionduetoleptonic de-caysofτ-leptons.

(3)

Pythia8v8.186 [38] wasusedforpartonshoweringand hadroniza-tion, wherein the NNPDF23LO [39] parton distribution functions (PDF)oftheprotonandasetoftunedparameters calledtheA14 Pythiatune [40] were used.Allsamplesofsimulatedeventswere rescaledtonext-to-leading-order(NLO) calculationsusingNLO/LO K -factors rangingfrom 1.1to1.4, decreasing asa function ofthe massofthe W_R boson,calculatedwith ZTOP [30].Signalsamples weregenerated between0.5and3 TeV insteps of250 GeV, and between3and5 TeV instepsof500 GeV.

Thebenchmarksignal modelusedforthiswork nominally as-sumesthattheW_R bosoncouplingstrengthtofermions, g,isthe sameasforthe W boson: g=g,whereg istheSMSU(2)L cou-pling.The couplingof left chiralfermions tothe W_R is assumed tobe zero.Thetotal widthofthe W_R boson increasesfrom2to 130 GeV formasses between0.5 and5 TeV [29] for g=g and scalesasthe squareof theratio g/g. Inorderto explorethe al-lowedrangeofthe W_R coupling g, sampleswerealsogenerated forvaluesof g/g upto5.0, forseveral W_R bosonmass hypothe-ses,allowingtheeffectofincreasedW_R widthtoalsobeincluded. Simulated top-quark pair and single-top-quark processes (t-channel,s-channelandW t)wereproducedusingtheNLO Powheg-Box_[₄₁_,₄₂_{] generator}_with_the_CT10_{PDF [}₄₃_]._The_parton_shower and the underlying event were added using Pythia v6.42 [44] with the Perugia 2012 tune [45]. The top-quark pair produc-tion MC sample is normalized to an inclusive cross section of σt¯t=832+₋4651 pb for a top-quarkmass of 172.5 GeV as obtained from next-to-NLO (NNLO) plus next-to-next-to-leading-logarithm (NNLL)QCDcalculationswiththeTop++2.0program [46–52].

The background contributions from W and Z boson produc-tion in association with jets were simulated using the Sherpa v2.2.1 [53] generator. Matrix elements were calculated for up to twopartonsatNLO andfourpartons atLOandmergedwiththe SherpapartonshowerusingtheME+PS@NLOprescription [54–56]. TheW/Z +jetssamplesarenormalizedtotheinclusiveNNLOcross sectionscalculatedwithFEWZ [57,58].

Theproductionofvector-bosonpairs (W W , W Z or Z Z ) with at least one charged lepton in the ﬁnal state was simulated by the Powheg-Box generatorin combinationwith Pythia8 and the leading-order CTEQ6L1PDF [59].Thenon-perturbativeeffectswere modeled withthe AZNLO setoftunedparameters [60].

Forall MadGraph and Powheg samples,the EvtGen v1.2.0 pro-gram [61] wasusedforthebottomandcharmhadrondecays.

Allsimulated eventsamplesinclude theeffect ofmultiple pp interactionsinthesameandneighboring bunchcrossings(pile-up) byoverlaying,oneachsimulatedsignalorbackgroundevent, sim-ulatedminimum-biaseventsgenerated using Pythia8,the A2 set oftunedparameters [62] andthe MSTW2008LO PDFset [63].

Simulatedsamples were processed through the Geant4-based ATLAS detectorsimulationorthrough a fastersimulation making useof parameterized showers in the calorimeters [64,65]. Simu-latedevents were then processed using the same reconstruction algorithmsandanalysischainasusedfordata.

4. Eventselectionandbackgroundestimation

This search makes use of the reconstruction of multi-particle vertices,theidentiﬁcationandthekinematicpropertiesof recon-structedelectrons, muons, jets, andthe determinationofmissing transversemomentum.

Collisionverticesarereconstructedfromatleasttwo IDtracks withtransversemomentum pT>400 MeV. Theprimary vertexis selectedastheonewiththehighestpT2,calculatedconsidering allassociatedtracks.

Electrons are reconstructed from ID tracks that are matched tonoise-suppressedtopologicalclustersofenergydepositions [66]

inthe electromagneticcalorimeter.Theclustersare reconstructed using the standard ATLAS sliding-window algorithm, which clus-ters calorimeter cells within ﬁxed-size rectangles [67]. Electron candidates are requiredto satisfy criteriaforthe electromagnetic showershape,trackquality,andtrack–clustermatching;these cri-teria areappliedusinga likelihood-basedapproach.Electron can-didatesmustmeetthe“Tight”workingpointrequirementsdeﬁned inRef. [68] andare furtherrequiredto have pT > 25 GeV anda pseudorapidityofthecalorimeterclusterposition,|ηcluster|,smaller than2.47. Events withelectrons falling inthecalorimeterbarrel– endcaptransitionregion,1.37<|ηcluster| <1.52,whichhaslimited instrumentation,arerejected.

Muons are identified by matching tracks found in the ID to either full tracks or track segments reconstructed in the muon spectrometer(“combinedmuons”),orbystand-alonetracksinthe muon spectrometer [69]. They are requiredto pass identification requirementsbasedonqualitycriteriaappliedtotheIDandmuon spectrometer tracks. Muon candidates must meet the “Medium” identificationworkingpointrequirementsdefinedinRef. [69],have atransversemomentum pT >25 GeV,andsatisfy|η|<2.5.

Electron andmuon candidates must further satisfy additional isolationcriteriathatimproverejectionofcandidatesarisingfrom sourcesotherthanpromptW/Z bosondecays(e.g.hadrons mim-ickinganelectronsignature,heavy-flavor hadrondecaysorphoton conversions).Muonsarerequiredtobeisolatedusingthe require-mentthatthescalarsumofthe pT ofthetracksinavariable-size conearoundthemuondirection(excludingthetrackidentifiedas the muon) be less than 6% of the transverse momentum of the muon. The trackisolation cone size is givenby the minimum of R=10 GeV/pμ_T andR=0.3.Electronsarealsorequiredtobe isolatedusingthesametrack-basedvariableasformuons, except thatthemaximumR inthiscaseis0.2.Forthepurposeof mul-tijet background estimation (see Section 5) electrons and muons satisfyingaloosersetofidentificationcriteria,inparticular with-outanisolationrequirement,arealsoconsidered.

Jetsarereconstructedfromtopologicalcalorimeterclusters us-ingtheanti-kt algorithm [70] witharadiusparameterof R=0.4, and must satisfy pT > 25 GeV and |η| < 2.5. To suppress jets originating from in-time pile-up interactions, jets in the range pT<60 GeV and |η|<2.4 are required to pass the jet vertex tagger [71] selection, which has an eﬃciency of about 90% for jetsoriginatingfromthe primaryvertex. Theclosest jets overlap-ping with selectedelectron candidates within a cone ofsize R equal to 0.2 are removed from events, as the jet and the elec-tron very likely correspond to the same reconstructed object. If a remaining jet with pT > 25 GeV isfound close toan electron withinaconeofsizeR=0.4,thentheelectroncandidateis dis-carded.SelectedmuoncandidatesnearjetsthatsatisfyR(muon, jet)<0.04+10 GeV/p_Tμare rejectedifthejet hasatleastthree tracksoriginatingfromtheprimaryvertex.Anyjetswithlessthan threetracksthatoverlapwithamuonarerejected.

Theidentificationofjetsoriginatingfromthe hadronizationof b-quarks(“b-tagging”)isbasedonpropertiesspecifictob-hadrons, such aslong lifetimeandlargemass. Suchjetsare identified us-ing the multivariate MV2c10 b-tagging algorithm [72,73], which makes useof informationaboutthejet kinematic properties,the characteristicsoftrackswithinjets,andthepresenceofdisplaced secondary vertices. The algorithm is used at the 77% efficiency working point and provides a rejection factor of 134 (6.21) for jetsoriginatingfromlight-quarksorgluons(charmquarks),as de-termined in simulatedt¯t events.Jets satisfyingthese criteriaare referredtoas“b-tagged”jets.

The presence of neutrinos can be inferred from an apparent momentumimbalanceinthetransverseplane.Themissing trans-versemomentum(Emiss

(4)

neg-ative vectorial sum of the transverse momentum of all recon-structed objects (electrons, muons, jets) aswell as speciﬁc “soft terms”consideringtracksassociatedwiththeprimaryvertexthat donotmatchtheselectedreconstructedobjects [74].

Candidateeventsarerequiredtohaveexactlyonecharged lep-ton,twotofourjetswithatleastoneofthemb-taggedanda mini-mumEmiss_T thresholdthatdependsontheleptonflavor.Fromthese objects,W bosonandtop-quarkcandidatesarereconstructedand finalrequirementsontheeventkinematicpropertiesareappliedto defineseveralorthogonalregions withenriched signalcontent,as wellassignal-depletedregions tovalidatedatamodeling.Thejet, b-tag andlepton requirements define basic selections, which are labeled as X -jet Y -tag where X=2,3,4 and Y=1,2, separated forelectronandmuonchannelselections.

The W boson candidateis reconstructed fromthe lepton and Emiss

T ,withtheassumptionthatonlyoneneutrinoispresentinthe event.Thez componentoftheneutrinomomentum(pz)is calcu-latedfromtheinvariantmassofthelepton–Emiss_T systemwiththe constraintthatmW =80.4 GeV.The constraintyields aquadratic equationandinthecaseoftworealsolutions,thesmallest|pz| so-lutionischosen.Ifthetransversemass,mW_T ,ofthereconstructed W boson islargerthanthe valuemW used intheconstraint,the two solutions are imaginary.This casecan be dueto the resolu-tionofthemissingtransversemomentummeasurement.Here,the Emiss_x_,_y componentsareadjustedtosatisfymW_T =mW,yieldinga sin-glerealsolution.

The four-momentum of the top-quark candidate is recon-structed by adding the four-momenta ofthe W -boson candidate andofthejet,amongallselectedjetsintheevent,thatyieldsthe invariant massclosestto thetop-quarkmass(mtop=172.5 GeV). Thereafter, this jet is referred to as “btop”, and may not be the jet actually b-tagged. Finally, the four-momentum of the candi-dateW bosonisreconstructedbyaddingthefour-momentumof the reconstructed top-quark candidate and the four-momentum of the highest-pT remaining jet (referred to as “b1”). The W four-momentumisused toevaluatetheinvariant massofthe re-constructed W→tb system (mtb),whichisthevariableusedfor backgrounddiscriminationforthissearch.

Aneventselectioncommontoallsignalandvalidationregions is deﬁnedas: lepton pT>50 GeV, pT(b1)>200 GeV, pT(top)> 200 GeV,andE_Tmiss>30 GeV.Inordertokeepthemultijet back-ground ata low level an additional selection is imposed, in the muon channel, on the sum of mW_T and Emiss_T : m_TW + E_Tmiss> 100 GeV.Intheelectronchannelthesamerequirementisapplied tokeeptheselectioninbothchannelsassimilaraspossible,and, inadditiontheEmiss

T thresholdisraisedto80 GeV tofurther sup-pressthemultijetbackground.Thisphasespaceisthensubdivided into a signal region (SR), a validation region enriched with the W +jets background (VRpretag), a validation region enriched with the t¯t background (VR_t_¯_t), anda validation region enriched with the W +heavy-flavor jetsbackground(VRHF). Allregionsconsist of events withtwo or three jets, except for the VR_t_¯_t where events withexactly four jetsare selected. The SR andVR_t_¯_t requirethat one or two jets are b-tagged,while only one b-tagged jet is re-quired in the VRHF. No b-tagging requirement is applied in the VRpretag.Specific selectionsarethen appliedinthe twofollowing cases.The SRis definedby requiringthat the angularseparation ofthe lepton andbtop be small:R(,btop)<1.0. An additional criterion mtb>500 GeV is applied toremove a smallnumber of low-massW +jetsandtt events.¯ TheVRHFconsistsofeventswhere thelepton–jet andjet–jetseparations arelarge: R(,btop)>2.0 andR(b1,btop)>1.5.Theapplicationofthesetwoselections re-ducesthet¯t backgroundintheVRHF regionby90%.Theexpected signalcontaminationinthevalidationregionsisatmost5%forlow

Table 2

Summaryoftheeventselectioncriteriausedtodeﬁnesignalandvalidationregions. TheEmiss

T selectioncutisharderforeventswithelectronsthanwithmuons.

Common selection

pT() >50 GeV, pT(b1) >200 GeV, pT(top) >200 GeV EmissT >30 (80) GeV, mWT +E

miss

T >100 GeV

Signal region VRpretag VRt¯t VRHF

2 or 3 jets 2 or 3 jets 4 jets 2 or 3 jets 1 or 2 b-jets pretag 1 or 2 b-jets 1 b-jet

R(,btop) <1.0 R(,btop) >2.0

mtb>500 GeV R(b1,btop) >1.5

Fig. 2. Signal selection efficiency(efficiencyisdefined as thenumber ofevents passingallselectionsdividedbythetotalnumberofsimulatedW→tb¯→ νbb¯ events)inthesignalregionasafunctionofthesimulatedWR mass.Efficienciesare

shownfor:allchannelscombined(fullcircle),electronchannelsonly(fullsquare) andmuonchannelsonly(fulltriangle).Forreference,signaleﬃciencycurvesare alsoshownwithouttherequirementonb-tagging(pretagselection:dottedlines).

Wmasses,andfallsbelow10−4 forWmassesabove3 TeV.The eventselectioncriteriaforeachregionaresummarizedinTable2. Thesignalselectionefficiency(definedasthenumberofevents passing allselection requirementsdividedby thetotalnumberof simulated W→tb¯→ νbb events)¯ inthesignal regionisshown, as a function ofthe simulated W_R mass, in Fig. 2. Selection ef-ficiency curves are shown for the electron and muon channels separately,aswellasforthepretagselection.DuetothejetpTand R(,btop)requirements,thesignal hasvanishingefficiencyfora W_R massof500 GeV,buttheefficiencyrisesasthedecayproducts becomemoreboosted.ThemaximumSRsignalefficiency,11.3%,is obtained fora mass of1.5 TeV, then the efficiencydecreases for highermassesto5.3%at5 TeV.Theapplicationoftheb-tagging re-quirementhasalargerimpactonthesignalefficiencyathighW_R bosonmassvalues.Intheelectronchanneltheelectron–jetoverlap criteriondoesnotallowtheelectrontobeclose(R(,jet)<0.4) tothejet.Inthemuonchannel,thiscriterionisrelaxedbyusinga variableR conesize,resultinginanimprovedsignalacceptance. 5. Backgroundestimation

The t¯t, single-top-quark, diboson and W/Z +jets backgrounds are modeled usingthesimulatedMCsamplesandarenormalized tothetheorypredictionsoftheinclusivecrosssections,whilethe multijetbackgroundisestimatedusingthedataasdescribedbelow inthissection.Eachofthesebackgroundsamplesgivesriseto indi-vidualdifferentialmtb templatespredictingtheiruniquekinematic properties. These initial background normalizations are taken as

(5)

startingvalues,andthefinal normalizationisdeterminedthrough amaximum-likelihoodfitofthebackgroundtemplatestothedata inwhichthebackgroundnormalizationsare parametersofthefit (describedinSection7).Becausethesignalregionsaredominated by tt and¯ W +jets production, the normalization of these back-grounds isallowed to float freely in the maximum-likelihood fit withnoprior.

The background arising from multijet production consists of events with a jet that is misreconstructed as a lepton or with a non-prompt lepton that satisfies the lepton identification cri-teria. The simulation ofthis backgroundsource is challenging as it suffers from large systematic uncertainties and does not reli-ablyreproducetheobserveddatainregionsenrichedwithmultijet events.Thereforethe multijetbackgroundis estimatedfromdata with the so-called matrixmethod, which is used to disentangle the mixture of non-prompt leptons found in the multijet back-ground and prompt leptons originating from W / Z bosons [75]. Thismethodusesadatasample, withloosened identification cri-teria,dominatedbymultijetproductionandwithasmall contam-inationofelectroweak(EW)W/Z +jetsproduction.Theprobability thata jet frommultijetproductionwhichpassesthe loose selec-tion also satisfies the tight selection criteria is estimated inthis controlregion. The multijetpurityin thissample isimproved by subtracting,usingMCsimulation,theEWcontaminationtoremove biasdueto prompt-leptonsources.The efficiencyforprompt lep-tons passing the loose selection to also pass the tight selection isdetermined using tt MC¯ samples, corrected usingcomparisons of MC and data Z → events. The number of multijet back-ground events satisfying the selection criteria is estimated from these efficiencies using data events that satisfy all criteria, ex-cept that loose lepton identification criteriaare used. While this data-driven method is a significant improvement on the use of MCsimulation,thelownumberofeventsandinherentsystematic variations oftheEW contributionlead toa significant systematic uncertainty. Systematic uncertainties on the multijet background areevaluated [76] usingvariousdefinitionsofmultijetcontrol re-gionsandbyconsideringsystematicuncertainties associatedwith objectreconstructionandMC simulation.The uncertaintyonthis backgroundistakenas50%ofthetotalrateandtreatedas uncor-relatedbetweenselectedregions.

Fig. 3 shows the distributions of the reconstructed invariant massoftheWbosoncandidatefordataandforbackground pre-dictions in the 2-jet 1-tag VRHF and 4-jet 2-tag VR_t_¯_t validation regions.BackgroundtemplatesareﬁttodataineachVRusingthe same statistical method as for the signal region except that the normalizations oft¯t and W +jets backgrounds are constrained to thepost-ﬁtratesobtainedinthesignalregion(seeSection7). 6. Systematicuncertainties

Two primary sources of systematic uncertainty, experimental andmodeling, affect the reconstruction of the mtb distributions. Experimental uncertainties arise due tothe trigger selection, the objectreconstruction andidentiﬁcation, aswell astheobject en-ergy,momentumandmasscalibrationsandtheirresolutions. Mod-eling uncertaintiesresultinshapeandnormalizationuncertainties ofthe different MC samplesused to model the signal and back-grounds.These stem fromuncertainties inthe generator matrix-element calculation,the choice of partonshower and hadroniza-tionmodelsandtheirparametervalues,thePDFsetandthechoice ofrenormalizationandfactorizationscales.Theimpactonthe sig-nal and background event yields of the main systematic uncer-taintiesissummarizedinTable3,whereintheuncertaintyonthe overall yield is presented foreach backgroundsource. All values aregivenasapercentagechangeinoverallyieldandrepresentthe

priorvaluesassignedbeforeﬁtting.Thesourceofeachuncertainty isdescribedinthissection,anduncertaintiesareconsidered fully correlatedacrossalleightsignalregionsandamongprocesses, un-lessspeciﬁcallynoted.

The selection of jets and Emiss_T has an associated uncertainty relatedto thecalorimetercalibrationof theenergyscale andthe calorimeterresolution,aswell astothe identification/reconstruc-tion efficiencies of objects reconstructed using the calorimeter, sample flavor composition and corrections for pile-up and neu-trinosproduced inhadrondecays.Theuncertaintycontributedby eachsource istypically1–5%oftheexpectedeventratesandcan impacttheshapeofdifferentialdistributions.Inaddition,theEmiss_T calculationleadstoatypicaluncertaintyintheeventyieldofless than1%.

The process of b-tagging jets hasan uncertainty in the scale factorsrequiredtomatchthetaggingeﬃciencybetweendataand simulation. These uncertainties are evaluated independently for jetsarisingfromb-quarks,c-quarksandlight-quarksorgluons.The uncertaintyintheselectioneﬃciencyfortaggingb-quarksis typi-callysmall(1–5%perjet)exceptforveryhighpTjetswhereitcan increase to6% perjet,andthemis-taggingofc-/light-quarks and gluons can be as large as10%. Thesesources ofuncertainty can additionallyinducenon-uniformvariationsindifferential distribu-tionsofupto10%.

Theuncertaintyinthereconstructionefficiencyandacceptance ofleptons dueto trigger,reconstruction andselectionefficiencies in simulated samplesis roughly 1% of the total eventyield. The energy/momentumscaleandresolutionforleptonsiscorrectedin simulationtomatchdatameasurements,andtheresulting uncer-tainty intheefficiencyarising fromthesecorrectionsis lessthan 1–2%.

Thenormalizationofsimulatedsampleshasan associated un-certaintythatvariesbyproductionprocess.Theuncertaintyinthe cross section timesbranching fractionforsingle-top anddiboson productionis takenas 6% [77–79] and11% [80], respectively. An uncertaintyof20%isassumedfor Z +jetsrate, whichrepresentsa very smallbackground,in linewiththe modeling uncertainty as-signed for W +jets (see belowin thissection).The cross sections forthe tt and¯ W +jets samplesare normalizedusing freely ﬂoat-ingparameterswhosevaluesaredeterminedbyﬁttingtodata.All simulated samples that are normalized to the ATLAS luminosity measurement are assigneda luminosity uncertaintyof 2.1%. This uncertaintyisderived,followingamethodologysimilartothat de-tailedinRef. [81],fromacalibrationoftheluminosityscaleusing x– y beam-separation scans performed in August 2015 and May 2016.

Differences due to the choice of MC generator, fragmenta-tion/hadronization model, and initial/ﬁnal-state radiation model are treated as a source of uncertainty for the tt and¯ t-channel single-top-quark simulations. The uncertainty due to the choice of MC generator is evaluated as the difference in yield between the nominal choice of Powheg-Box and the alternative Mad-Graph5_aMC@NLO [82] generator, using Herwig++ [83,84] for showering in both instances. The uncertainty due to the frag-mentation/hadronizationmodelisevaluatedbycomparing Pythia6 and Herwig++ simulated samples. Variations of the amount of additional radiation are studied by changing the scale of the hard-scatter process and the scales in parton-shower simulation simultaneously using the Powheg-Box+Pythia6 set-up. In these samples,avariationofthefactorizationandrenormalizationscales by a factor oftwo is combined withthe Perugia2012radLo tune and a variation of both scales by a factor of 0.5 is combined withthePerugia2012radHi tune [45].Inthecaseoft¯t production the Powheg-Box hdamp parameter, which controls the transverse momentumoftheﬁrst additionalemission beyondthe Born

(6)

con-Fig. 3. DistributionsofthereconstructedinvariantmassoftheWbosoncandidateinthe(top)2-jet1-tagVRHFand(bottom)4-jet2-tagVRt¯tvalidationregions.Background

templatesarefittodataineachVRusingthesamestatisticalmethodasforthesignalregionexceptthatthenormalizationsoftt and¯ W +jetsbackgroundsareconstrained tothepost-fitratesobtainedinthesignalregion(seeSection7).Thepre-fitlinepresentsthebackgroundpredictionbeforethefitisperformed.Uncertaintybandsinclude allthesystematicandstatisticaluncertainties.TheresidualdifferencebetweenthedataandMCyieldsisshownasaratiointhebottomportionofeachfigure,whereinthe errorbarsonthedatapointscorrespondtothedataPoissonuncertainty.

ﬁguration, is also changed simultaneously, using values of mtop and2×mtop,respectively.AnuncertaintyassociatedwiththeNLO calculation of W t production [85] isevaluated by comparingthe baselinesample generatedwiththediagramremovalschemetoa W t samplegeneratedwiththediagramsubtractionscheme.

Thesedifferencesyieldrelativevariationsinshapeand normal-ization of1–3% on average, although the variation can be larger than 10% in the highest mtb regions probed. The normalization componentofthesemodeling uncertaintiesis removedforthett¯ samples becausethe overall normalizationis determined via the datainthiscase.

Differences between the predictions for the ratio of 2-jet to 3-jet yields from different showering simulations were studied forthett and¯ W +jetssimulation.Thesedifferencesare estimated by simultaneously varying the renormalization and factorization scales,andbyusingdifferentMCgenerators.Whileonlysmall dif-ferenceswereobservedfort¯t simulation,theratiooftheyieldsof

2-jetto3-jetselectionsinW +jetssimulationvariedbyupto20%. Thus, an additional uncertaintyof20% is assignedtothe W +jets yieldinthe3-jetselection.4

Uncertainties in W +jets modeling are determined by compar-ing the nominal Sherpa simulation with an alternative sample produced withthe MadGraph5_aMC@NLO generatorinterfacedto Pythia8 forpartonshoweringandhadronization.The uncertainty in our knowledge of the ﬂavor fraction in the W +jets sample is tested by splittingthe W +jets sample intolight-quark/gluon and heavy-ﬂavor components andby decorrelating the W +jets shape uncertaintybetween2-jetand3-jetevents.Ineachcase,no signif-icanteffectontheextractedresultsisobserved.

4 _For_the_{Z +jets}_background_a_similar_variation_could_be_expected,_but_since_this

(7)

Table 3

Impactofthemainsourcesofuncertaintyonthesignalandbackgroundeventyields.Allvaluesaregivenasapercentage changeintheoverallyieldandrepresentthepriorvaluesassignedbeforeﬁtting.Uncertaintiesforthesignalaregivenfora WR masshypothesisof2 TeV.Uncertaintiesinthebackgroundarethesameforallsignalmasses.Systematicuncertainties

inthenormalization,2-jetvs3-jetregioncross-extrapolation,andreconstructedmtbshapeofthesignalandbackground

processesaredescribedinthetext.Sourcesofuncertaintymayaffectboththetotaleventyieldandtheshapeofthemtb

distribution.An“S”indicatesthatashapevariationhasbeenincluded,inadditiontotheratevariation,duetothesources listed.“U”referstoregionsthatarenotcorrelatedwithoneanotherand“F”referstoanormalizationthatﬂoatsfreely. Incertaininstancesoffreelyﬂoatingnormalizations,theratevariationofsystematiceffectsisremoved,thusleavingonly ashapevariation.Suchcasesareindicatedwitha“*”symbol.The“Jets”columnincludesuncertaintiesrelatedtoEmiss

T .A

rangeofvaluescorrespondtothelowestandthehighestvaluesdeterminedacrossdifferentchannelsintheSR.Theﬁnal columndescribestheuncertaintyinextrapolatingeventyieldsbetweenthe2-jetand3-jetselections.

Process Norm. Lumi. b-Tagging [S] Jets [S] Leptons Modeling [S] 2j/3j Extrap.

W_R – 2.1 8–12 1–4 1–2 – –

t¯t F – 2–6 4–8 1–2 0 [*] –

W +jets F – 6–15 2–12 1–3 0 [*] 20

Z +jets 20 2.1 6–12 2–9 1–3 – –

Diboson 11 2.1 3–10 2–8 1–2 – –

Single top quark 6 2.1 2–7 1–4 1–2 6–22 –

Multijet 50 [U] – – – – – –

Table 4

Thenumbersofsignaland backgroundeventsand thenumbersofobserveddataeventsareshowninthe2-jet1-tagand 3-jet1-tagsignalregions. Forsignal,the valuescorrespondtoexpectedeventyieldsandquoteduncertaintiesaccountfor thestatisticaluncertaintyofthenumberofeventsinthesimulatedsamples.Thenumberofbackgroundeventsisobtained followingaMLﬁttothedataanduncertaintiescontainstatisticalandsystematicuncertainties.

2-jet 1-tag (e±) 2-jet 1-tag (μ±) 3-jet 1-tag (e±) 3-jet 1-tag (μ±) W_R (1.0 TeV) 1517 ± 32 2030 ± 40 1159 ± 31 1665 ± 35 WR (2.0 TeV) 83.4 ± 1.7 132.9 ± 2.1 105.0 ± 1.9 167.4 ± 2.2 WR (3.0 TeV) 4.7 ± 0.1 10.4 ± 0.2 7.0 ± 0.2 15.7 ± 0.2 WR (4.0 TeV) 0.43 ± 0.01 1.01 ± 0.02 0.64 ± 0.02 1.62 ± 0.03 WR (5.0 TeV) 0.076 ± 0.002 0.153 ± 0.003 0.096 ± 0.003 0.232 ± 0.004 t¯t 1112 ± 23 1505 ± 28 3220 ± 50 4090 ± 70 Single-top 472 ± 20 657 ± 25 482 ± 21 624 ± 24 W +jets 520 ± 50 1280 ± 120 550 ± 40 1130 ± 90 Multijets 358 ± 35 630 ± 100 196 ± 20 390 ± 60 Z +jets, diboson 129 ± 14 211 ± 19 128 ± 12 242 ± 20 Total background 2590 ± 60 4290 ± 160 4580 ± 70 6470 ± 130 Data 2622 4260 4555 6433

Theuncertaintyintheyieldofsimulatedt¯t backgroundevents due to the choice of PDF is evaluated using the PDF4LHC rec-ommendations [86]. The statisticaluncertainty ofthelimited MC samplesisincludedineachhistogrambinofthemtbdistribution. 7.Results

In orderto test for the presence ofa massive resonance,the mtb templatesobtainedfromthesignalandbackgroundsimulated eventsamplesareﬁttodatausinga binnedmaximum-likelihood (ML) approach based on the RooStats framework [87–89]. Each signal region selection is considered simultaneously as an inde-pendentsearchchannel,foratotalofeightregions corresponding tomutually exclusive categoriesof electron andmuon, 2-jet and 3-jet,and1-b-tagand2-b-tags.

Thenormalizationsofthet¯t and W +jetsbackgroundsare free parameters inthe fit,while other backgroundnormalizations are assignedGaussian priors based ontheir respective normalization uncertainties.The systematicuncertainties described inSection 6 are incorporated in the fit as nuisance parameters with correla-tions acrossregions andprocesses takenintoaccount. The signal normalizationisafreeparameterinthefit.

The expected and observed event yields after the ML ﬁt are showninTables4and5andcorrespondtoanintegrated luminos-ityof 36.1 fb−1. The ﬁttedt¯t and W +jets rates relative to their nominalpredictions arefoundtobe 0.98±0.04 and0.78±0.19, respectively. Forthese two backgrounds the total uncertainty re-portedintheeventyieldtablesissmallerthantheuncertaintyin

the ﬁtted normalizationfactor because there are anticorrelations betweennuisanceparametersinthelikelihoodﬁt.

Themtb distributionsfortheSRafterthe MLﬁtare shownin Figs. 4and5.An expectedsignalcontributioncorresponding toa W_R bosonwithamassof2.0 TeV isshownasadashedhistogram overlay.The binningofthemtb distributionischosen tooptimize thesearchsensitivitywhileminimizingstatisticalﬂuctuations. Re-quirements are imposed on the expectednumber of background eventsperbin,andthebinwidthisadaptedtoaresolution func-tionthatrepresentsthewidthofthereconstructedmasspeakfor eachstudiedW_R bosonsignalsample.

Fora W_R boson witha massof 2 TeV and nominal g/g=1 coupling the total expected uncertainty in estimating the signal strength5 _is _12%. _The _total _systematic_uncertainty _is _9%, _and _the

largest uncertainties are due to the tt generator¯ (4.0%), jet en-ergyscale(JES)(2.8%),tt showering¯ (2.5%),t¯t normalization (2.0%) andJES η intercalibrationmodeling (1.3%). Forresonanceswitha mass of2.5 TeV or above,the data Poissonuncertainty becomes thelargestuncertaintyinestimatingthesignal rate,whilethe to-talsystematicuncertaintyisdominatedbytheuncertaintyonthe b-taggingeﬃciency.

As nosigniﬁcantexcess overthebackground predictionis ob-served, upper limits at the 95% conﬁdence level (CL) are set on the production cross section times the branching fraction for

5 _The_signal_strength_is_deﬁned_as_the_ratio_of_the_signal_cross_section_estimated

(8)

Table 5

Thenumbersofsignalandbackgroundeventsandthenumbersofobserveddataeventsareshowninthe2-jet2-tagand 3-jet2-tagsignalregions.Forsignal, thevaluescorrespondtoexpectedeventyieldsand quoteduncertaintiesaccountfor thestatisticaluncertaintyofthenumberofeventsinthesimulatedsamples.Thenumberofbackgroundeventsisobtained followingaMLﬁttothedataanduncertaintiescontainstatisticalandsystematicuncertainties.

2-jet 2-tag (e±) 2-jet 2-tag (μ±) 3-jet 2-tag (e±) 3-jet 2-tag (μ±) W_R(1.0 TeV) 1584 ± 35 2060 ± 40 1241 ± 30 1749 ± 34 W_R(2.0 TeV) 33.5 ± 1.0 55.5 ± 1.2 51.6 ± 1.2 84.3 ± 1.5 W_R(3.0 TeV) 1.4 ± 0.1 2.6 ± 0.1 2.5 ± 0.1 5.1 ± 0.1 WR(4.0 TeV) 0.131 ± 0.007 0.25 ± 0.01 0.21 ± 0.01 0.46 ± 0.01 WR(5.0 TeV) 0.035 ± 0.002 0.053 ± 0.002 0.044 ± 0.002 0.080 ± 0.002 tt¯ 536 ± 14 789 ± 16 2459 ± 31 3200 ± 40 Single-top 121 ± 6 176 ± 10 235 ± 12 347 ± 17 W +jets 28 ± 6 42 ± 4.0 50 ± 5 97 ± 9 Multijets 36 ± 6 71 ± 13 95 ± 11 135 ± 22 Z +jets, diboson 2.5 ± 0.4 11.5 ± 1.3 21.2 ± 2.1 26.9 ± 2.3 Total background 723 ± 16 1088 ± 21 2859 ± 33 3810 ± 50 Data 683 1091 2869 3797

Fig. 4. Post-ﬁtdistributionsofthereconstructedmassofthe WR bosoncandidateinthe(top)2-jet1-tagand(bottom)2-jet2-tagsignalregions,for(left)electronand

(right)muonchannels.AnexpectedsignalcontributioncorrespondingtoaWRbosonmassof2.0 TeV enhanced20timesisshown.Thepre-ﬁtlinepresentsthebackground

predictionbeforetheﬁtisperformed.Uncertaintybandsincludeallthesystematicandstatisticaluncertainties.TheresidualdifferencebetweenthedataandMCyieldsis shownasaratiointhebottomportionofeachﬁgure,whereintheerrorbarsonthedatapointscorrespondtothedataPoissonuncertainty.

(9)

Fig. 5. Post-ﬁtdistributionsofthereconstructedmassoftheWR bosoncandidateinthe(top)3-jet1-tagand(bottom)3-jet2-tagsignalregions,for(left)electronand

(right)muonchannels.AnexpectedsignalcontributioncorrespondingtoaWRbosonmassof2.0 TeV enhanced20timesisshown.Thepre-ﬁtlinepresentsthebackground

predictionbeforetheﬁtisperformed.Uncertaintybandsincludeallthesystematicandstatisticaluncertainties.TheresidualdifferencebetweenthedataandMCyieldsis shownasaratiointhebottomportionofeachﬁgure,whereintheerrorbarsonthedatapointscorrespondtothedataPoissonuncertainty.

each model. The limits are evaluated using a modiﬁed frequen-tistmethodknownasCLs [90] with aproﬁle-likelihood-ratiotest statistic [91] usingtheasymptoticapproximation.

The95% CLupperlimitsontheproductioncross section mul-tipliedbythebranchingfractionforW_R →tb are¯ showninFig.6 asa function ofthe resonancemass. Theobserved andexpected limitsarederived using alinear interpolationbetweensimulated signalmasshypotheses.Theexclusionlimitsrangebetween4.9 pb and2.9×10−2_pb_for_W

Rbosonmassesfrom0.5 TeV to5 TeV.The lowerobservedlimitsforW_R massesaround2.5 TeV areduetoa deﬁcitofdataeventsinthe2–2.5 TeV m_t_b_¯ rangeinthe2-jet1-tag and3-jet1tag(muon)signalregions.TheexistenceofW_R bosons withmasses_mW

R<3.15 TeV isexcludedforthe ZTOPbenchmark modelforW_R,assumingthat the W_R coupling gis equaltothe SMweakcouplingconstant g.

Limitsonthe ratioofcouplings g/g asa functionofthe W_R bosonmasscanbederivedfromthelimitsontheW_R bosoncross section.Limitscanalsobesetforg/g>1,asmodelsremain per-turbativeuptoaratioofaboutﬁve [30].TheW_R bosoncross

sec-tionhasa dependenceonthecoupling g,comingfromthe vari-ation oftheresonance width. Thescaling ofthe W_R boson cross section asafunction of g/g andmW isestimatedatNLO using the ZTOP generator. Inaddition, speciﬁc signal samplesare used inordertotakeintoaccount theeffectontheacceptanceandon kinematicaldistributionsoftheincreasedsignalwidth(compared withthenominalsamples)forvaluesofg/g>1.Fig.7showsthe excludedparameterspaceasafunctionoftheW_R resonancemass, whereintheeffectofincreasing W_R widthforcouplingvaluesof g/g>1 isincludedforsignalacceptanceanddifferential distribu-tions.Thelowestobserved(expected)limitong/g,obtainedfora W_R bosonmassof0.75 TeV,is0.13(0.13).

The ATLAS experimenthas recently searched for W_R →tb in¯ thefullyhadronicﬁnal state [27] using 36.1 fb−1,corresponding tothesamedatacollectionperiodastheanalysispresentedhere. As thesetwo searches are complementary and use mutually or-thogonaleventselections,amoregeneralandpowerfulsearchfor W_R→tb production¯ canbeobtainedviatheirstatistical combina-tion.Thesignal simulationwasproducedinthesamemannerfor

(10)

Fig. 6. Upperlimitsatthe95%CLonthe WR productioncrosssectiontimesthe W_R→tb branching¯ fractionasafunctionofresonancemass,assumingg/g=1. Thesolid curvecorrespondsto theobserved limit,whilethe dashed curveand shadedbandscorrespondtothelimitexpectedintheabsenceofsignalandthe re-gionsenclosingone/twostandarddeviation(s.d.)ﬂuctuationsoftheexpectedlimit. Thepredictionmadebythebenchmarkmodelgenerator ZTOP [30],anditswidth thatcorrespondtovariationsduetoscaleandPDFuncertainty,arealsoshown.

Fig. 7. Observedand expected95%CLlimit onthe ratio g/g,asa functionof resonancemass,for right-handedW coupling.Theﬁlledareacorrespondtothe observedlimitwhilethedashed lineand theonestandarddeviation(s.d.) band correspondtheexpectedlimit.TheimpactoftheincreasedWR widthforcoupling

valuesofg/g>1 ontheacceptanceandonkinematicaldistributionsistakeninto account.

bothsearches,andthesimulationofsharedbackgroundsourcesis obtainedwithidenticalorsimilartools.Thefullyhadronicsearch hasabackgrounddominatedbyQCDmultijetproduction,whichis estimatedviadata-drivenmethods.Thesmallercontributionfrom tt and¯ singlyproducedtopquarksiscommontothetwoanalyses, andthusallsystematicuncertaintiesrelatedtoshared reconstruc-tionorselectionmethodsaretreatedasfullycorrelated.

Theresultofthecombinationofthecrosssectiontimes branch-ing fraction limits of the leptonic and fully hadronic analyses is shownin Fig. 8. The individual limits andtheir combinationare showninFig.9.Theexpectedlimitsproducedbythetwosearches are similar above a resonance mass of 2 TeV, below which the fullyhadronicsearchsuffers duetoineﬃciency fromdijettrigger thresholdscausingitnottocontributeforresonancemassesbelow 1 TeV. Thus, theexpected limitson the productioncross section multiplied by the branching fraction improve by approximately 35%above1 TeV andthecombinedresultraisesthelowerlimiton

Fig. 8. Observedandexpected95%CLupperlimitontheW_R productioncross sec-tiontimesthe WR→tb branching¯ fractionasafunctionofresonancemassfor

the combinationofsemileptonic andhadronic [27] W→tb searches,¯ assuming g/g=1.Thehadronicsearchcoversamassrangebetween1.0and5.0 TeV.The solidblackcurvecorrespondstotheobservedlimit,whilethedashed curveand shadedbandscorrespondtothelimitexpectedintheabsenceofsignalandthe re-gionsenclosingone/twostandarddeviation(s.d.)ﬂuctuationsoftheexpectedlimit. Thepredictionmadebythebenchmarkmodelgenerator ZTOP [30],anditswidth thatcorrespondtovariationsduetoscaleandPDFuncertainty,arealsoshown.

Fig. 9. Observedandexpected95%CLupperlimitontheWR productioncross

sec-tiontimestheWR→tb branching¯ fractionasafunctionofresonancemass,forthe

semileptonicandhadronic [27] W→tb searches,¯ aswellastheircombination.The solidcurvescorrespondtotheobservedupperlimits,whilethedashedlinesarethe expectedlimits.

the W_R massto3.25 TeV.Ontheother hand,thegainfrom com-bining the observed crosssection times branching fractionlimits israthermodest,comparedwiththeresultoftheleptonicanalysis only,becauseofupwardﬂuctuationsobservedinthefullyhadronic analysisdata.

8. Conclusion

Asearchfor W_R→tb in¯ theleptonplus jetsﬁnalstate is per-formedusing36.1 fb−1 of13 TeV pp collisiondatacollectedwith the ATLAS detector at the LHC. No signiﬁcant excess of events is observedabove the SM predictions.Upper limitsare placedat the95%CLonthecrosssectiontimesbranchingfraction, σ(pp→ W_R→tb¯),rangingbetween4.9 pband2.9×10−2 pbinthemass range of 0.5 TeV to 5 TeV for a right-handed W boson. Exclu-sion limitsare alsocalculatedforthe ratioofthe couplings g/g

(11)

andthelowest observedlimit, obtainedfora W_R boson massof 0.75 TeV,is0.13.Astatisticalcombinationofthecross-section lim-itsisperformedwiththeresultsobtainedwhenthefullyhadronic decaysofW_R →tb are¯ considered.Theupperlimitsonthecross section times branching fraction improve by approximately 35% above 1 TeV.Masses below3.15(3.25) TeV are excluded for W_R bosonsinthebenchmark ZTOPmodelforthesemileptonic (com-binedsemileptonicandhadronic)scenarios.

Acknowledgements

We thankCERN for thevery successful operation ofthe LHC, aswell asthe support stafffromour institutions without whom ATLAScouldnotbeoperatedeﬃciently.

WeacknowledgethesupportofANPCyT,Argentina;YerPhI, Ar-menia; ARC, Australia; BMWFW and FWF, Austria; ANAS, Azer-baijan;SSTC, Belarus; CNPq and FAPESP, Brazil; NSERC, NRC and CFI,Canada; CERN; CONICYT,Chile; CAS, MOSTandNSFC, China; COLCIENCIAS, Colombia; MSMT CR, MPO CR and VSC CR, Czech Republic;DNRFandDNSRC,Denmark;IN2P3-CNRS,CEA-DRF/IRFU, France; SRNSFG, Georgia; BMBF, HGF, andMPG, Germany; GSRT, Greece;RGC,HongKong SAR,China;ISFandBenoziyo Center, Is-rael; INFN, Italy; MEXT and JSPS, Japan; CNRST, Morocco; NWO, Netherlands;RCN, Norway;MNiSW andNCN, Poland;FCT, Portu-gal; MNE/IFA, Romania; MES of Russiaand NRC KI, Russian Fed-eration; JINR; MESTD, Serbia; MSSR, Slovakia; ARRS and MIZŠ, Slovenia;DST/NRF,SouthAfrica;MINECO,Spain;SRCand Wallen-berg Foundation, Sweden; SERI, SNSF and Cantons of Bern and Geneva, Switzerland; MOST, Taiwan; TAEK, Turkey; STFC, United Kingdom;DOEandNSF, UnitedStatesofAmerica. Inaddition, in-dividualgroupsandmembershavereceivedsupportfromBCKDF, theCanadaCouncil,Canarie,CRC,ComputeCanada,FQRNT,andthe OntarioInnovationTrust, Canada;EPLANET, ERC,ERDF,FP7, Hori-zon 2020 andMarie Skłodowska-Curie Actions, European Union; Investissements d’Avenir Labex and Idex, ANR, Région Auvergne andFondationPartagerleSavoir,France;DFGandAvHFoundation, Germany;Herakleitos,ThalesandAristeiaprogrammesco-ﬁnanced byEU-ESFandtheGreekNSRF;BSF,GIF andMinerva,Israel;BRF, Norway; CERCA Programme Generalitat de Catalunya, Generalitat Valenciana,Spain;theRoyalSocietyandLeverhulmeTrust,United Kingdom.

The crucialcomputing support fromall WLCG partners is ac-knowledged gratefully, in particular from CERN, the ATLAS Tier-1 facilities at TRIUMF (Canada), NDGF (Denmark, Norway, Swe-den),CC-IN2P3(France),KIT/GridKA(Germany),INFN-CNAF(Italy), NL-T1(Netherlands),PIC(Spain),ASGC(Taiwan),RAL(UK)andBNL (USA),theTier-2facilitiesworldwideandlargenon-WLCGresource providers.Majorcontributorsofcomputingresources arelistedin Ref. [92].

References

[1]G.Burdman,B.A.Dobrescu, E.Ponton, Resonancesfrom twouniversalextra dimensions,Phys.Rev.D74(2006)075008,arXiv:hep-ph/0601186.

[2]H.-C.Cheng,C.T.Hill,S.Pokorski,J.Wang,Standardmodelinthe latticized bulk,Phys.Rev.D64(2001)065007,arXiv:hep-th/0104179.

[3]T.Appelquist,H.-C.Cheng,B.A.Dobrescu,Boundsonuniversalextra dimen-sions,Phys.Rev.D64(2001)035002,arXiv:hep-ph/0012100.

[4]J.C.Pati,A.Salam,Leptonnumberasthefourth“color”,Phys.Rev.D10(1974) 275–289.

[5]R.N. Mohapatra, J.C. Pati, Left-rightgauge symmetryand an “isoconjugate” modelofCPviolation,Phys.Rev.D11(1975)566–571.

[6]G.Senjanovic,R.N.Mohapatra,Exactleft-rightsymmetryandspontaneous vi-olationofparity,Phys.Rev.D12(1975)1502–1505.

[7]M.Perelstein,LittleHiggsmodelsandtheirphenomenology,Prog.Part.Nucl. Phys.58(2007)247–291,arXiv:hep-ph/0512128.

[8]M.J.Dugan,H.Georgi,D.B.Kaplan,AnatomyofacompositeHiggsmodel,Nucl. Phys.B254(1985)299.

[9]K.Agashe,R.Contino,A.Pomarol,TheminimalcompositeHiggsmodel,Nucl. Phys.B719(2005)165–187,arXiv:hep-ph/0412089.

[10]D0Collaboration,SearchforWbosonsdecayingtoanelectronandaneutrino withtheD0detector,Phys.Rev.Lett.100(2008)031804,eprint,arXiv:0710. 2966.

[11]CDFCollaboration,SearchforanewheavygaugebosonWwithevent signa-tureelectron+missingtransverseenergyinpp collisions¯ at√s=1.96 TeV, Phys.Rev.D83(2011)031102,arXiv:1012.5145.

[12]CMSCollaboration,Searchforphysicsbeyondthestandardmodelinﬁnalstates with alepton andmissingtransverseenergyinproton–proton collisionsat

√

s=8 TeV,Phys.Rev.D91(2015)092005,arXiv:1408.2745 [hep-ex].

[13]ATLASCollaboration,Searchfornewparticlesineventswithoneleptonand missingtransversemomentuminpp collisionsat√s=8 TeV withtheATLAS detector,J.HighEnergyPhys.09(2014)037,arXiv:1407.7494 [hep-ex].

[14]ATLASCollaboration,Searchfornewresonancesineventswithoneleptonand missingtransversemomentuminpp collisionsat√s=13 TeV withtheATLAS detector,Phys.Lett.B762(2016)334,arXiv:1606.03977 [hep-ex].

[15]CMSCollaboration,SearchforheavygaugeW bosonineventswith an en-ergeticleptonandlargemissingtransversemomentumat√s=13 TeV,Phys. Lett.B770(2017)278,arXiv:1612.09274 [hep-ex].

[16]C.Patrignani,etal.,Rev.ParticlePhys.,Chin.Phys.C40(2016)100001.

[17]D.J.Muller,S.Nandi,Topﬂavor:aseparateSU(2)forthethirdfamily,Phys.Lett. B(ISSN 0370-2693)383(1996)345–350.

[18]E.Malkawi,T.M.Tait,C.Yuan,Amodelofstrongﬂavordynamicsforthetop quark,Phys.Lett.B385(1996)304–310,arXiv:hep-ph/9603349.

[19]D0Collaboration,SearchforW→tb resonanceswithleft- andright-handed couplingstofermions,Phys.Lett.B699(2011)145–150,arXiv:1101.0806.

[20]CDFCollaboration,Searchfortheproductionofnarrowt anti-b resonancesin 1.9fb−1_of_p_{p collisions}_¯ _at√_s₌₁_._{96 TeV,}_Phys._Rev._Lett.₁₀₃₍₂₀₀₉₎_041801,

arXiv:0902.3276.

[21]ATLAS_√ Collaboration,Searchfortb resonancesinproton–protoncollisionsat

s=7 TeV withtheATLASdetector,Phys.Rev.Lett.109(2012)081801,arXiv: 1205.1016.

[22]CMSCollaboration,SearchforaW bosondecayingtoabottomquarkanda topquarkinpp collisionsat√s=7 TeV,Phys.Lett.B718(2013)1229–1251, arXiv:1208.0956.

[23]CMSCollaboration,SearchforW→tb decaysinthelepton+jetsﬁnalstate inpp collisionsat√s=8 TeV,J.HighEnergyPhys.05(2014)108,arXiv:1402. 2176.

[24]ATLASCollaboration,SearchforW→tb intheleptonplusjetsﬁnalstatein proton–protoncollisionsat acentre-of-massenergyof√s=8 TeV withthe ATLASdetector,Phys.Lett.B743(2015)235–255,arXiv:1410.4103 [hep-ex].

[25]CMSCollaboration,Searchforheavyresonancesdecayingtoatopquarkanda bottomquarkinthelepton+jetsﬁnalstateinproton–protoncollisionsat13 TeV,Phys.Lett.B777(2018)39–63,arXiv:1708.08539 [hep-ex].

[26]ATLAS_√ Collaboration,Searchfor W→tb→qqbb decaysin pp collisionsat

s=8 TeV withtheATLASdetector,Eur.Phys.J.C75(2015)165,arXiv:1408. 0886 [hep-ex].

[27]ATLAS Collaboration,Searchfor W→tb decaysinthehadronicﬁnalstate usingpp collisionsat√s=13 TeV withtheATLASdetector,Phys.Lett.B781 (2018)327–348,arXiv:1801.07893 [hep-ex].

[28]ATLASCollaboration,TheATLASexperimentattheCERNlargehadroncollider, J.Instrum.3(2008)S08003.

[29]Z.Sullivan,FullydifferentialWproductionanddecayatnext-to-leadingorder inQCD,Phys.Rev.D66(2002)075011,arXiv:hep-ph/0207290.

[30]D. Duffty,Z.Sullivan,Modelindependentreachfor W -primebosonsatthe LHC,Phys.Rev.D86(2012)075018,arXiv:1208.4858.

[31] ATLASCollaboration,ATLASInsertableB-LayerTechnicalDesignReport, CERN-LHCC-2010-013.ATLAS-TDR-19,https://cds.cern.ch/record/1291633,2010. [32] ATLASCollaboration,ATLASInsertableB-LayerTechnicalDesignReport

Adden-dum,AddendumtoCERN-LHCC-2010-013,ATLAS-TDR-019,https://cds.cern.ch/ record/1451888,2012.

[33] ATLASCollaboration,Performanceofthe ATLAStriggersystemin2015,Eur. Phys.J.C77(2017)317,https://doi.org/10.1140/epjc/s10052-017-4852-3. [34]J.Alwall,M.Herquet,F.Maltoni, O.Mattelaer,T.Stelzer,MadGraph5:going

beyond,J.HighEnergyPhys.06(2011)128,arXiv:1106.0522.

[35]J.Alwall,etal.,Theautomatedcomputationoftree-levelandnext-to-leading orderdifferentialcrosssections,andtheirmatchingtopartonshower simula-tions,J.HighEnergyPhys.07(2014)079,arXiv:1405.0301 [hep-ph].

[36]C.Degrande,etal.,UFO–TheUniversalFeynRulesOutput,Comput.Phys. Com-mun.183(2012)1201–1214,arXiv:1108.2040.

[37]A.Alloul,N.D.Christensen,C.Degrande,C.Duhr,B.Fuks,FeynRules 2.0–a completetoolboxfortree-levelphenomenology,Comput.Phys.Commun.185 (2014)2250–2300,arXiv:1310.1921.

[38]T.Sjöstrand,S.Mrenna,P.Z.Skands,AbriefintroductiontoPYTHIA8.1,Comput. Phys.Commun.178(2008)852–867,arXiv:0710.3820.

[39]R.D.Ball,L.DelDebbio,S.Forte,A.Guffanti,J.I.Latorre,etal.,Aﬁrstunbiased globalNLOdeterminationofpartondistributionsandtheiruncertainties,Nucl. Phys.B838(2010)136–206,arXiv:1002.4407 [hep-ph].

(12)

[40] ATLAS Collaboration, ATLAS Pythia 8 tunes to 7TeV data, ATL-PHYS-PUB-2014-021,https://cds.cern.ch/record/1966419,2014.

[41]S.Alioli,P.Nason,C.Oleari,E.Re,AgeneralframeworkforimplementingNLO calculationsinshowerMonteCarloprograms:thePOWHEGBOX,J.High En-ergyPhys.06(2010)043,arXiv:1002.2581.

[42]S.Frixione,P.Nason,G.Ridolﬁ,Apositive-weightnext-to-leading-orderMonte Carloforheavyﬂavourhadroproduction,J.HighEnergyPhys.09(2007)126, arXiv:0707.3088.

[43]H.-L.Lai,etal.,Newpartondistributionsforcolliderphysics,Phys.Rev.D82 (2010)074024,arXiv:1007.2241 [hep-ph].

[44]T. Sjöstrand, et al., High-energy physics event generation with PYTHIA 6.1, Comput.Phys.Commun.135(2001)238–259,arXiv:hep-ph/0010017.

[45]P.Z.Skands,TuningMonteCarlogenerators:ThePerugiaTunes,Phys.Rev.D82 (2010)074018,arXiv:1005.3457 [hep-ph].

[46]M.Cacciari,M.Czakon,M.Mangano,A.Mitov,P.Nason,Top-pairproduction athadroncolliderswithnext-to-next-to-leadinglogarithmicsoft-gluon resum-mation,Phys.Lett.B710(2012)612–622,arXiv:1111.5869.

[47]M.Beneke,P.Falgari,S.Klein,C.Schwinn,Hadronictop-quarkpairproduction withNNLLthresholdresummation,Nucl.Phys.B855(2012)695–741,arXiv: 1109.1536.

[48]P.Baernreuther, M. Czakon,A.Mitov, Percent-level-precisionphysics at the tevatron:next-to-next-to-leadingorderQCDcorrectionstoqq¯→t¯t+X ,Phys. Rev.Lett.109(2012)132001,arXiv:1204.5201.

[49]M.Czakon,P.Fiedler,A.Mitov,Totaltop-quarkpair-productioncrosssection at hadroncollidersthrough O(α4

S),Phys.Rev.Lett.110 (25)(2013)252004,

arXiv:1303.6254.

[50]M.Czakon,A.Mitov,NNLOcorrectionstotoppairproductionathadron col-liders:thequark–gluonreaction,J.HighEnergyPhys. 01(2013)080,arXiv: 1210.6832.

[51]M.Czakon,A.Mitov,NNLOcorrectionstotop-pairproductionathadron collid-ers:theall-fermionicscatteringchannels,J.HighEnergyPhys.12(2012)054, arXiv:1207.0236.

[52]M.Czakon,A.Mitov,Top++:aprogramforthecalculationofthetop-pair cross-section at hadron colliders,Comput. Phys. Commun.(ISSN 0010-4655)185 (2014)2930–2938.

[53]T.Gleisberg,etal.,EventgenerationwithSHERPA1.1,J.HighEnergyPhys.02 (2009)007,arXiv:0811.4622 [hep-ph].

[54]S.Höche,F.Krauss,S.Schumann,F.Siegert,QCDmatrixelementsand trun-catedshowers,J.HighEnergyPhys.05(2009)053,arXiv:0903.1219 [hep-ph].

[55]T.Gleisberg,S.Hoche,Comix,anewmatrixelementgenerator,J.HighEnergy Phys.12(2008)039,arXiv:0808.3674 [hep-ph].

[56]S.Schumann,F.Krauss,APartonshoweralgorithmbasedonCatani–Seymour dipolefactorisation,J.HighEnergyPhys.03(2008)038,arXiv:0709.1027 [hep -ph].

[57]R.Hamberg,W.VanNeerven,T.Matsuura,Acompletecalculationoftheorder

a2

s correctiontotheDrell–YanKfactor,Nucl.Phys.B359(1991)343–405;

R.Hamberg,W.VanNeerven,T.Matsuura,Nucl.Phys.B644(2002)403 (Erra-tum).

[58]C.Anastasiou,L.J.Dixon,K.Melnikov,F.Petriello,HighprecisionQCDathadron colliders:electroweakgaugebosonrapiditydistributionsatNNLO,Phys.Rev.D 69(2004)094008,arXiv:hep-ph/0312266.

[59]J.Pumplin, etal.,Newgeneration ofpartondistributionswith uncertainties fromglobalQCDanalysis,J.HighEnergyPhys.07(2002)012,arXiv:hep-ph/ 0201195 [hep-ph].

[60]ATLASCollaboration,MeasurementoftheZ/y∗bosontransversemomentum distributioninpp collisionsat √s=7 TeV withtheATLAS detector,J.High EnergyPhys.09(2014)55.

[61]D.J.Lange,TheEvtGenparticledecaysimulationpackage,Nucl.Instrum. Meth-odsA462(2001)152–155.

[62] ATLAS Collaboration, Summary of ATLAS Pythia 8 tunes, ATL-PHYS-PUB-2012-003,https://cds.cern.ch/record/1474107,2012.

[63]A.D.Martin,W.J.Stirling,R.S.Thorne,G.Watt,PartondistributionsfortheLHC, Eur.Phys.J.C63(2009)189–285,arXiv:0901.0002 [hep-ph].

[64]ATLAS Collaboration,TheATLAS simulationinfrastructure,Eur.Phys.J. C70 (2010)823–874,arXiv:1005.4568.

[65]S.Agostinelli,etal.,GEANT4–asimulationtoolkit,Nucl.Instrum.MethodsA 506(2003)250–303.

[66] W.Lampl, et al., Calorimeter clustering algorithms:description and perfor-mance,ATL-LARG-PUB-2008-002,https://cds.cern.ch/record/1099735,2008.

[67]ATLASCollaboration,ElectronandphotonenergycalibrationwiththeATLAS detectorusingLHCRun1data,Eur.Phys.J.C74(2014)3071,arXiv:1407.5063 [hep-ex].

[68]ATLASCollaboration,Electronreconstructionandidentiﬁcationeﬃciency mea-surementswiththeATLASdetectorusingthe2011LHCproton–protoncollision data,Eur.Phys.J.C74(2014)2941,arXiv:1404.2240 [hep-ex].

[69]ATLASCollaboration,MuonreconstructionperformanceoftheATLASdetector inproton–protoncollisiondataat√s=13 TeV,Eur.Phys.J.C76(2016)292, arXiv:1603.05598 [hep-ex].

[70]M.Cacciari,G.P.Salam,G.Soyez,Theanti-ktjetclusteringalgorithm,J.High

EnergyPhys.04(2008)063,arXiv:0802.1189.

[71]ATLASCollaboration,Performanceofpile-upmitigationtechniquesforjetsin

pp collisionsat√s=8 TeV usingtheATLASdetector,Eur.Phys.J.C76(2016) 581,arXiv:1510.03823 [hep-ex].

[72]ATLASCollaboration,Performanceofb-jetidentiﬁcationintheATLAS Experi-ment,J.Instrum.11(2016)P04008,arXiv:1512.01094 [hep-ex].

[73] ATLASCollaboration,OptimisationoftheATLASb-taggingperformanceforthe 2016 LHCRun, ATL-PHYS-PUB-2016-012, https://cds.cern.ch/record/2160731, 2016.

[74]M.Aaboud,etal.,Performanceofmissingtransversemomentum reconstruc-tionwiththeATLASdetectorusingproton–protoncollisionsat√s=13 TeV, arXiv:1802.08168 [hep-ex],2018.

[75]ATLAS Collaboration, Measurementofthe topquark pair production cross-sectionwithATLASinthesingleleptonchannel,Phys.Lett.B711(2012)244, arXiv:1201.1889 [hep-ex].

[76]ATLASCollaboration,Searchforheavyparticlesdecayingintotop-quarkpairs usinglepton-plus-jetseventsinproton–proton collisionsat√s=13 TeV with theATLASdetector,Eur.Phys.J.C78(2018)565,arXiv:1804.10823 [hep-ex].

[77]N. Kidonakis, Next-to-next-to-leading-order collinearand soft gluon correc-tionsfort-channelsingletopquarkproduction,Phys.Rev.D83(2011)091503, arXiv:1103.2792 [hep-ph].

[78]N.Kidonakis,NNLLresummationfor s-channelsingletopquarkproduction, Phys.Rev.D81(2010)054028,arXiv:1001.5034.

[79]N.Kidonakis,Two-loopsoftanomalousdimensionsforsingletopquark asso-ciatedproductionwithaW− orH−,Phys.Rev.D82(2010)054018,arXiv: 1005.4451.

[80]J.M.Campbell,R.Ellis,MCFMfortheTevatronandtheLHC,Nucl.Phys.B,Proc. Suppl.205–206(2010),arXiv:1007.3492 [hep-ph].

[81]ATLASCollaboration,Luminositydeterminationinpp collisionsat√s=8 TeV usingtheATLASdetectorattheLHC,Eur.Phys.J.C76(2016)653,arXiv:1608. 03953 [hep-ex].

[82]J.Alwall,R.Frederix,S.Frixione,V.Hirschi,F.Maltoni,etal.,Theautomated computationoftree-levelandnext-to-leadingorderdifferentialcrosssections, andtheirmatchingtopartonshowersimulations,arXiv:1405.0301 [hep-ph], 2014.

[83]M.Bahr,etal.,Herwig++physicsandmanual,Eur.Phys.J.C58(2008)639–707, arXiv:0803.0883 [hep-ph].

[84]J.Bellm,etal.,Herwig7.0/Herwig++3.0releasenote,Eur.Phys.J.C76(2016) 196,arXiv:1512.01178 [hep-ph].

[85]S.Frixione,etal.,Single-tophadroproductioninassociationwithaW boson,J. HighEnergyPhys.07(2008)029,arXiv:0805.3067.

[86]J.Butterworth,etal.,PDF4LHCrecommendationsforLHCRunII,J.Phys.G43 (2016)023001,arXiv:1510.03865 [hep-ph].

[87]L.Moneta,K.Cranmer,G.Schott,W.Verkerke,TheRooStatsproject,57,arXiv: 1009.1003 [physics.data-an],2010.

[88]W.Verkerke,D.P.Kirkby,TheRooFittoolkitfordatamodeling,eConfC0303241 (2003),MOLT007,186,arXiv:physics/0306116 [physics],2003.

[89]M.Baak,etal.,HistFittersoftwareframeworkforstatisticaldataanalysis,Eur. Phys.J.C75(2015)153,arXiv:1410.1280 [hep-ex].

[90]A.L.Read,Presentationofsearchresults:theCLstechnique,J.Phys.G28(2002) 2693–2704,11(2002).

[91]G.Cowan,K.Cranmer,E.Gross,O.Vitells,Asymptoticformulaefor likelihood-basedtestsofnewphysics,Eur.Phys.J.C71(2011)1554;Erratum,Eur.Phys. J.C73(2013)2501,arXiv:1007.1727 [physics.data-an].

[92] ATLAS Collaboration, ATLAS computing acknowledgements, ATL-GEN-PUB-2016-002https://cds.cern.ch/record/2202407.

TheATLASCollaboration

M. Aaboud34d,G. Aad99,B. Abbott124, O. Abdinov13,∗,B. Abeloos128,D.K. Abhayasinghe91, S.H. Abidi164, O.S. AbouZeid39,N.L. Abraham153, H. Abramowicz158,H. Abreu157,Y. Abulaiti6,