Measurement of $W^{+}W^{-}$ production in association with one jet in proton-proton collisions at $\sqrt{s}=8$ TeV with the ATLAS detector

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Measurement of W

W

production in association with one jet in proton–proton collisions at √

s = ^{8 TeV with the ATLAS detector}

.TheATLAS Collaboration^

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

Articlehistory:

Received11August2016

Receivedinrevisedform7October2016 Accepted8October2016

Availableonline14October2016 Editor:W.-D.Schlatter

TheproductionofW bosonpairsinassociationwithonejetinpp collisionsat√

s=^{8 TeV isstudied} usingdatacorrespondingtoanintegratedluminosityof20.3 fb⁻¹collectedbytheATLASdetectorduring 2012attheCERNLargeHadronCollider.Thecrosssectionismeasuredinafiducialphase-spaceregion defined by the presence ofexactly one electron and one muon, missing transverse momentum and exactlyonejetwithatransversemomentumabove25 GeVandapseudorapidityof|η_|<4.5.Theleptons are required to have oppositeelectric charge and to pass transverse momentumand pseudorapidity requirements.Thefiducialcrosssectionisfoundtobeσ_{W W}^fid,1-jet₌136±⁶(stat)±¹⁴(syst)±³(lumi) fb.

Incombinationwithapreviousmeasurementrestrictedtoleptonicfinalstateswithnoassociatedjets, the fiducial cross section of W W production with zero or one jet is measured to be σ_{W W}^{fid,≤1-jet}₌

511±⁹(stat)±²⁶(syst)±¹⁰(lumi) fb.Theratiooffiducialcrosssectionsinfinalstateswithoneand zero jetsis determinedto be0.36±⁰.05.Finally, atotalcrosssectionextrapolatedfrom thefiducial measurement ofW W productionwith zero oroneassociated jet isreported. The measurements are comparedtotheoreticalpredictionsandfoundingoodagreement.

©^{2016TheAuthor.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense} (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

1. Introduction

ThemeasurementoftheproductionoftwoW bosonsisacru- cial test of the non-Abelian gauge structure of the electroweak theory of the Standard Model (SM). The increasing precision of the experimental measurements at the LHC has elicited im- proved theoretical descriptions of the process. Progress has been made to extend the next-to-leading-order (NLO) [1] calculation of pp→^{W+W− production to include} next-to-next-to-leading- order(NNLO)effects[2]inperturbativequantumchromodynamics (QCD). A separate calculation of the loop-induced, non-resonant gg→^{W+W− productionprocesshasbeenmadeavailable ator-} derO(α_S³)[3]inthestrongcouplingconstant αS.ResonantW W^∗ production via the exchange of a Higgs boson has been calcu- lated to order O(α_S³) [4] and O(α_S⁴) [5]. These predictions can besummed togive anupdated predictionforthetotalcrosssec- tion of 65.0⁺₋¹₁^._.²₁ pb as further detailed in Section 7. In addition to these newcalculations, fullydifferential NNLO predictions [6]

havebecomeavailable,ashavededicatedNLOpredictions forjet- associated W W production [7,8] with up to three jets [9]. The resummationof logarithms arising froma selection onthe num-

 E-mailaddress:atlas.publications@cern.ch.

berofjetshasbeenpresentedatnext-to-next-to-leading-logarithm (NNLL)accuracyinRefs.[10,11].Itisthereforeinterestingtostudy W W productioninassociationwithjetstoconfrontthesecalcula- tionswithexperimentaldatafromtheLHC.

A measurement of the jet multiplicity in W W events at the CDF experimentwas published inRef. [12].At theLHC, theCMS Collaboration has included W W production in association with one jet in their measurement ofthe total W W production cross section at√

s=8 TeV[13],buthasnot publisheddedicatedfidu- cialcrosssectionsofjet-associatedW W production.

Thisletterpresentsameasurementofthefiducialcrosssection of W W productionusing thedecaychain W⁺W⁻→^e±νeμ^∓ν_μ

infinalstateswithoneassociatedhadronicjet,furtherreferredto as1-jetfinalstate.Thefiducialregionisdefinedusingstableparti- clesatthegeneratorlevelandischosentomatchtheexperimental selectionascloselyaspossible.

Onlyeventswithexactlyonereconstructedjetareselectedfor theanalysis,whileeventswithalargernumberofjetssufferfrom a large background from top-quark production andare not con- sidered. The selected W W candidate event sample is corrected forbackgroundprocesses,detectionefficienciesandresolutionef- fects,andthecrosssectionofW W+^{1-jetproductionisextracted} forthefiducialphase-spaceregion.Theresultsarecombinedwith a previous measurement reported in Ref. [14] restricted to final http://dx.doi.org/10.1016/j.physletb.2016.10.014

0370-2693/©^{2016TheAuthor.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense}(http://creativecommons.org/licenses/by/4.0/).Fundedby SCOAP³.

states without any reconstructed jets, referred to as 0-jet final state. The fiducial W W+≤^{1-jet cross section and the ratio R}1 ofthefiducialW W+^{1-jetandfiducialW W}+^{0-jetcrosssections} aredeterminedandcomparedto differenttheoreticalpredictions.

Themeasurementthereforeextendsthefiducialphasespaceofthe previousmeasurementoftheW W productioncrosssection.

2. DataandMonteCarlosamples

TheATLASdetector[15]isageneral-purposedetectormeasur- ing collisions at the Large Hadron Collider (LHC) with coverage over the full azimuthal angle φ. It consists of an inner detector surrounded by a 2 T solenoid to measure tracks withpseudora- piditiesof|η_|<2.5,¹electromagneticandhadroniccalorimetersto provideenergymeasurementsfor|η|<4.9,andamuonspectrom- eterwithatoroidalmagneticfieldtodetectmuonswith|η|=².7.

Athree-leveltriggersystemselectseventstobereadout.

The measurement uses datacollected withthe ATLAS experi- ment during the 2012 data-taking period. Only runs withstable protonbeamscollidingat√

s=^{8 TeV areusedin whichallrele-} vantdetectorcomponentswerefunctional.Thisdatasamplecorre- spondstoan integratedluminosity of20.3 fb⁻¹ determined with anuncertaintyof±¹.9% andderived frombeam-separationscans performedinNovember2012[16].

The analysis relies on event simulation to correct the mea- sured event yields for experimental effects and for the study of background processes. Different simulated event samples are usedto model the signal fromthe individual production mecha- nisms:qq¯→^{W+W− eventsare simulatedusing}the Powheg 1.0 generator [17–21], which is interfaced to Pythia 8.170 [22];

for the non-resonant gg-induced W W signal the gg2ww pro- gram (version 3.1.3) [23] is employed and interfaced to Her- wig 6.5/Jimmy 4.31 [24,25]; resonant W W^∗ production via a Higgs boson with a mass of mH =^{125 GeV is modelled using} Powheg+Pythia 8.170. The three event samples are simulated usingtheCT10NLO [26] partondistributionfunction (PDF).Pho- tonradiation ismodelledusing Photos [27]. Theparameter tune usedfortheunderlyingeventisAU2[28].Theeventsamplesare normalised to a cross section times branching ratio of 5.58 pb (q^q¯→^{W+W− [1]), 0.153 pb}(non-resonant gg→^{W+W− [23])} and0.435 pb(gg→^H→^{W+W− [4]). Thesumofthesecontri-} butions corresponds to a total W W cross-sectionof 58.7⁺₋⁴₃^._.²₈ pb wheretheuncertainties areduetoscaleandPDF uncertaintiesin the crosssection calculations. For additionalstudies a sample of simulatedqq¯→^{W+W− eventsproduced}with MC@NLO[18] and Jimmy[24,25]usingtheAUET2tune[29]andtheCT10PDFisused.

Production of pairs of top quarks, s-channel single top-quark productionand W -associatedtop-quark productionare modelled with the Powheg+Pythia 6 generator with the AU2 [28] tune.

Single top-quarkproduction inthe t-channel is described by the Acer3.7 [30] MC generatorinterfaced to Pythia 6 [31] withthe AUET2B tune [32]. These events samples are normalised to the respectiveNNLO+NNLL calculations[33–36]to obtain therelative contributiontothetotaltop-quarkbackground,whoseoverallnor- malisationisdeterminedfromdataasdetailedinSection4.

1 ATLASusesaright-handedcoordinatesystemwithitsoriginatthe nominal interactionpoint(IP)inthecentreofthedetectorandthez-axisalongthebeam pipe.Thex-axispointsfromtheIPtothecentreoftheLHCring,andthe y-axis pointsupward.Cylindricalcoordinates(r,φ)areusedinthe transverseplane,φ beingtheazimuthalanglearoundthez-axis.Thepseudorapidityisdefinedinterms ofthepolarangleθas η= −ln tan(θ/2).Thetransverseenergyiscomputedas ET=E·sinθ,whiletheradialdistancebetweentwoobjectsisdefinedas R=



( η)²+ ( φ)^2.

BackgroundfromW andZ bosonproductionismodelledusing Alpgen2.14[37] interfacedto Pythia 6andnormalised toNNLO calculations [38] where needed. The AUET2 tune is used forthe underlyingevent.ThedibosonbackgroundprocessesW Z and Z Z are generatedusingthesamesettings asemployedforthesimu- latedq^q¯→^{W+W− eventsamples.TheproductionofaW boson} andavirtualphoton(γ^∗)isgeneratedusingthe Sherpa generator (version1.4.2)[39].ForWγ production Alpgen+Herwig+Jimmy is employed.

In all simulated event samples, additional pp collisions ac- companying the hard-scatter interactions (pile-up) are modelled by overlaying minimum-biaseventsgenerated using Pythia 8.To simulate the detector response, the generated events are passed throughadetailedsimulationoftheATLASdetector[40]basedon Geant4[41]or Geant4combinedwithaparameterisedcalorime- tersimulation[42].

3. Objectreconstructionandeventselection

Events are selected using reconstructed jets, electrons, muons and missingtransverse momentum. The selection follows closely theoneinRef.[14]tofacilitatethecombinationwiththeW W+ 0-jet final state. Electrons and muons are identified based on tracks in the inner detector matched either to energy deposits in the electromagnetic calorimeter or combined with tracks in the muon spectrometer, respectively. Electrons are reconstructed within |η_{| <}2.47 excluding the transitionregion between barrel andendcapcalorimetersof1.37<|η| <1.52.Muonsarerequired toliewithin |η| <².4.Thesamereconstructionandidentification requirementsasin Ref. [14]are used, resultinginan eventsam- plewithminimalcontributionsfrombackgroundsduetoparticles misidentified asleptons, particularly from W+^{jets, multijetand} Wγ events.FortheselectionofW W candidate events,thepres- enceofexactlytwoisolated,oppositelychargedleptons(,^)with transversemomentaof p_T>25 GeV andp_T^>20 GeV isrequired.

Onlyfinalstateswithoneelectronandonemuonareused.Events withadditionalleptonswithpT>7 GeV arerejected,whichhelps tosuppressother dibosonprocesseswithmorethan twoleptons.

It is requiredthat atleast one ofthe leptons has met an online single-leptonselectionorbothhavepassedadileptontriggerwith reducedthresholdsandlessstringentobjectidentificationcriteria.

Thissetup has an efficiencyof99%–100% withrespect tothe of- flineleptonselection.

Jets are formed using calibrated topological clusters of en- ergy [43] reconstructed in the calorimeters using the anti-kt al- gorithm [44] with radius parameter R=⁰.4. Further corrections to the jet energy are applied based on simulation [45] and are followed by a pile-upsuppression [46].Jets are requiredto have pT>25 GeV and |η_{| <}4.5.More than 50% of the scalarsum of the p_T of all tracks contained within R=⁰.4 of the jet axis is required to be from tracks associated with the primary ver- tex to suppress contributions from additional pp interactions in theevent[47]ifthejetsatisfies p_T<50 GeV and|η| <².4.Only eventswithexactlyonejetmeetingtheabovecriteriaareselected.

Jets containing b-hadrons (so-called b-jets) are identified within the central region ofthe detector, |η| <².5, usinga multivariate approach [48,49] withan efficiency of 85%. To reduce the back- groundfrom top-quarkproduction, eventscontaining b-jets with p_T>20 GeV andwithin|η| <².5 arerejected.

Selectionrequirementsonthemissingtransversemomentumin thecandidateeventsareusedtoreducethecontributionofevents fromZ/γ^∗→τ τ (Drell–Yan)productionwhereboth τ-leptonsde- cay leptonically. Missing transverse momentum is reconstructed fromthevectorsumofthetransversemomentaofidentifiedparti- cles[50]towhicheitherreconstructedjetsandcalorimetricdepo-

Fig. 1. (a)Distributionsofthetransversemomentumoftheselectedjetinthecontrolregionenrichedineventsfromtop-quarkproduction.Thesuminquadratureof statistical,experimentalandtheoreticaluncertaintiesintheMCpredictionareshownasahatchedband.(b)Distributionsofthetransversemomentumoftheselectedjet afterfinaleventselection.DataareshowntogetherwiththeyieldsfromW W signalasestimatedfromsimulatedeventsampleswhicharescaledtoatotalcrosssection of58.7⁺₋⁴₃^._.²₈pb,andtheestimatedbackgroundcontributions.Thesuminquadratureofstatistical,experimentalandtheoreticaluncertaintiesisshownasahatchedband.

In bothfiguresthelastbinofthedistributionisanoverflowbin.

sitions notassociatedwithanyparticle areadded.Missingtrans- verse momentum induced by mismeasurements ofthe energy of leptons isfurther reducedinthe calorimeter-basedmeasurement byprojectingthemissingtransversemomentum E^miss_T ontonearby leptons,to calculatetheso-calledrelativemissingtransversemo- mentum E^miss_T,Rel. A lepton is considered nearby if the azimuthal separation to the E^miss_T direction is small, φ (E^miss_T ,)<π/2, andonlyin thiscase, E^miss_T ismodified to yield E^miss_T,Rel=^Emiss_T × sin( φ (E^miss_T ,)), otherwise E^miss_T,Rel= ^EmissT . The relative missing transversemomentum isrequired tobe E^miss_T,Rel>15 GeV. Anad- ditionaltrack-basedmeasureofthemissingtransversemomentum (p^miss_T )isconstructedbyaddingthemomentaoftracksassociated withtheprimary vertextothevector sumofthetransverse mo- mentaofidentifiedelectronsandmuons.Byconstruction, p^miss_T is lesssensitive toenergy depositsfromadditional interactions and itisrequiredtobe p^miss_T >20 GeV.Tofurtherreducethesensitiv- itytofluctuationsineitherofthe missingtransversemomentum variablesused,the azimuthalseparation betweenE^miss_T and p^miss_T mustsatisfy φ (E^miss_T ,p^miss_T )<2.0.

Theinvariantmassofthetwoselectedleptons,m,isrequired tobegreaterthan10 GeVtosuppresscontributionsfrommisiden- tifiedleptonsproducedinmultijetandW+jets events.Apartfrom therequirementsonthejetsand φ (E_T^miss,p^miss_T ), thiseventselec- tionisidenticaltotheoneemployedinRef.[14].

4. Determinationofbackgrounds

The experimental signature of exactly one electron and one muon with opposite electric charge, and missing transverse mo- mentum can be produced by a variety of SM processes which are treated as backgrounds. Top quarks decay almost exclusively toa b-quarkanda W boson.This makest¯t andsingle top-quark productionthedominantbackgroundto W W production, inpar- ticularforeventswithjetsinthefinalstate.Thebackgroundyield fromtop-quarkproductionisdeterminedusingamethodproposed in Ref. [51]. The eventyield is extrapolatedfrom a control sam- pleenrichedineventsfromtop-quarkproduction.Itisdefinedby thenominalselection requirementsbutmustcontain exactlyone identifiedb-jetwith pT>25 GeV andwithin|η_|<2.5,insteadof requiringthe absence ofidentified b-jets. The distribution ofthe transversemomentumoftheb-jetinthecontrolsampleisshown

in Fig. 1(a). The datais usedto constrain the largeexperimental and theoretical uncertainties shownby the error bands.The fac- tor to extrapolate fromthis control sample to the signal sample is determined as theratio of jetspassing or failingthe b-jet re- quirementin additionalcontrol samples,definedbythe presence of two jets, at leastone of which passes theb-tag requirement.

Systematiceffectsresultingfromthechoiceofthecontrolsample arecorrectedforbyanadditionalfactorestimatedfromsimulated eventsamples.Thecorrection introduces experimentalsystematic uncertaintiesof±³.1%,mainlyfromtheuncertaintyinthejeten- ergyscale.Theoreticaluncertaintiesarefoundtoamountto±².5%

and are dominated by differences in simulatedtt event samples producedwith Powheg and MC@NLO,anduncertaintiesintheW t productioncrosssection.Statistical uncertaintiesfromthelimited size of the control samples in data and simulation introduce an uncertaintyof±³.5%,resultinginanoverallprecision intheesti- matedtop-quarkbackgroundyieldof±⁵.2%.

The estimationof theremaining background processesclosely follows the methodologydescribed inRef. [14].Data-driven esti- mates of the yields of W+^{jets and multijet productionare de-} terminedinaneventsampleindatathatisselectedwithrelaxed identification andisolation criteriaforthe leptons. The composi- tion ofthiseventsample withgenuine andmisidentifiedleptons canbeinferredusingtheprobabilitiesofgenuineandmisidentified leptons selected withthe relaxed criteria to satisfy the nominal lepton selection criteria. The yield ofbackground fromDrell–Yan productionisobtainedfroma simultaneousfitofthedistribution ofsimulatedeventsamplestothe φ (E^miss_T ,p^miss_T )distributionof the data in the signal region and in a control sample, defined by a selection of 5 GeV<p^miss_T <20 GeV and no selection on φ (E^miss_T ,p^miss_T ). The yields of the diboson processes, W Z , Z Z andWγ production,aredeterminedusingsimulationandarenor- malised toNLO predictions [1].Theuncertainties assignedto the NLOpredictionsareinflatedtocoverdifferencesfromthecalcula- tions inRefs.[52,53].ForWγ productiona K -factoriscalculated fromRef.[54]andappliedtotheNLOprediction.

The observed data and the estimated signal and background yields are summarised in Table 1. Half of the events selected in data are estimated to originate from background processes, where top-quark production represents the largest contribution.

Thetransversemomentumdistributionoftheselectedjetafterthe finaleventselectionisshowninFig. 1(b),wheredataisshownto-