Search for dark matter produced in association with a hadronically decaying vector boson in pp collisions at √s = 13 TeV with the ATLAS detector

(1)

Article

Reference

Search for dark matter produced in association with a hadronically decaying vector boson in pp collisions at √s = 13 TeV with the ATLAS

detector

ATLAS Collaboration

ANCU, Lucian Stefan (Collab.), et al .

Abstract

A search is presented for dark matter produced in association with a hadronically decaying W or Z boson using 3.2 fb −1 of pp collisions at s=13 TeV recorded by the ATLAS detector at the Large Hadron Collider. Events with a hadronic jet compatible with a W or Z boson and with large missing transverse momentum are analysed. The data are consistent with the Standard Model predictions and are interpreted in terms of both an effective field theory and a simplified model containing dark matter.

ATLAS Collaboration, ANCU, Lucian Stefan (Collab.), et al . Search for dark matter produced in association with a hadronically decaying vector boson in pp collisions at √s = 13 TeV with the ATLAS detector. Physics Letters. B , 2016, vol. 763, p. 251-268

DOI : 10.1016/j.physletb.2016.10.042

Available at:

http://archive-ouverte.unige.ch/unige:88705

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

1 / 1

(2)

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Search for dark matter produced in association with a hadronically decaying vector boson 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:

Received9August2016

Receivedinrevisedform9October2016 Accepted17October2016

Availableonline20October2016 Editor:W.-D.Schlatter

A searchis presented for dark matterproducedin associationwith ahadronically decaying W or Z boson using 3.2 fb⁻¹ of pp collisions at √

s=13 TeV recordedby the ATLAS detector at the Large HadronCollider.EventswithahadronicjetcompatiblewithaW or Z bosonandwith largemissing transverse momentumareanalysed.The dataare consistentwiththeStandardModel predictionsand areinterpretedintermsofbothaneffectiveﬁeldtheoryandasimpliﬁedmodelcontainingdarkmatter.

©²⁰¹⁶^TheÂuthor.^Published^byÊlsevier^B.V.^Thisîsânôpenâccessârticleûnder^the^CC^BY^license (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

Darkmatteris thedominantcomponentofmatter intheuni- verse, but its particle nature remains a mystery. Searches for a weaklyinteractingmassiveparticle(WIMP),denotedby χ,andfor interactions betweenχ andStandard Model (SM) particles are a centralcomponentofthecurrentsetofdark-matterexperiments.

At particle colliders, dark-matter particles may be produced in pairs via some unknown intermediate state. While in many modelsdirect detectionexperiments havethe greatestsensitivity for dark-matter masses mχ between 10 and 100 GeV, searches fordark matter atparticle collidersare mostpowerfulfor lower masses [1–3]. The ﬁnal-state WIMPs are not directly detectable, buttheirpresencecanbeinferredfromtherecoilagainstavisible particle[1].TwoexampleprocessesareshowninFig. 1.

The Tevatron and LHC collaborations have reported limits on the cross section of p^p¯ →χχ¯ +^X ^and ^pp→χχ¯ +^X, ^respectively,where X isahadronicjet[1–3],a photon(γ)[4,5],aW/Z boson[6,7],oraHiggsboson [8,9].Inmanycases,resultsarere- portedin terms oflimits on the parameters of an effective ﬁeld theory(EFT)formulatedasafour-pointcontactinteraction[10–18]

betweenquarks andWIMPs. Forsuch models, thestrongest lim- itscomefromdatainwhichtherecoilingobjectisajet. Inother models,however,theinteractionisbetweendarkmatter andvec- torbosons[19],suchthattheprimarydiscoverymodewouldbein ﬁnalstatessuchasthoseanalysedhere,wheretherecoilingobject isaW orZ boson.

E-mailaddress:[email protected].

In thisLetter,a search is reportedfor theproduction ofa W or Z bosondecaying hadronically (toqq¯ orqq,¯ respectively) and reconstructed as a single massive jet in association with large missing transverse momentum from the undetected χχ¯ particles in data collected by the ATLAS detector from pp collisions withcentre-of-mass energy √

s=^{13 TeV.} ^This ^search ^is^sensitive to WIMPpair production,aswell asto other dark-matter-related models which predict invisible Higgs boson decays (W H or Z H productionwithH→χχ¯).

TheATLASdetector[20]attheLHCcoversthepseudorapidity¹ range |η|<4.9 and the full azimuthal angleφ. It consistsof an innertrackingdetectorsurroundedbyasuperconductingsolenoid, electromagneticandhadroniccalorimeters,andanexternalmuon spectrometer incorporating large superconducting toroidal mag- nets.A two-leveltriggersystemisusedtoselectinterestingevents toberecordedforsubsequentoﬄineanalysis.Onlydataforwhich beamswerestableandallsubsystemsdescribedabovewereoper- ationalareused.Applyingtheserequirementsto ppcollisiondata, recordedduringthe2015LHCrun,resultsinadatasamplewitha time-integratedluminosityof3.2 fb⁻¹.Thesystematicuncertainty

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

Thex-axispointsfromtheIPtothecentreoftheLHCring,andthey-axispoints upward.Polarcoordinates(r,φ)areusedinthetransverse(x,y)plane,φbeingthe azimuthalanglearoundthebeampipe.Thepseudorapidityisdeﬁnedintermsof thepolarangleθasη= −ln tan(θ/2).

http://dx.doi.org/10.1016/j.physletb.2016.10.042

0370-2693/©²⁰¹⁶^TheÂuthor.^Published^byÊlsevier^B.V.^Thisîsânôpenâccessârticleûnder^the^CC^BY^license(http://creativecommons.org/licenses/by/4.0/).Fundedby SCOAP³.

(3)

Fig. 1.PairproductionofWIMPs(χχ¯)inproton–protoncollisionsat theLHCinassociationwithavectorboson(V,meaning W or Z)viatwohypotheticalprocesses:

(a) productionviaaneffectiveV Vχ χinteractionor(b) viaasimpliﬁedmodelwhichincludesans-channelmediator.

of2.1%intheluminosityisderivedfollowingthesamemethodol- ogyasthatdetailedinRef.[21].

Threenon-exclusivecategoriesofjetcandidatesarebuilt,each using the anti-k_⊥ clustering algorithm [22]. Two categories use clusters of energy deposits in calorimeter cells seeded by those with energies signiﬁcantly above the measured noise and cali- bratedatthehadronicenergyscale[25].Theyaredistinguishedby theirradiusparameters;jetswithradiusparameterof1.0(0.4)are referred toaslarge-Rjets(narrowjets). Largeandnarrowjetscan shareafractionoftheir energydeposits.A thirdtypeofjetcandi- dateisreconstructedfrominner-detectortracksusingtheanti-k_⊥ algorithm with R=⁰.2,referred to astrackjets. Large-R jetsare trimmed[26] toremoveenergydepositedbypile-upjets,theun- derlyingevent,andsoftradiation.Inthisprocess,theconstituents oflarge-R jetsarereclusteredusingthek_⊥ algorithm[23,24]with adistanceparameterof0.2,andsubjetswithtransversemomen- tum pT less than5% ofthe large-R jet pT are removed. Large-R jets are required to satisfy pT>200 GeV and |η_|<2.0. These large-R jetsareintendedtocapturethehadronicproductsofboth quarksfromthedecayofa W or Z boson,whilethenarrowjets andtrackjetsare helpfulinbackgroundsuppression.Theinternal structureofthelarge-Rjetischaracterizedintermsoftwoquanti- ties:D2[27,28],whichidentiﬁesjetswithtwodistinctconcentra- tionsofenergy[29,30],andmjet,whichisthecalculatedinvariant massofthejet.Narrowjetsarerequiredtosatisfy pT>20 GeV for

|η|<2.5 orp_T>30 GeV for2.5<|η|<4.5.Trackjetsarerequired tosatisfy p_T>10 GeV and|η|<2.5.Forboththelarge-Randnar- rowjets, jetmomenta are calculatedby performing afour-vector sumoverthesecomponentclusters,treatingeachtopologicalclus- ter[25]asan(E,p)four-vectorwithzeromass,andarecalibrated tothehadronicscale.Fornarrowjets,thedirectionofpisgivenby thelinejoiningthereconstructedvertexwiththebarycentreofthe energycluster. The missingtransverse momentum E^miss_T iscalcu- latedasthenegativeofthevectorsumofthetransversemomenta ofreconstructed jets, leptons, andthose tracks whichare associ- atedwiththereconstructedvertexbutnotwithanyjetorlepton.

A closelyrelatedquantity,E^miss_T_,_noμ,iscalculatedinthesamewaybut excludingreconstructedmuons.A thirdvariant,p^miss_T ,isthemiss- ing transverse momentum measured using inner detectortracks.

The magnitudesofthe threemissing-transverse-momentumvari- antsaredenotedbyE^miss_T ,E^miss_T_,_noμ,andp^miss_T ,respectively.Electrons, muons,jets, andE^miss_T arereconstructedasdescribedinRefs.[25, 31–33],respectively.

Candidate signal events are selected by an inclusive E^miss_T trigger that is more than 99% eﬃcient for events with E^miss_T >

200 GeV. Events triggered by detector noise and non-collision backgrounds are rejected as described in Ref. [34]. In addition, eventsarerequiredtosatisfytherequirementsofE^miss_T >250 GeV, no reconstructed electrons or muons, and at least one large-R jet with pT>200 GeV, |η|<2.0,mjet and D2 consistent witha W or Z boson decay as in Ref. [35]. To further suppress back-

grounds from multijet and t¯t production, events are required to satisfy p^miss_T >30 GeV, a minimum azimuthal angular distance, φ, of 0.6 between the E^miss_T and the nearest narrow jet, and φ (E^miss_T ,p^miss_T )<π/2. Within a fiducialvolume defined atpar- ton level by similar selection requirements (except those on D2 and p^miss_T ),thereconstruction efficiencyforthesignalmodelsde- scribedabovevariesfrom38%to49%.

Thedominantsourceofbackgroundeventsis Z→νν¯ produc- tioninassociationwithjets.A secondarycontributioncomesfrom the productionof jetsinassociation witha leptonically decaying W or Z boson in which the charged leptons are not identiﬁed or the τ leptons decayhadronically.The third major background contributioncomesfromtop-quarkpairproduction.Thekinematic distributions ofthesethreelargestbackgroundsare estimatedus- ing simulatedeventsamplesbutthe normalizationis determined usingcontrol regionswherethedark-mattersignal isexpectedto be negligible. Each control region requires E^miss_T >200 GeV and p^miss_T >30 GeV aswellasonelarge-R jetsatisfyingthesubstruc- ture requirement on D₂ as applied in the signal region. The Z boson control region requires exactly two muons with dimuon invariant mass 66<mμμ<116 GeV. The W boson (top quark) control regionrequires exactlyone muon,andzero(atleastone) b-tagged trackjet not associated withthe large-R jet. Validation of the reconstruction of hadronic W boson decays with large-R jets is performed in the top-quark control region, as shown in Fig. 2,whichalsopresentsthedistributionofthe D2 substructure variable.Othersources ofbackgroundaredibosonproductionand single-top-quarkproduction.Thecontributiontothesignal region frommultijetproductionisnegligible.

Samplesof simulatedW + ^jetsând ^Z +^jetsêvents âre^generated using Sherpa 2.1.1 [36]. Matrix elements are calculated for up to two partons at next-to-leading order (NLO) and four partons at leading order (LO) using the Comix [37] and Open- Loops[38]matrixelementgeneratorsandmergedwiththeSherpa parton shower [39] using the ME+PS@NLO prescription [40]. The CT10 [41] PDF set is used in conjunction with dedicated parton shower tuning developed by the Sherpa authors. The W/Z production rates are normalized to a next-to-next-to-leading order (NNLO) calculation [42]. The production of t¯t andsingle-top processes, including s-channel, t-channel and W t production is modelledwiththePowheg-Boxv2generator[43–45]interfacedto Pythia6.428[46].Inthesegeneratorsthe CT10andCTEQ6L1[47]

PDF sets areused, respectively.Top-quark pairproductionis normalized to NNLO with next-to-next-to-leading-logarithm correc- tions [48] in QCD while single-top processes are normalized at NLO [49,50] in QCD. The diboson (W W,W Z,Z Z) processes are simulatedusingSherpa2.1.1withtheCT10PDFandnormalizedat NLO [51,52]inQCD.The multijetprocess isdescribedusingsam- ples simulated with Pythia8.186 [53] and the NNPDF2.3LO [54]

PDF at leading order in QCD; these multijet samples were used to develop thebackgroundestimationstrategy butnotforthe ﬁ- nalbackgroundprediction.

(4)

Fig. 2.Pane(a)Distributionofmjetinthedataandforthepredictedbackgroundinthetop-quarkcontrolregion.Pane(b)DistributionofjetsubstructurevariableD2inthe dataandforthepredictedbackgroundineventssatisfyingallsignalregionrequirementsotherthanthoseonD2.Alsoshownisthedistributionforthesimpliﬁedmodel withavector-bosonmediator,scaledbyafactorof10⁴forgivenvaluesofmχ andmmed,themediatormass.(Forinterpretationofthereferencestocolorinthisﬁgure legend,thereaderisreferredtothewebversionofthisarticle.)

Fig. 3.TheE^miss_T_,_no_μdistributionoftheeventsinthecontrolregionsaftertheprofile-likelihoodfittothedataunderthebackground-onlyhypothesis.Pane(a)showsthet¯t controlregion,pane(b)showsthe Z +jetscontrolregion,andpane(c)showsthe W +jetscontrolregion.Thetotalbackgroundpredictionbeforethefitisshownas adashedline.Theinsetatthebottomofeachplotshowstheratioofthedatatothetotalpost-fitbackground.Thehatchedbandsrepresentthetotaluncertaintyinthe background.(Forinterpretationofthereferencestocolorinthisfigurelegend,thereaderisreferredtothewebversionofthisarticle.)

(5)

Fig. 4.The E^miss_T distributionoftheeventsinthesignalregionaftertheprofile- likelihoodfit tothedataunderthebackground-onlyhypothesis.Theinsetshows theratioofthedatatothetotalbackground.AlsoshownistheE^miss_T distribution forthesimplifiedmodelwithavector-bosonmediator,scaledbyafactorof10⁴for mχ=^{10 GeV and}^mmed=^{10 TeV.}^The^total^background^before^the^fitîs^shownâsâ dashedline.Thehatchedbandsrepresentthetotaluncertaintyinthebackground.

(Forinterpretationofthereferencestocolorinthisﬁgurelegend,thereaderisreferredtothewebversionofthisarticle.)

SamplesofsimulatedWχχ¯ andZχχ¯ eventsaregeneratedus- ingMadGraph5_aMC@NLO[55],andtheunderlyingeventandpar- tonshoweringare simulatedwithPythia8.186[53]. Twotheoret- icalmodelsareusedasbenchmarks:aseven-dimensional V Vχ χ

EFT[19] model(V meaning W or Z)andavector-mediatedsim- plifiedmodel[56].ThestrengthoftheEFTinteractioniscontrolled by a mass scale, M, and the strength of the simplified model interaction is controlled by the product of the couplings of the mediatorto theSM andthedarkmatter(DM)particles, g_SMg_DM. The EFT model samples were generated with M=^{3000 GeV,} andthesimplifiedmodelsamples weregeneratedwithcouplings g_SM=⁰.25 and g_DM=^1. ^The^samples ^were^generated âsâ^func- tionofdark-matter particlemassmχ fortheEFT modelandina gridofmediatormassmmedandmχ forthesimplifiedmodel.

Majorsourcesofsystematicuncertaintyareuncertaintiesinthe modellingoflarge-R jet observables, whichhavea 5–13%impact on the expected background and signal yields, and the energy scale ofthe narrow jets, which contribute a 1–5% uncertaintyto theexpectedyields.Othersources ofuncertaintyincludetheoret- ical uncertainties in the simulated event samplesused to model the background processes (1–10%), parton distribution functions (10–15%),andlepton reconstructionandidentiﬁcationeﬃciencies (upto2%).

Aproﬁle-likelihoodﬁt[57]totheE^miss_T (E^miss_T_,_noμ)distributionin thesignalregion(controlregions)isusedtoconstrainthe W boson, Z boson,andtt¯ backgroundsandextractthesignalstrength,

μ,foreachmodelasanoverallnormalizationfactorforthesignal prediction.Besidesthesignalstrength,threeoverallnormalization factors for the W boson, Z boson, and tt¯ backgrounds are parameters in the ﬁt. The diboson and single-top backgrounds are estimatedfromsimulation, andthe multijetbackgroundis negligible.The likelihoodfunctionisdeﬁnedastheproductofPoisson distributions overall binsin E^miss_T and E_T^miss_,_noμ,andthelikelihood issimultaneouslymaximizedoverthesignalandcontrolregions.

Variationsof theexpectedsignal andbackgroundto allow for theirsystematicuncertaintiesaredescribedwithnuisanceparame- tersconstrainedbyGaussianprobabilitydistributionfunctions,and correlations across signal and background processes and regions aretakenintoaccount.

Table 1

Predictedandobservednumberofeventsinthesignalre- gion.Theyieldsanduncertaintiesofthebackgroundsare shownaftertheproﬁle-likelihoodﬁttothedataunderthe background-onlyhypothesis.Forcomparison,theexpected yieldinthe V Vχ χ EFTmodelwith M=600 GeV and mχ=^{500 GeV is}¹⁰.1±⁰.4 events.

Process Events

Z+^jets ⁵⁴⁴±³³

W +^jets ²⁷⁵±²⁴

tt¯and single-top 211±19

Diboson 89±12

Total background 1120±47

Data 1121

Table 2

Background normalization factors relative to the initial theoreticalprediction,extractedfromtheproﬁle-likelihood ﬁtunderthebackground-onlyhypothesis.

Process Normalization factor

Z+^jets ¹.01±⁰.16

W +^jets ⁰.90±⁰.16

tt¯ 0.91±⁰.18

A background-only (μ =⁰⁾ ^fit, ^shows ^no ^deviation ^from ^SM predictions, and Figs. 3 and 4 show kinematic distributions af- tertheprofile-likelihoodfit.Thefloatingbackground-normalization parameters are consistent withunity within one standard deviation.Tables 1 and 2showtheexpectedeventyieldsafterapplying the signal selection and the backgroundnormalization scale factors, respectively.Thevaluesinthesetablesareestimatedforthe background-onlyhypothesis.

Upperlimitsat95% confidencelevel(C.L.)on μ arecalculated usingtheCLsmethod[58].FortheV Vχ χ EFTmodel,theselim- its aretranslated intoconstraintsonthemassscale, M.Fig. 5(a) showsthelimitonthemassscale,M,intheEFTmodel,asafunc- tion of mχ.Fig. 5(b)shows the limitson the signal strength, μ, for a vector-mediatedsimplified model generatedwith couplings gSM=⁰.25 and gDM=^{1 in}^the^planeôf^mχ andm_med.

In conclusion, thisLetter reportsATLAS limits on dark-matter production in events with a hadronically decaying W or Z boson and large missing transverse momentum. These limits from 3.2 fb⁻¹ of13 TeVppcollisionsattheLHCimproveonearlierAT- LAS results.No statisticallysigniﬁcant excessisobservedoverthe StandardModelprediction.

Acknowledgements

We thank CERN forthe very successful operation ofthe LHC, as well asthe supportstaff fromour institutions withoutwhom ATLAScouldnotbeoperatedeﬃciently.

WeacknowledgethesupportofANPCyT,Argentina;YerPhI,Ar- menia;ARC,Australia;BMWFWandFWF,Austria; ANAS,Azerbai- jan; SSTC,Belarus;CNPqandFAPESP,Brazil; NSERC,NRCandCFI, Canada;CERN; CONICYT,Chile;CAS,MOSTandNSFC,China;COL- CIENCIAS, Colombia; MSMT CR, MPO CR and VSC CR, Czech Re- public; DNRFand DNSRC, Denmark;IN2P3-CNRS, CEA-DSM/IRFU, France; GNSF, Georgia; BMBF, HGF, and MPG, Germany; GSRT, Greece; RGC, Hong Kong SAR, China; ISF, I-CORE and Benoziyo Center, Israel; INFN, Italy; MEXT and JSPS, Japan; CNRST, Mo- rocco; FOM and NWO, Netherlands; RCN, Norway; MNiSW and NCN,Poland;FCT,Portugal;MNE/IFA,Romania;MESofRussiaand NRC KI,Russian Federation;JINR;MESTD,Serbia; MSSR,Slovakia;

ARRS andMIZŠ,Slovenia; DST/NRF,South Africa; MINECO,Spain;

SRCandWallenberg Foundation,Sweden;SERI,SNSFandCantons