Measurement of electroweak WZ boson production and search for new physics in WZ + two jets events in pp collisions at s=13TeV

Contents lists available atScienceDirect

Physics

Letters

B

www.elsevier.com/locate/physletb

Measurement

of

electroweak

WZ boson

production

and

search

for

new

physics

in

WZ

+

two

jets

events

in

pp collisions

at

√

s

=

13 TeV

.

The

CMS

Collaboration



CERN,Switzerland

a

r

t

i

c

l

e

i

n

f

o

a

b

s

t

r

a

c

t

Articlehistory:

Received13January2019

Receivedinrevisedform30April2019 Accepted27May2019

Availableonline31May2019 Editor:M.Doser Keywords: CMS Physics SM WZ VBS

A measurement of WZ electroweak (EW) vector boson scattering is presented. The measurement is performed inthe leptonic decaymodes WZ→ 

ν

, where ,=e,

μ

.The analysis is based ona datasampleofproton-protoncollisionsat√s=13 TeVattheLHCcollectedwiththeCMSdetectorand correspondingtoanintegratedluminosityof35.9 fb−1_{.TheWZ plustwojetproductioncrosssectionis}

measuredinfiducialregionswithenhancedcontributionsfromEWproductionandfoundtobeconsistent withstandardmodelpredictions.TheEW WZ productioninassociationwithtwojetsismeasuredwith anobserved(expected)significanceof2.2(2.5)standarddeviations.ConstraintsonchargedHiggsboson productionandonanomalousquarticgaugecouplingsintermsofdimension-eighteffectivefieldtheory operatorsarealsopresented.

©2019TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP3_.

1. Introduction

Thediscoveryofascalarbosonwithcouplingsconsistentwith thoseofthe standard model(SM) Higgsboson (H)by the ATLAS andCMSCollaborations [1–3] at theCERNLHCprovidesevidence thattheW andZ bosonsacquiremassthroughthe Brout-Englert-Higgs mechanism [4–9]. However, current measurements of the Higgs boson couplings [10,11] do not preclude the existence of scalarisospin doublets, triplets, or higherisospin representations alongsidethesingle isospindoubletfield responsibleforbreaking theelectroweak(EW)symmetryintheSM [12,13].Inadditionto theircouplingstotheHiggsboson,thenon-Abeliannatureofthe EWsectoroftheSMleadstoquarticandtripleself-interactionsof themassivevectorbosons.PhysicsbeyondtheSMintheEW sec-toris expectedtoinclude interactions withthevector andHiggs bosons that modify their effective couplings. Characterizing the self-interactionsofthevectorbosonsisthusofgreatimportance.

The total WZ production cross section in proton-proton (pp) collisionshasbeenmeasured intheleptonic decaymodesbythe ATLAS and CMS Collaborations at 7, 8,and 13 TeV [14–18], and limitsonanomalous triple gaugecouplings [19] are presentedin Refs. [15,17,20].Constraintsonanomalousquarticgaugecouplings (aQGC) [21] are presented by the ATLAS Collaboration at 8 TeV in Ref. [15]. At the LHC, quartic WZ interactions are accessible

 _{E-mailaddress:cms-publication-committee-chair@cern.ch.}

through triple vector boson productionor via vector boson scat-tering(VBS),wherevector bosonsareradiatedfromtheincoming quarksbefore interacting,asillustrated inFig.1 (upperleft).The VBS processes form a distinct experimental signature character-ized by theW and Z bosonswithtwo forward,high-momentum jets, arising fromthe hadronization oftwo quarks. Theyare part ofanimportantsubclassofprocessescontributingtoWZ plustwo jet(WZjj)productionthatproceedsviatheEWinteractionattree level,

O(

α

4

₎

_{,referred toasEW-inducedWZjj production, or}

sim-ply EW WZ production. An additional contribution to the WZjj state proceeds via quantum chromodynamics (QCD) radiation of partonsfromanincomingquarkorgluon,showninFig.1(upper right),leadingtotree-levelcontributionsat

O(

α

2

_α

2

S

)

.Thisclassof

processesisreferred toasQCD-inducedWZjj production(orQCD WZ).

The first study of EW WZ production at the LHC was per-formedbytheATLAS Collaborationat8 TeV[15].Ameasurement at 13 TeV with an observed statistical significance for the EW WZ process greaterthan 5standard deviationshas recentlybeen reported and submitted for publication by the ATLAS Collabora-tion [22]. This letter reports searches for EW WZ production in theSM andfornewphysics modifyingtheWWZZ couplinginpp collisionsat

√

s

=

13 TeV.TwofiducialWZjj crosssectionsare pre-sented,bothinphasespaceswithenhancedcontributionsfromthe EW WZ process.Thedatasamplecorrespondstoanintegrated lu-minosityof35.9 fb−1 collectedwiththeCMSdetector [23] atthe CERN LHCin2016. Theanalysisselectseventswithexactly three

https://doi.org/10.1016/j.physletb.2019.05.042

0370-2693/©2019TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).Fundedby SCOAP3_.

Fig. 1. RepresentativeFeynmandiagramsforWZjj productionintheSMandbeyond theSM.TheEW-inducedcomponentofWZ productionincludesquarticinteractions (upperleft) ofthe vectorbosons.ThisisdistinguishablefromQCD-induced pro-duction(upperright)throughkinematicvariables.NewphysicsintheEWsector modifyingthequarticcouplingcanbeparameterizedintermsofdimension-eight effectivefieldtheoryoperators(lowerleft).Specificmodelsmodifyingthis interac-tionincludethosepredictingchargedHiggsbosons(lowerright).

leptons(electronsormuons),missingtransversemomentumpmiss T ,

andtwo jetsat high pseudorapidity

η

with a large dijet system invariant massmjj,characteristicofVBS processes.The kinematic

variablesof thetwo forwardandhighmomentum jets, including

η

separationandmjj,are usedtoidentifytheEW WZ component

of WZjj production. An excess of events with respect to the SM predictioncould indicatecontributions fromadditionalgauge bo-sonorvectorresonances [24],chargedscalarorHiggsbosons [25], or it could suggest that the gauge or Higgs bosons are not el-ementary [26]. We study such deviations in terms of aQGCs in thegeneralized frameworkof dimension-eighteffectivefield the-ory operators, Fig. 1 (lower left),and in terms of charged Higgs bosons,Fig. 1(lowerright),andwe placelimitsontheir produc-tioncrosssectionsandoperatorcouplings.

2. TheCMSdetector

Thecentralfeature oftheCMSapparatusisasuperconducting solenoid of 6 m internal diameter, providing a magnetic field of 3.8 T.Withinthesolenoidvolumearesiliconpixelandstrip track-ingdetectors,alead tungstatecrystalelectromagneticcalorimeter (ECAL), and a brass and scintillator hadron calorimeter (HCAL), each composed of a barrel and two endcap sections. Forward calorimeters extend the

η

coverage provided by the barrel and endcap detectors up to

|

η

|

<

5. Muons are measured in gas-ionization detectors embedded in the steelflux-return yoke out-sidethesolenoid.

Events of interest are selected using a two-level trigger sys-tem [27]. The first levelof the CMStrigger system, composed of customhardware processors,usesinformationfromthe calorime-tersandmuondetectorstoselecteventsofinterestinafixedtime intervalof3.2

μ

s.Thehigh-leveltriggerprocessorfarmfurther de-creases theevent ratefrom around 100 kHz to lessthan 1 kHz, beforedatastorage [27].

AmoredetaileddescriptionoftheCMSdetector,togetherwith adefinitionofthecoordinatesystemused andthe relevant kine-maticvariables,canbefoundinRef. [23].

3. Signalandbackgroundsimulation

SeveralMonteCarlo(MC)eventgeneratorsareusedtosimulate thesignalandbackgroundprocesses.

The EW-inducedproductionofWZ boson pairsandtwo final-state quarks,Fig.1(upperleft),wheretheW andZ bosonsdecay leptonically,issimulatedatleadingorder(LO)inperturbativeQCD using MadGraph5_amc@nlo v2.4.2 [28]. The MC simulation in-cludes all contributions to the three-lepton final state at

O(

α

6

₎

_,

with thecondition that the massof W bosonbe within 30 GeV of its on-shell value from Ref. [29]. The resonant W boson is decayed using MadSpin [30]. Triboson processes, where the WZ boson pair is accompanied by a third vector boson that decays intojets, areincluded intheMCsimulation,butaccount forwell below 1% of the event yield for the selectionsdescribed in Sec-tion 5. Contributions with an initial-state b quark are excluded fromthisMCsimulationsincetheyareconsideredpartofthetZq background process. The predictions from MadGraph5_amc@nlo are cross-checked withLOpredictions from theevent generators Vbfnlo 3.0 [31] and sherpa v2.2.4 [32,33], and with fixed-order calculationsfrom MoCaNLO+Recola [34,35].Agreementisobtained whenusingequivalentconfigurationsofinputparameters, includ-ingcouplings,particlemassesandwidths,andthechoiceof renor-malization(

μ

R)andfactorizationscales(

μ

F).

Several MC simulations ofthe QCD WZ process, Fig.1 (upper right),areconsidered.Thesimulationsareinclusiveinthenumber ofjetsassociatedwiththe leptonicallydecayingW and Z bosons, andthereforecomprisethefullWZjj state.TheprimaryMC simu-lationis simulatedatLOwith MadGraph5_amc@nlo v2.4.2,with contributionstoWZ productionwithuptothreeoutgoingpartons includedinthematrixelementcalculation.Thedifferentjet multi-plicitiesaremergedusingtheMLMscheme [36].Anext-to-leading order (NLO) MC simulation from MadGraph5_amc@nlo v2.3.3 with zero or one outgoing partons at Born level, merged us-ing the FxFx scheme [37], andan inclusive NLO simulationfrom powheg_{2.0[38–41] arealsoutilized.TheLOMCsimulationwith} MLMmerging,referredtoastheMLM-mergedsimulation,isused asthecentralpredictionfortheanalysisbecauseofitsinclusionof WZ plus three-partoncontributionsat treelevel, whichare rele-vanttoWZjj production.TheotherMCsimulations,usedtoassess the modeling uncertainty in the QCD WZ process, are referred to asthe FxFx-merged andthe powheg simulations, respectively. Each MC simulationis normalizedtothe NLO crosssection from powheg2.0.

In addition to the EW WZ and QCD WZ processes, which at tree level are

O(

α

4

₎

_and

_O(

_α

2

_α

2

S

)

respectively, a smaller

con-tribution at

O(

α

3

_α

S

)

contributes to the WZjj state. We refer

to thiscontribution asthe interference term. It is evaluated us-ing MC simulations ofparticle-level events generatedwith Mad-Graph5_amc@nlo v2.6.0. The process is simulated with the dy-namic

μ

R and

μ

F set to the maximum outgoing quark pT per

event, and with fixed scales

μ

R

=

μ

F

=

mW, where mW is the

worldaveragevalueoftheW bosonmass,takenfromRef. [29]. TheassociatedproductionofaZ bosonandasingletopquark, referred to as tZq production, is simulated at NLO in the four-flavor scheme using MadGraph5_amc@nlo v2.3.3. The MC simu-lation isnormalizedusinga crosssection computedatNLO with MadGraph5_amc@nlo inthefive-flavorscheme,followingthe pro-cedure ofRef. [42].The productionofZ boson pairsvia qq anni-hilationisgeneratedatNLOinperturbativeQCDwith powheg 2.0 whilethegg

→

ZZ processissimulatedatLOwith mcfm 7.0 [43]. The ZZ simulationsarenormalizedtothecrosssection calculated at next-to-next-to-leading order for qq

→

ZZ with MATRIX [44,

45] (K factor 1.1) and at NLO for gg

→

ZZ [46] (K factor 1.7). The EW production of Z boson pairs and two final-state quarks, where the Z bosons decay leptonically, is simulatedat LO using MadGraph5_amc@nlo v2.3.3.BackgroundfromZ

γ

,ttV (ttW,ttZ), andtribosoneventsVVV (WWZ, WZZ,ZZZ)aregeneratedatNLO

with MadGraph5_amc@nlo v2.3.3,withthevector bosons gener-atedon-shellanddecayedvia MadSpin.

The simulation of the aQGC processes is performed at LO using MadGraph5_amc@nlo v2.4.2 and employs matrix element reweightingto obtainafinely spacedgridofparameters foreach oftheanomalous couplingsoperatorsprobedbytheanalysis. The configuration of input parameters is equivalent to that used for the EW WZ simulation described previously. The production of charged Higgs bosons in the Georgi–Machacek (GM) model [47] is simulated at LO using MadGraph5_amc@nlo v2.3.3 and nor-malizedusingthenext-to-next-to-leadingordercrosssections re-portedinRef. [48].

The pythia v8.212 [49,50] packageis usedforparton shower-ing,hadronization,andunderlyingeventsimulation,with parame-terssetbytheCUETP8M1tune [51] forallsimulatedsamples.For theEW WZ process,comparisons aremadeatparticle-level with the parton shower and hadronization of sherpa and with her-wigv7.1 [52,53].Forall MCsimulations usedinthisanalysis, the NNPDF3.0 [54] set ofpartondistributionfunctions(PDFs)isused, withPDFscalculatedtothesameorderinperturbativeQCDasthe hardscatteringprocess.

Thedetectorresponseissimulatedusingadetaileddescription oftheCMSdetectorimplementedinthe Geant4 package [55,56]. Thesimulatedeventsarereconstructedusingthesamealgorithms usedforthedata.Thesimulatedsamplesincludeadditional inter-actionsin thesameandneighboring bunchcrossings, referred to aspileup. Simulated eventsare weighted so the pileup distribu-tionreproducesthatobservedinthedata,whichhasanaverageof about23interactionsperbunchcrossing.

4. Eventreconstruction

Inthisanalysis,theparticle-flow(PF)eventreconstruction algo-rithm [57] isused.ThePFalgorithmaimstoreconstructand iden-tifyeachindividualparticleasaphysicsobjectinanevent,withan optimizedcombinationof informationfromthe various elements oftheCMSdetector. Theenergyof photonsisobtainedfromthe ECALmeasurement.Theenergyofelectronsisdeterminedfroma combinationoftheelectronmomentumattheprimaryinteraction vertexasdeterminedbythetracker,theenergyofthe correspond-ingECAL cluster, andtheenergy sumofall bremsstrahlung pho-tonsspatiallycompatiblewithoriginatingfromtheelectrontrack. Theenergyofmuonsisobtainedfromthecurvatureofthe corre-spondingtrack.Theenergyofchargedhadronsisdeterminedfrom a combination of their momentum measured in the tracker and thematchingECALandHCALenergydeposits,correctedfor zero-suppressioneffectsandfortheresponsefunctionofthe calorime-terstohadronicshowers.Finally,theenergyofneutralhadronsis obtainedfromthe corresponding correctedECAL andHCAL ener-gies.

The reconstructed vertex with the largest value of summed physics-object p2_T (where pT is thetransverse momentum)isthe

primary pp interaction vertex. The physics objects are the jets, clusteredusingajetfinding algorithm [58,59] withthetracks as-signedto thevertexasinputs,andthe associated pmiss_T , takenas thenegativevectorsumofthe pj_Tofthosejets.

Electronsare reconstructed within the geometricalacceptance

|

η

e

_|

_<

₂

_.

_{5.Thereconstructioncombinestheinformationfrom}

clus-ters of energy deposits in the ECAL and the trajectory in the tracker [60].Toreducetheelectronmisidentificationrate,electron candidatesaresubjectedtoadditionalidentificationcriteriabased onthedistributionoftheelectromagneticshowerintheECAL,the relative amount ofenergy deposited in the HCAL,a matchingof thetrajectoryofanelectrontrackwiththeclusterintheECAL,and itsconsistencywithoriginatingfromtheselectedprimary vertex.

Candidatesthat are identifiedasoriginatingfrom photon conver-sionsinthedetectormaterialareremoved.

Muons are reconstructed within

|

η

μ

_|

_<

₂

_.

_{4 [61]. The} recon-structioncombinestheinformationfromboththetrackerandthe muon spectrometer. The muonsare selectedfrom amongthe re-constructed muon track candidates by applying minimal quality requirements on the track components in the muon system and byensuringthatmuonsareassociatedwithsmallenergydeposits inthecalorimeters.

Foreach lepton track, the distanceof closest approach to the primary vertexinthetransverseplane isrequiredtobelessthan 0.05 (0.10) cm for electrons in the barrel (endcap) region and 0.02 cmformuons. Thedistancealongthebeamlinemustbeless than 0.1(0.2) cmforelectronsin thebarrel(endcap)and0.1 cm formuons.

Jets are reconstructed using PF objects. The anti-kT jet

clus-tering algorithm [58] witha distanceparameter R

=

0

.

4 is used. Toexcludeelectrons andmuonsfromthejet sample,thejetsare required to be separated from the identified leptons by



R

=



(

η

)

2

_{+ (φ)}

2

_>

₀

_.

_{4, where}

_φ

_{is the azimuthal angle in}

ra-dians. The CMS standard method for jet energy corrections [62] is applied. These include corrections to the pileup contribution that keep the jet energy correction and the corresponding un-certainty almost independent of the number of pileup interac-tions.Inordertoreject jetscomingfrompileupcollisions(pileup jets), a multivariate-based jet identification algorithm [63] is ap-plied.This algorithm takesadvantage of differencesin theshape ofenergydepositsinajetconebetweenjetsfromhard-scattering andfrom pileupinteractions. The jets are requiredto have pj_T

>

30 GeV and

|

η

j

_|

_<

₄

_.

_{7. We identify potential top quark}

back-grounds by identifying the b quark produced in its decay via thecombinedsecondaryvertexb-taggingalgorithmwiththetight working point [64]. The efficiency for selecting b quark jets is

≈

49%witha misidentificationprobability of

≈

4% forc quarkjets and

≈

0.1%forlight-quarkandgluonjets.

The isolation ofindividual electronsormuonsis defined rela-tive totheir p

T by summingoverthe pT ofchargedhadronsand

neutralparticleswithinaconewithradius



R

<

0

.

3

(

0

.

4

)

around theelectron(muon)directionattheinteractionvertex:

I

=





pcharged_T

+

max



0

,



pneutral_T

+



pγ_T

−

pPU_T



p_T

.

Here,

pcharged_T is the scalar pT sum of charged hadrons

orig-inating from the primary vertex. The

pneutral_T and

pγ_T are thescalar pT sumsforneutralhadronsandphotons,respectively.

The neutralcontribution totheisolation frompileupevents, pPU_T , is estimated differently for electrons and muons. For electrons, pPU

T

≡

ρ

Aeff,wherethe averagetransverse momentumflow

den-sity

ρ

iscalculatedineacheventusingthe“jetarea”method [65], which defines

ρ

as the median of the ratio of the jet trans-verse momentumto thejet area, pj_T

/

Aj,forall pileupjetsin the

event. The effective area Aeff is the geometric area of the

isola-tionconetimesan

η

-dependentcorrectionfactorthataccountsfor theresidualdependenceoftheisolationonthepileup.Formuons, p_TPU

≡

0

.

5

_ip_TPU,i, where i runs over the charged hadrons orig-inating from pileup vertices and the factor 0.5 corrects for the ratio ofcharged to neutral particlecontributions in the isolation cone. Electrons are considered isolated if Ie

_<

₀

_.

₀₃₆

_,

₍

₀

_.

₀₉₄

₎

_for

thebarrel(endcap)region,whereasmuonsareconsideredisolated if Iμ

<

0

.

15,where thevaluesare optimizedforaggressive back-groundrejection while maintaininga reconstruction efficiency of

≈

70%.Relaxedidentificationcriteriaare definedby Iμ

<

0

.

40 for muons andby relaxed trackquality anddetector-based isolation

forelectrons. The overall efficiencies of the reconstruction, iden-tification, and isolation requirements for the prompt e or

μ

are measured in data andsimulation in bins of p

T and

|

η



|

using a

“tag-and-probe”technique [66] appliedto an inclusivesample of Z events.Thedatatosimulationefficiencyratiosareusedasscale factorstocorrectthesimulatedeventyields.

5. Eventselection

Collisioneventsare selectedby triggers that requirethe pres-enceofoneortwo electrons ormuons.The p_T threshold forthe single leptontrigger is25(20) GeV for theelectron (muon) trig-ger.For the dilepton triggers, withthe same ordifferent flavors, theminimump_Toftheleadingandsubleadingleptonsare17(17) and12(8) GeV forelectrons (muons),respectively.The combina-tionofthesetriggerpathsbringsthetriggerefficiencyforselected three-leptoneventstonearly 100%.Partialmistimingofsignalsin theforwardregionoftheelectromagneticcalorimeter(ECAL) end-caps(2

.

5

<

|

η

|

<

3

.

0)ledtoearlyreadoutforasignificantfraction ofeventswithforwardjetactivity,andacorrespondingreduction inthelevel 1triggerefficiency. Acorrection forthiseffectis de-terminedinbinsofjet pj_T and

η

j _{usinganunbiaseddatasample.}

Thislossofefficiencyisabout1%formjjof200 GeV,increasingto

about15%formjj

>

2 TeV.

A selected event is required to have three lepton candidates







,where

,





=

e,

μ

.All leptons must pass the identification and isolation requirements described in Section 4. The electrons and muons can be directly produced froma W or Z boson de-cayorfromaW orZ bosonwithanintermediate

τ

leptondecay. The







pairconsistsoftwoleptonswithoppositechargeandthe sameflavor,asexpectedforaZ bosoncandidate.Oneofthe lep-tonsfromtheZ bosoncandidateisrequiredtohave p1

T

>

25 GeV

and the other p2

T

>

15 GeV. For events with three same-flavor

leptons,twooppositelycharged,same-flavorcombinationsare pos-sible. Thepairwithinvariant massclosest tomZ

=

91

.

2 GeV,the

nominalZ bosonmass fromRef. [29], isselected asthe Z boson candidate.The remaining lepton isassociated with the W boson andmusthave p

T

>

20 GeV.Eventscontaining additionalleptons

satisfyingtherelaxedidentificationcriteriawithp_T

>

10 GeV are rejected.Becauseoftheneutrinointhefinalstate,theeventsare requiredto have pmiss_T

>

30 GeV.Toreduce contributionsfromtt events,theleptonsconstitutingtheZ bosoncandidatearerequired to have an invariant mass satisfying

|

m

−

mZ

|

<

15 GeV and

eventswith a b taggedjet with pb

T

>

30 GeV and

|

η

b

|

<

2

.

4 are

vetoed.

The invariant mass of any dilepton pair m must be greater

than4 GeV.Such arequirementisnecessaryintheoretical calcu-lationstoavoiddivergencesfromcollinearemissionofsame-flavor opposite-sign dilepton pairs, and 4 GeV is chosen to avoid low mass resonances. The selection is extended to all dilepton pairs toreducecontributionsfrombackgroundswithsoftleptonswhile havinganegligibleeffectonsignal efficiency.Thetrilepton invari-antmass,m3,isrequiredtobemorethan 100 GeV toexcludea

regionwhereproductionofZ bosonswithfinal-statephoton radi-ationisexpectedtocontribute.

Furthermore,theeventmusthaveatleasttwo jetswith pj_T

>

50 GeV and

|

η

j

_|

_<

₄

_.

_{7.The jet withthe highest p}j

T is calledthe

leading jet and the jet with the second-highest pj_T the sublead-ing jet.Toexploit theunique signature oftheVBS process, these two jets are required to have mjj

>

500 GeV and

η

separation

|

η

(

j1

,

j2

)

|

≡ |

η

jj

|

>

2

.

5. The variable

η

3∗

=

η

3

_{− (}

_η

j1

+

_η

j2

_)/

₂

ofthethree-leptonsystemisadditionallyrequiredtobe between

−

2

.

5 and2.5.Thisselectionisreferredtoasthe“EWsignal selec-tion.”Thesamesetofselections,butwithnorequirementon

η

∗_3

Table 1

Summaryofeventselectionsandfiducialregiondefinitionsfor theanalysis.The selectionslabeled“EW signal”and“Higgsboson”areappliedtodataand recon-structedsimulatedevents.TheEWsignalselection isusedfor allmeasurements exceptfor thecharged Higgsboson search thatuses the selection indicatedin thecolumnlabeled“Higgsboson.”TheWZjj crosssectionisreportedinthe fidu-cial regionsdefined bythe selectionsspecifiedinthe lasttwocolumns applied toparticle-levelsimulatedevents.Thevariablesnjandnbrefertothenumberof

anti-kT jetsandthenumberofanti-kT b-taggedjets,respectively.Othervariables

aredefinedinthetext.

EW signal Higgs boson Tight fiducial Loose fiducial p1 T [GeV] >25 >25 >25 >20 p2 T [GeV] >15 >15 >15 >20 p T[GeV] >20 >20 >20 >20 |ημ_{| <2.4 <2.4 <2.5 <2.5} |ηe_| <2.5 <2.5 <2.5 <2.5 |m−mZ|[GeV] <15 <15 <15 <15 m3[GeV] >100 >100 >100 >100 m[GeV] >4 >4 >4 >4 pmissT [GeV] >30 >30 — — |ηj_{| <4.7 <4.7 <4.7 <4.7} pjT[GeV] >50 >30 >50 >30 |R(j, )| >0.4 >0.4 >0.4 >0.4 nj ≥2 ≥2 ≥2 ≥2 pb T[GeV] >30 >30 — — |ηb_| <2.4 <2.4 — — nb =0 =0 — — mjj >500 >500 >500 >500 |ηjj| >2.5 >2.5 >2.5 >2.5 |η3_{− (η}j1+_ηj2_)/2| _{<2.5 — <2.5 —}

andwiththerelaxedrequirementp_Tj

>

30 GeV,isusedinsearches forchargedHiggsbosonsandthereforecalledthe“Higgsboson se-lection.”AsummaryoftheseselectionsisshowninTable1.

Sideband regions of events with a similar topology to sig-nal events, but outside the signal region, are used to constrain the normalization of the QCD WZ process in the EW WZ mea-surement and in searches for new physics. We refer to this re-gion asthe “QCDWZ sideband region.”It consistsofeventswith mjj

>

100 GeV satisfyingallrequirementsappliedtosignalevents,

butfailingat leastone ofthe signaldiscriminating variables, i.e., mjj

<

500 GeV or

|

η

jj

|

<

2

.

5. For the EW WZ measurement,

events satisfying

|

η

∗_3

|

>

2

.

5 arealso selectedinthe sideband re-gion.

Toreducethedependenceontheoreticalpredictions, measure-mentsarereportedintwofiducialregions,definedinTable1.The “tight fiducialregion” isdefinedto beascloseaspossibleto the measurement phase space, whereas the“loose fiducial region” is designedtobe easilyreproducibleintheoreticalcalculationsorin MC simulations, following the procedure of Ref. [34]. The fidu-cial predictions are defined through selections on particle-level simulatedeventsusingthe Rivet [67] framework,whichprovides a toolkit foranalyzing simulated events in a model-independent way.Electronsandmuonsarerequiredtobeprompt(i.e.,notfrom hadron decays), and those produced in the decay of a

τ

lepton are not considered in the definition of the fiducial phase space. The momentaofprompt photonslocatedwithin acone ofradius



R

=

0

.

1 areaddedtotheleptonmomentumtocorrectfor final-statephotonradiation,referredtoas“dressing.”Thethreehighest pT leptons areselected andassociatedwiththe W andZ bosons

withthe same procedureused inthe dataselection. The fiducial cross section in the QCD WZ sideband region is defined follow-ing the tight fiducial region ofTable 1, with mjj

>

100 GeV and mjj

<

500 GeV or

|

η

jj

|

<

2

.

5 or

|

η

∗3

|

>

2

.

5. Theoretical

predic-tions are evaluated using MadGraph5_amc@nlo at LOinterfaced to pythia withthesamplesdescribedinSection3.

6. Backgroundestimation

Backgroundcontributionsin thisanalysisaredivided intotwo categories: background processes with prompt isolated leptons, e.g.,ZZ, tZq, ttZ;andbackgroundprocesses withnonprompt lep-tonsfromhadronsdecayingtoleptons insidejetsorjets misiden-tifiedasisolatedleptons, primarilytt and Z+jets.Thebackground processes with prompt leptons are estimated from MC simula-tion,whereasbackgroundswithnonpromptleptonsfromhadronic activityare estimatedfromdata usingcontrol samples.The non-promptcomponentoftheZ

γ

process,inwhichthephoton experi-encesconversionintoleptonsinthetracker,isevaluatedusingMC simulation.

The contribution from QCD WZ production is estimated with MC simulation. It isconsidered signal forthe WZjj cross section measurement, but is the dominant background for the EW WZ measurement and in searches for new physics. For the EW WZ measurementandnewphysicssearches,thenormalizationofthe QCDWZ process isconstrainedbydata intheQCD WZ sideband region. The cross section predicted by the MLM-merged sample inthe QCD WZ sidebandregion is 18

.

6₋₂+2_..9₃ (scale)

±

1

.

0(PDF) fb, where the scale and PDF uncertainties are calculated using the proceduredescribedinSection7.Inthisregionthenormalization correction, which is derived froma fit to the data, is consistent withunity.TheEW WZ process,consideredsignalfortheWZjj and EW WZ measurementsbutbackgroundtonewphysicssearches,is alsoestimatedusingMCsimulation.

The contribution from background processes with nonprompt leptons is evaluated with datacontrol samplesof events satisfy-ingrelaxedleptonidentificationrequirementsusingthetechnique described in Refs. [16,68]. Events satisfying the full analysis se-lection, with the exception that one, two, or three leptons pass relaxedidentificationrequirementsbutfail themorestringent re-quirementsapplied tosignal events,are selected toformrelaxed leptoncontrol samples.Thesecontrolsamplesare mutually inde-pendentand, additionally, independent fromthe signal selection. Thesmallcontributiontotherelaxedleptoncontrolsamplesfrom eventswiththreepromptleptonsisestimatedwithMCsimulation andsubtractedfromtheeventsamples.

Theexpectedcontributioninthesignalregionisestimated us-ing “loose-to-tight” efficiency factors applied to the lepton can-didates failing the analysis requirements in the control region events. The efficiency factors are calculated from a sample of Z

+ 

cand events, where Z denotes a pair of oppositely charged,

same-flavorleptons satisfyingthe full identificationrequirements and

|

m+−

−

mZ

|

<

10 GeV, and



candisa leptoncandidate

satis-fyingtherelaxedidentification.Theloose-to-tightefficiencyfactors areobtainedfromratiosofeventswherethe



candobjectsatisfies

thefullidentificationrequirementstoeventswhereall identifica-tioncriteriaarenotsatisfied,andisparameterizedasafunctionof pT and

η

.Across-checkofthetechniqueisperformedby

repeat-ingtheprocedurewithefficiencyfactorsderivedfromasampleof eventsdominatedbydijetproduction.Theloose-to-tightefficiency factorsobtainedinthetworegionsagreetowithin30%forthefull pTand

η

range.

Thismethod is validated in nonoverlapping data samples en-richedinDrell–Yanandtt contributions. TheDrell–Yansample is definedbyinvertingtheselectionrequirementinpmiss_T ,andthett sample is defined by requiring atleast one b-tagged jet and re-jecting eventswith

|

m

−

mZ

|

<

5 GeV while keepingall other

requirements forthe signal region. The predictions derived from therelaxedlepton datacontrol samplesagreewiththe measure-mentsintheDrell–Yanandtt datasamplestowithin20%.

Thesmallsize ofthelooseleptoncontrolsamplesandZ

γ

MC simulation limit differential predictions in the EW signal region.

Therefore, the combined shape of the estimated nonprompt and Z

γ

backgrounds forboth electrons andmuonsare used as back-ground for the EW WZ measurement and in the extraction of constraints on aQGCs. The normalizationof the distributions per channelaretakenfromtheratioofthenonprompt(Z

γ

)yieldina single channelto thetotalnonprompt(Z

γ

)eventyieldmeasured in WZjj events with no requirements on the dijetsystem. These ratios are consistentwithin the statisticaluncertaintywithratios measuredwhenrelaxingthejet pTrequirementinWZjj events,in

WZ eventsinclusiveinthenumberofjets,andineventssatisfying theEWsignalandQCDWZ sidebandselections.

7. Systematicuncertainties

Thedominantuncertaintiesinboththecrosssection measure-ment andnewphysics searchesare those associatedwiththejet energyscale(JES)andresolution(JER).TheJESandJER uncertain-ties are evaluated in simulated events by smearing and scaling therelevant observablesandpropagating theeffectsto theevent selectionandthekinematicvariablesusedintheanalysis.The un-certainty intheeventyield inthe EWsignalselection duetothe JESandJER is9% forQCD WZ and5% forEW WZ processes. For the QCD WZ (EW WZ)process, the JESuncertainty variesin the rangeof5–25%(3–15%)withincreasingvaluesofmjjand

|

η

jj

|

.

Theuncertaintiesinsignalandbackgroundprocessesestimated withMCsimulation areevaluatedfromthetheoretical uncertain-ties of the predictions.Event weights in the MC simulations are usedto evaluatevariations ofthe centralprediction.Scale uncer-tainties are estimatedby independently varying

μ

R and

μ

F by a

factor oftwo from their nominalvalues, withthe condition that 1

/

2

≤

μ

R

/

μ

F

≤

2. The maximal and minimal variations are

ob-tainedperbintoformashape-dependentvariationband.ThePDF uncertainties are evaluated by combiningthe predictions per bin from the fit and

α

s variations of the NNPDF3.0 set accordingto

theproceduredescribedinRef. [69] forMC replicasets.Thescale and PDF uncertainties are uncorrelated for different signal and background process and 100% correlated across bins for the dis-tributions used toextract results.ForMC simulations normalized toa crosssectioncomputedata higherorderinQCD,the uncer-taintiesarecalculatedfromtheorderoftheMCsimulation.

The uncertainty in modeling the EW WZ and QCD WZ pro-cesses has a large impact in the EW WZ measurement. In addi-tion tothe uncertainties fromscaleandPDF choice, comparisons of alternative matrix element and parton shower generators are considered. The uncertainty in the QCD WZ process is derived bycomparing thepredictionsoftheMLM-mergedsimulationand those obtainedwiththe FxFx-merged simulation,after fixing the normalizationto the observed datain the QCD WZ sideband re-gion. Differences between the predictions of the MC simulations inthe signal regionandin theratio oftheQCD WZ sideband to thesignal region eventyields areconsidered inthe comparisons. The differences inpredictions are generally within the scale and PDFuncertaintiesoftheMCsimulations,anda10%normalization uncertaintyisassignedtoaccountfortheobserveddiscrepancies. Theresultsobtainedusingthe powheg simulation,whichpredicts aslightlysoftermjjspectrum,arealsolargelycontainedwithinthe

theoreticaluncertaintiesconsidered.However,becauseWZjj events from this simulation arise from soft radiation from the parton shower,itisnotexplicitlyconsideredintheuncertaintyevaluation. FortheEW WZ process,theMCsimulationsdescribedinSection3

agree within the theoretical uncertainties from the PDF and the choiceof

μ

R and

μ

Fforthekinematicvariablesconsideredinthe

analysis,sonoadditionaluncertaintyisassigned.

The interference termis evaluated on particle-level simulated events selected fromthe MC simulations described in Section 3.

Itis positive,androughly12% ofthe EW WZ contributioninthe QCDWZ sidebandregionand4%intheEWsignalregionforboth MCsimulationsconsidered,consistentwiththeresultsreportedin Ref. [34]. The ratio of the interference to the EW WZ decreases withincreasingmjj,consistent withtheobservationsof Refs. [34, 70].Thesevaluesareusedasasymmetricshapeuncertaintyinthe EW WZ prediction.Thisuncertaintyislowerthanothertheoretical uncertainties andhas anegligiblecontribution totheuncertainty intheEW WZ measurement.

Higher-orderEWcorrectionsinVBSprocessesareknowntobe negativeandattheleveloftensofpercent,withthecorrection in-creasingin magnitude with increasing mjj andmVV [71]. We do

not apply corrections to the WZjj MC simulation, but we have verified that the significance of the EW WZ measurement is in-sensitivetohigher-orderEWcorrectionsbyperforming thesignal extractiondescribedinSection8withthemjjpredictedbytheEW

WZ MCsimulationmodified bythe correctionsfromRef. [72].As the relative effect of the EW corrections on SM and anomalous WZjj productionis unknown, wedo not apply correctionstothe SM backgrounds or new physics signals for our results. Because correctionstotheSMWZjjproductionthatdecreasetheexpected number ofevents at highmWZ lead to more stringent limitson

newphysics,thisisaconservativeapproach.

The uncertainties related to the finite number of simulated events,ortothelimitednumberofeventsindatacontrolregions, affectthesignalandbackgroundpredictions.Theyareuncorrelated across differentsamples, andacross bins ofa single distribution. The limitednumberofeventsin therelaxedlepton control sam-plesusedforthenonpromptbackgroundestimateisthedominant contributiontothisuncertainty.

Thenonpromptbackgroundestimateisalsoaffectedby system-atic uncertainties from the jet flavor composition of the relaxed leptoncontrolsamplesandloose-to-tightextrapolationfactors.The systematic uncertainty in the nonprompt event yield is 30% for bothelectronsandmuons, uncorrelatedbetweenchannels. It cov-ers the largest difference observed between the estimated and measured numbersofeventsin datacontrolsamples enrichedin tt andDrell–Yancontributionsandthe differencesbetweenusing extrapolationfactorsderivedinZ

+

jet anddijetevents.

Systematic uncertainties are less than 1% forthe trigger effi-ciencyand1–3%fortheleptonidentificationandisolation require-ments, depending on the lepton flavors. Other systematic uncer-tainties are related to the use of simulated samples: 1% for the effectsofpileupand1–2%forthe pmiss_T reconstruction, estimated byvaryingtheenergiesofthePFobjectswithintheiruncertainties. The uncertainty in the b tagging efficiencyis 2% forWZ events, which accounts for differences in b tagging efficiencies between MCsimulations anddata.Theuncertaintyintheintegrated lumi-nosityofthedatasampleis2.5% [73].Thisuncertaintyaffectsboth thesignalandthesimulatedportionofthebackgroundestimation, butdoesnotaffectthebackgroundestimationfromdata.

For the extraction of results, log-normal probability density functionsare assumedfor the nuisance parameters affecting the eventyieldsofthevariousbackgroundcontributions,whereas sys-tematicuncertaintiesthat affecttheshapeofthedistributionsare represented by nuisance parameters whose variation resultsin a continuous perturbation of the spectrum [74] and are assumed to have a Gaussian probability density function. A summary of thecontribution ofeach systematicuncertainty tothe total WZjj cross section measurement is presented in Table 2. The impact of each systematic uncertainty in the WZjj cross section mea-surement is obtained by freezing the set of associated nuisance parameterstotheirbest-fitvaluesandcomparingthetotal uncer-taintyinthesignalstrengthtotheresultfromthenominalfit.The prompt background normalization uncertainty includes the scale

Table 2

ThedominantsystematicuncertaintycontributionsinthefiducialWZjj crosssection measurement.

Source of syst. uncertainty Relative uncertainty inσWZjj[%]

Jet energy scale +11/−8.1

Jet energy resolution +1.9/−2.1 Prompt background normalization +2.2/−2.2 Nonprompt normalization +2.5/−2.5 Nonprompt event count +6.0/−5.8 Lepton energy scale and eff. +3.5/−2.7

b tagging +2.0/−1.7

Integrated luminosity +3.6/−3.0

andPDF uncertainties inthe backgroundprocessesestimated us-ingMCsimulations.

8. FiducialWZjj crosssectionmeasurementandsearchforEW WZ production

The cross section for WZjj production, without separating by production mechanism, is measured with a combined maximum likelihood fit to the observed event yields for the EW signal se-lection. The likelihood is a combination of individual likelihoods for the four leptonic decaychannels (eee, ee

μ

,

μμ

e,

μμμ

) for the signal and background hypotheses with the statistical and systematic uncertainties in the form of nuisance parameters. To minimize the dependence of the result on theoretical predic-tions, the likelihood function is built from the event yields per channel without considering information about the distribution of events in kinematic variables. The expected event yields for the EW- and QCD-induced WZjj processes are taken from the MadGraph5_amc@nlo v2.4.2predictions.TheWZjj signalstrength

μ

WZjj, whichis theratio ofthemeasured signal yield to the

ex-pected numberofsignal events,is treatedasa free parameterin thefit.

Thebest-fitvaluefortheWZjj signalstrengthisusedtoobtain a crosssectioninthetightfiducialregiondefinedinTable1.The measuredfiducialWZjj crosssectioninthisregionis

σ

_WZjjfid

=

3

.

18+0₋₀._.57₅₂(stat)₋+0₀._.43₃₆(syst) fb

=

3

.

18+0₋₀._.71₆₃fb

.

Thisresultcanbecomparedwiththepredictedvalueof3

.

27+₋0₀._.39₃₂ (scale)

±

0

.

15 (PDF) fb.TheEW WZ andQCDWZ contributionsare calculatedindependentlyfromthesamplesdescribed inSection 3

andtheir uncertainties are combinedinquadrature toobtain the WZjj cross sectionprediction.The predictedEW WZ crosssection is1

.

25+₋0_0..₀₉11(scale)

±

0

.

06 (PDF) fb,andtheinterferenceterm con-tributioninthisregionislessthan1%ofthetotalcrosssection.

Results arealsoobtainedinalooserfiducialregion,definedin Table 1followingRef. [34], tosimplifycomparisonswith theoret-ical calculations. The acceptance from the loose to tight fiducial regionis

(

72

.

4

±

0

.

8

)

%,computedusing MadGraph5_amc@nlo in-terfacedto pythia.Theuncertaintyintheacceptanceisevaluated by combiningthescaleandPDFuncertainties intheEW WZ and QCD WZ predictions in quadrature. The scale uncertainty in the QCD WZ contribution is the dominant component of the uncer-tainty.TheresultingWZjj loosefiducialcrosssectionis

σ

_WZjjfid,loose

=

4

.

39+0_−0..78₇₂(stat)+0₋₀._.60₅₀(syst) fb

=

4

.

39+0₋₀._.98₈₇fb

,

compared with the predicted value of 4

.

51₋+0₀._.59₄₅ (scale)

±

0

.

18 (PDF) fb. The EW WZ and QCD WZ contributionsandtheir uncertaintiesaretreatedindependentlywiththesameapproachas describedforthetightfiducialregion.ThepredictedEW WZ cross sectioninthelooseregionis1

.

48+₋₀0_..13₁₁(scale)

±

0

.

07 (PDF) fb,and therelativecontributionfromtheinterferencetermislessthe1%.

Fig. 2. Themjj(upper)and|ηjj|(lower)ofthetwoleadingjetsforevents

satis-fyingtheEWsignalselection.Thelastbincontainsalleventswithmjj>2500 GeV

(upper)and|ηjj|>7.5 (lower).ThedashedlineshowstheexpectedEW WZ

con-tributionstackedontopofthebackgroundsthatareshownasfilledhistograms.The hatchedbandsrepresentthetotalandrelativestatisticaluncertaintiesonthe pre-dictedyields.Thebottompanelshowstheratioofthenumberofeventsmeasured indatatothetotalnumberofexpectedevents.Thepredictedyieldsareshownwith theirpre-fitnormalizations.

Separating the EW- and QCD-induced components of WZjj events requires exploiting the different kinematic signatures of the two processes. The relative fraction of the EW WZ process withrespecttotheQCDWZ processandotherbackgroundsgrows withincreasingvaluesofthemjjand

|

η

jj

|

oftheleadingjets, as

demonstratedinFig.2.Thismotivatestheuseofa2Ddistribution builtfrom these variables for the extraction of the EW WZ sig-nalvia a maximumlikelihood fit.This 2D distribution,shownas aone-dimensionalhistograminFig.3,alongwiththeyieldinthe QCDWZ sidebandregion,arecombinedinabinnedlikelihood in-volvingtheexpectedandobservednumbersofeventsineachbin. The likelihood is a combination of individual likelihoods for the fourdecaychannels.

Fig. 3. Theone-dimensionalrepresentationofthe2Ddistributionofmjjand|ηjj|,

usedfor theEW signalextraction.Thexaxis showsthe mjj distribution inthe

indicatedbins,splitintothreebinsofηjj:ηjj∈ [2.5,4],[4,5],≥5.Thedashed

linerepresentsthe EW WZ contributionstackedontopofthebackgroundsthat areshownasfilledhistograms.Thehatchedbandsrepresentthetotalandrelative systematicuncertaintiesonthepredictedyields.Thebottompanelshowstheratio ofthenumberofeventsmeasuredindatatothetotalnumberofexpectedevents. Thepredictedyieldsareshownwiththeirbest-fitnormalizations.

The systematic uncertainties are represented by nuisance pa-rameters that are allowed to vary according to their proba-bility density functions, and correlation across bins and be-tween different sources of uncertainty is taken into account. The expected number of signal events is taken from the Mad-Graph_{5_amc@nlo v2.4.2 prediction atLO, multiplied by a signal} strength

μ

EWwhichistreatedasafreeparameterinthefit.

Thebest-fitvalueforthesignalstrength

μ

EWis

μ

EW

=

0

.

82+₋00..5143

,

consistentwiththeSMexpectationatLOof

μ

EW,LO

=

1,with

re-spect to the predicted cross section for the EW WZ process in the tight fiducial region. The significance of the signal is quan-tified by calculating the local p-value for an upward fluctuation of the data relative to the background prediction usinga profile likelihoodratioteststatisticandasymptoticformulae [75].The ob-served(expected)statisticalsignificanceforEW WZ productionis 2.2(2.5)standarddeviations.Amodificationtothepredictedcross sectionusedinthefittriviallyrescalesthesignalstrengthbutdoes not impactthe significance ofthe result.The total uncertaintyof themeasurementisdominatedbythestatisticaluncertaintyofthe data.Thepost-fityieldsforthesignalandbackground correspond-ingtothebest-fitsignalstrengthforEW WZ productionareshown inTable3.

9. Limitsonanomalousquarticgaugecouplings

Events satisfying the EW signal selection are used to con-strain aQGCs in the effective field theory approach [76]. Results are obtainedfollowingthe formulationofRef. [21] that proposes nine independent dimension-eight operators, which assume the SU(2)

×

U(1)symmetryoftheEWgaugesectoraswellasthe pres-enceof an SM Higgsboson. Alloperators are charge conjugation andparity-conserving. The WZjj channelis mostsensitive to the T0,T1,andT2operatorsthatareconstructedpurelyfromtheSU(2) gaugefields,theS0andS1operatorsthatinvolveinteractionswith theHiggsfield,andtheM0andM1operatorsthat involvea mix-tureofgaugeandHiggsfieldinteractions.

Table 3

Post-fiteventyieldsafterthesignalextractionfittoeventssatisfyingtheEWsignalselection.TheEW WZ processiscorrectedfortheobservedvalueofμEW.

Process μμμ μμe eeμ eee Total yield

QCD WZ 13.5±0.8 9.1±0.5 6.8±0.4 4.6±0.3 34.1±1.1 t+V/VVV 5.6±0.4 3.1±0.2 2.5±0.2 1.7±0.1 12.9±0.5 Nonprompt 5.2±2.0 2.4±0.9 1.5±0.6 0.7±0.3 9.8±2.3 VV 0.8±0.1 1.6±0.2 0.4±0.0 0.7±0.1 3.5±0.2 Zγ 0.3±0.1 1.2±0.8 <0.1 0.6±0.2 2.2±0.8 Pred. background 25.5±2.1 17.4±1.5 11.2±0.8 8.3±0.6 62.4±2.8 EW WZ signal 6.0±1.2 4.2±0.8 2.9±0.6 2.1±0.4 15.1±1.6 Data 38 15 12 10 75

Fig. 4. mT(WZ)foreventssatisfyingtheEW signalselection,usedtoplace

con-straintsontheanomalouscouplingparameters.Thedashedlinesshowpredictions forseveralaQGCparametersvaluesthatmodifytheEW WZ process.Thelastbin containsalleventswithmT(WZ)>2000GeV.Thehatchedbandsrepresentthe

to-talandrelativesystematicuncertaintiesonthepredictedyields.Thebottompanel showstheratioofthenumberofeventsmeasuredindatatothetotalnumberof expectedevents.Thepredictedyieldsareshownwiththeirbest-fitnormalizations fromthebackground-onlyfit.

ThepresenceofnonzeroaQGCswouldenhancetheproduction ofeventswithhighWZmass.Thismotivatestheuseofthe trans-versemassoftheWZ system,definedas

mT

(

WZ

)

=

[ET

(

W

)

+

ET

(

Z

)

]2

−





pT

(

W

)

+ 

pT

(

Z

)



2

,

withET

=

√

m2

+

p2_T,wheretheW candidate isconstructedfrom the



pmiss

T andtheleptonassociatedwiththeW boson,andm isthe

invariantmassoftheW or Z candidate,to constrainthe parame-ters f_Oi

/

4.Inthisformulation, fOiisadimensionlesscoefficient

fortheoperator

O

iand

istheenergyscaleofnewphysics.The mT

(

WZ

)

foreventssatisfyingtheEW signalselection isshownin

Fig.4.ThepredictionsofseveralindicativeaQGCoperatorsand co-efficientsarealsoshown.

TheMCsimulationsofnonzeroaQGCsincludetheSMEW WZ process,withanincreaseintheyieldathighmT

(

WZ

)

arisingfrom

parametersdifferentfromtheirSMvalues.Becausetheincreaseof theexpectedyieldovertheSMpredictionexhibitsaquadratic de-pendenceontheoperatorcoefficient,aparabolicfunctionisfitted to the predicted yields per bin to obtain a smooth interpolation between the discrete operator coefficients considered in the MC simulation.The one-dimensional 95% confidence level (CL)limits are extractedusingthe CLs criterion [77,78,75], withall

parame-ters except for the coefficient being probed set to zero. The SM

Table 4

Observedandexpected95%CL limitsforeachoperatorcoefficient(inTeV−4_)while

allotherparametersaresettozero.

Parameters Exp. limit Obs. limit

fM0/ 4 [−11.2,11.6] [−9.15,9.15] fM1/ 4 [−10.9,11.6] [−9.15,9.45] fS0/ 4 [−32.5,34.5] [−26.5,27.5] fS1/ 4 [−50.2,53.2] [−41.2,42.8] fT0/ 4 [−0.87,0.89] [−0.75,0.81] fT1/ 4 [−0.56,0.60] [−0.49,0.55] fT2/ 4 [−1.78,2.00] [−1.49,1.85]

prediction,includingtheEW WZ process,istreatedasthenull hy-pothesis.Theexpectedpromptbackgroundsarenormalizedtothe predictionsoftheMCsimulations,withnocorrectionsappliedfor the results of the EW WZ or WZjj measurements. No deviation from the SM prediction is observed, and the resulting observed andexpectedlimitsaresummarizedinTable4.

Constraints are also placed on aQGC parameters using a two-dimensionalscan,wheretwoparametersareprobedinthefitwith all others set tozero. Thisapproach is motivatedby correlations between operators and physical couplings, and for comparisons withalternativeformulationsofdimension-eightoperators.In par-ticular,thequarticgaugeinteractionsofthemassivegaugebosons isafunctionofS0andS1,whilecombinationsoftheM0andM1 operators canbecompared withtheformulationofRef. [79].The resulting 2D 95% CL intervals forthese parameters are shownin Fig.5.

10. LimitsonchargedHiggsbosonproduction

Theories with Higgs sectors including SU(2) triplets can give risetochargedHiggsbosons(H±)withlargecouplingstothe vec-tor bosons of the SM. A prominent one is the GM model [47], wherethe Higgssectorisextended byone realandone complex SU(2) triplet to preserve custodialsymmetry at treelevel for ar-bitrary vacuumexpectationvalues.Inthismodel,thecouplingsof H± andthe vector bosons depend on m

(

H±

)

andthe parameter sin

θ

H,orsH,whichrepresentsthemixingangleofthevacuum

ex-pectationvaluesinthemodel,anddeterminesthefractionofthe W andZ bosonmassesgeneratedbythevacuumexpectation val-uesofthetriplets.Thisanalysisextendsthepreviousstudy ofH± productionviavectorbosonfusionbytheCMSCollaborationinthe samechannel [68].

A combined fitof the predictedsignal and backgroundyields to the data inthe Higgs bosonselection isperformed inbins of mT

(

WZ

)

,simultaneouslywiththeeventyieldintheQCDWZ

side-band region,toderive model-independentexpectedandobserved upper limitson

σ

(

H

+

jj

)(

H±

)

B(

H±

→

WZ

)

at 95% CL using the CLs criterion. The distribution andbinning of the mT

(

WZ

)

distri-bution used inthe fitare shownin Fig.6.The upper limitsasa functionofm

(

H±

)

areshowninFig.7(upper).Theresultsassume thattheintrinsicwidthoftheH±is



0.05m

(

H±

)

,whichisbelow theexperimentalresolutioninthephasespaceconsidered.

Fig. 5. Two-dimensionalobserved95%CL intervals(solidcontour)andexpected68, 95,and99%CL intervals(dashedcontour)ontheselectedaQGCparameters.The valuesofcoefficientsoutsideofcontoursareexcludedatthecorrespondingCL.

The model-independent upper limits are compared with the predictedcrosssectionsatnext-to-next-to-leadingorderintheGM modelinthesH-m

(

H±

)

plane,undertheassumptions definedfor

the “H5plane” in Ref. [48]. For the probed parameter space and mT

(

WZ

)

distributionusedforsignal extraction,thevaryingwidth

asafunctionofsHisassumedtohavenegligibleeffectonthe

re-sult.Thevalueofthebranchingfraction

B(

H±

→

WZ

)

isassumed tobe unity.InFig.7(lower),theexcluded sHvaluesasafunction

ofm

(

H±

)

areshown.Theblueshadedregionshowstheparameter spaceforwhichtheH±totalwidthexceeds10%ofm

(

H±

)

,where themodelisnotapplicable becauseofperturbativityandvacuum stabilityrequirements [48].

11. Summary

AmeasurementoftheproductionofaW andaZ bosonin as-sociation withtwo jets has been presented, using events where

Fig. 6. mT(WZ)foreventssatisfyingtheHiggsbosonselection,usedtoplace

con-straintsontheproductionofchargedHiggsbosons.Thelastbincontainsallevents withmT(WZ)>2000 GeV.ThedashedlinesshowpredictionsfromtheGMmodel

withm(H±)=400(900)GeV andsH=0.3(0.5).Thebottompanelshowstheratio

ofthenumberofeventsmeasuredindatatothetotalnumberofexpectedevents. Thehatchedbandsrepresentthetotalandrelativesystematicuncertaintiesonthe predictedbackgroundyields.Thepredictedyieldsareshownwiththeirbest-fit nor-malizationsfromthebackground-onlyfit.

both bosons decay leptonically. Results are based on data corre-sponding to an integrated luminosity of 35

.

9 fb−1 recorded in proton-protoncollisionsat

√

s

=

13 TeV withtheCMSdetectorat theLHC in2016.The crosssection inatightfiducial regionwith enhanced contributions fromelectroweak (EW)WZ production is

σ

fid

WZjj

=

3

.

18+ 0.71

−0.63fb,consistentwiththestandardmodel(SM)

pre-diction. The dijet mass anddijet rapidity separation are used to measurethesignalstrengthofEW WZ productionwithrespectto theSMexpectation,resultingin

μ

EW

=

0

.

82₋+00..5143.Thesignificance

of thisresultis 2.2standard deviations with2.5 standard devia-tionsexpected.

Constraintsareplacedonanomalousquarticgaugecouplingsin termsofdimension-eighteffectivefieldtheoryoperators, and up-perlimitsaregivenontheproductioncrosssectiontimes branch-ingfractionofchargedHiggsbosons.Theupperlimitsoncharged Higgs boson production via vector boson fusion with decayto a W anda Z bosonextend theresults previously published by the CMSCollaboration [68] andarecomparabletothoseoftheATLAS Collaboration [80]. These are the first limits fordimension-eight effectivefieldtheoryoperatorsintheWZ channelat13 TeV.

Acknowledgements

We congratulate our colleagues in the CERN accelerator de-partments for the excellent performance of the LHC and thank thetechnicalandadministrativestaffs atCERNandatother CMS institutes for their contributions to the success of the CMS ef-fort.Inaddition,wegratefullyacknowledgethecomputingcenters and personnel of the Worldwide LHC Computing Grid for deliv-ering so effectivelythe computing infrastructureessential to our analyses. Finally, we acknowledge the enduring support for the constructionandoperationoftheLHCandtheCMSdetector pro-videdby the followingfunding agencies:BMBWF andFWF (Aus-tria); FNRS and FWO (Belgium); CNPq, CAPES, FAPERJ, FAPERGS, andFAPESP(Brazil); MES (Bulgaria); CERN;CAS, MoST,andNSFC (China); COLCIENCIAS (Colombia); MSES and CSF (Croatia); RPF

Fig. 7. Expected(dashedlines)andobserved(solidlines)upperlimitsat95%CL for themodelindependentσ(H±)B(H±→W±Z)asafunctionofm(H±)(upper)and forsHasafunctionofmH intheGMmodel(lower).Theblueshadedareacovers

thetheoreticallynotallowedparameterspace [48].

(Cyprus);SENESCYT(Ecuador);MoER,ERCIUT,andERDF(Estonia); AcademyofFinland,MEC,andHIP(Finland);CEAandCNRS/IN2P3 (France); BMBF, DFG, and HGF (Germany); GSRT (Greece); NK-FIA (Hungary); DAE and DST (India); IPM (Iran); SFI (Ireland); INFN(Italy);MSIPandNRF(RepublicofKorea);MES(Latvia);LAS (Lithuania);MOEandUM(Malaysia);BUAP,CINVESTAV,CONACYT, LNS,SEP,andUASLP-FAI(Mexico);MOS(Montenegro);MBIE(New Zealand);PAEC(Pakistan);MSHEandNSC(Poland);FCT(Portugal); JINR (Dubna); MON, RosAtom, RAS, RFBR, and NRC KI (Russia); MESTD (Serbia); SEIDI, CPAN, PCTI, and FEDER (Spain); MOSTR (Sri Lanka); Swiss Funding Agencies (Switzerland); MST (Taipei); ThEPCenter,IPST,STAR,andNSTDA(Thailand);TUBITAKandTAEK (Turkey);NASUandSFFR(Ukraine); STFC(United Kingdom);DOE andNSF(USA).

Rachada-pisek Individuals have received support from the Marie-CurieprogramandtheEuropeanResearchCounciland Hori-zon2020 Grant,contract No. 675440 (EuropeanUnion);the Lev-entis Foundation; the A.P. Sloan Foundation; the Alexander von Humboldt Foundation; the Belgian Federal Science Policy Office; the Fonds pour la Formation à la Recherche dans l’Industrie et dans l’Agriculture (FRIA-Belgium); the Agentschap voor Innovatie door Wetenschap en Technologie (IWT-Belgium); the F.R.S.-FNRS and FWO (Belgium) under the “Excellence of Science – EOS” – be.h project n. 30820817; the Ministry of Education, Youth and Sports (MEYS) of the Czech Republic; the Lendület (“Momen-tum”) Programme and the János Bolyai Research Scholarship of the Hungarian Academy of Sciences, the New National

Excel-lence Program ÚNKP,the NKFIAresearch grants 123842, 123959, 124845, 124850, and 125105 (Hungary); the Council of Science and Industrial Research, India; the HOMING PLUS programme of the Foundation for Polish Science, cofinanced from European Union, RegionalDevelopmentFund, theMobilityPlusprogramme oftheMinistryofScienceandHigherEducation,theNational Sci-ence Center (Poland), contracts Harmonia 2014/14/M/ST2/00428, Opus 2014/13/B/ST2/02543, 2014/15/B/ST2/03998, and 2015/19/B/ST2/02861, Sonata-bis 2012/07/E/ST2/01406; the Na-tional Priorities Research Program by Qatar National Research Fund; thePrograma Estatalde Fomento de laInvestigación Cien-tífica yTécnicadeExcelenciaMaríadeMaeztu,grant MDM-2015-0509andthePrograma SeveroOchoadel Principadode Asturias; the Thalis and Aristeia programmes cofinanced by EU-ESF and the Greek NSRF; the Rachadapisek Sompot Fund for Postdoctoral Fellowship, ChulalongkornUniversity andthe Chulalongkorn Aca-demic into Its 2nd Century Project Advancement Project (Thai-land); the Welch Foundation, contract C-1845; and the Weston HavensFoundation(USA).

References

[1] ATLASCollaboration,Observationofanewparticleinthesearchforthe stan-dardmodelHiggsbosonwiththeATLASdetectorattheLHC,Phys.Lett.B716 (2012)1,https://doi.org/10.1016/j.physletb.2012.08.020,arXiv:1207.7214. [2] CMSCollaboration,Observationofanewbosonatamassof125GeVwith

theCMSexperimentattheLHC,Phys.Lett.B716(2012)30,https://doi.org/10. 1016/j.physletb.2012.08.021,arXiv:1207.7235.

[3] CMSCollaboration,Observationofanewbosonwithmassnear125GeVin ppcollisionsat√s=7 and8TeV,J.HighEnergyPhys.06(2013)081,https:// doi.org/10.1007/JHEP06(2013)081,arXiv:1303.4571.

[4] F.Englert,R.Brout,Brokensymmetryandthemassofgaugevectormesons, Phys.Rev.Lett.13(1964)321,https://doi.org/10.1103/PhysRevLett.13.321. [5] P.W.Higgs,Brokensymmetries,masslessparticlesandgaugefields,Phys.Lett.

12(1964)132,https://doi.org/10.1016/0031-9163(64)91136-9.

[6] P.W.Higgs,Brokensymmetriesandthemassesofgaugebosons,Phys.Rev.Lett. 13(1964)508,https://doi.org/10.1103/PhysRevLett.13.508.

[7] G.S.Guralnik,C.R.Hagen,T.W.B.Kibble,Globalconservationlawsandmassless particles,Phys.Rev.Lett.13(1964)585,https://doi.org/10.1103/PhysRevLett.13. 585.

[8] P.W.Higgs,Spontaneoussymmetrybreakdownwithoutmasslessbosons,Phys. Rev.145(1966)1156,https://doi.org/10.1103/PhysRev.145.1156.

[9] T.W.B.Kibble,Symmetrybreakinginnon-Abeliangaugetheories,Phys.Rev.155 (1967)1554,https://doi.org/10.1103/PhysRev.155.1554.

[10] ATLAS and CMS Collaborations, Measurements ofthe Higgsboson produc-tionand decayratesand constraintson itscouplingsfrom acombined AT-LAS and CMSanalysis ofthe LHC pp collisiondata at √s=7 and8 TeV, J.HighEnergyPhys.08(2016)045,https://doi.org/10.1007/JHEP08(2016)045, arXiv:1606.02266.

[11] CMS Collaboration, Combined measurements of Higgs boson couplings in proton-protoncollisionsat√s=13 TeV,Eur.Phys.J.C79(2019)421,https:// doi.org/10.1140/epjc/s10052-019-6909-y,arXiv:1809.10733.

[12] C.-W.Chiang,G.Cottin,O.Eberhardt,GlobalfitsintheGeorgi-Machacekmodel, Phys. Rev.D99 (2019) 015001, https://doi.org/10.1103/PhysRevD.99.015001, arXiv:1807.10660.

[13] D. Chowdhury,O.Eberhardt,Update ofglobaltwo-Higgs-doubletmodelfits, J.HighEnergyPhys.05(2018)161,https://doi.org/10.1007/JHEP05(2018)161, arXiv:1711.02095.

[14] ATLASCollaboration,MeasurementofWZproduction inproton-proton colli-sionsat√s=7 TeVwiththeATLASdetector,Eur.Phys.J.C72(2012)2173, https://doi.org/10.1140/epjc/s10052-012-2173-0,arXiv:1208.1390.

[15] ATLASCollaboration,MeasurementsofW±Z productioncrosssectionsinpp collisionsat √s=8 TeVwiththe ATLASdetector andlimits onanomalous gaugebosonself-couplings,Phys.Rev.D93(2016)092004,https://doi.org/10. 1103/PhysRevD.93.092004,arXiv:1603.02151.

[16] CMSCollaboration,MeasurementoftheWZproductioncrosssectioninpp col-lisionsat√s=13 TeV,Phys.Lett.B766(2017)268,https://doi.org/10.1016/j. physletb.2017.01.011,arXiv:1607.06943.

[17] CMSCollaboration,MeasurementoftheWZproductioncrosssectioninpp col-lisionsat√s=7 and8TeVandsearchforanomaloustriplegaugecouplingsat √

s=8 TeV,Eur.Phys.J.C77(2017)236,https://doi.org/10.1140/epjc/s10052 -017-4730-z,arXiv:1609.05721.

[18]ATLAS Collaboration, Measurement of W±Z production cross sections and gaugebosonpolarisationinpp collisionsat√s=13 TeVwiththeATLAS de-tector,Eur.Phys.J.C(2019),inpress,arXiv:1902.05759.