Contents
\addbibresource

Acknowledgements.bib \addbibresourceEXOT-2016-35-PAPER.bib \addbibresourceATLAS.bib \addbibresourceATLAS-errata.bib \addbibresourceCMS.bib \addbibresourceConfNotes.bib \addbibresourcePubNotes.bib \AtlasTitleSearch for pair- and single-production of vector-like quarks in final states with at least one boson decaying into a pair of electrons or muons in collision data collected with the ATLAS detector at \PreprintIdNumberCERN-EP-2018-145 \AtlasJournalPhys. Rev. D \AtlasAbstract A search for vector-like quarks is presented, which targets their decay into a boson and a third-generation Standard Model quark. In the case of a vector-like quark () with charge (), the decay searched for is (). Data for this analysis were taken during 2015 and 2016 with the ATLAS detector at the Large Hadron Collider and correspond to \lumi of collisions at . The final state used is characterized by the presence of a boson with high transverse momentum, which is reconstructed from a pair of opposite-sign same-flavor leptons, as well as -tagged jets. Pair- and single-production of vector-like quarks are both taken into account and are each searched for using optimized dileptonic exclusive and trileptonic inclusive event selections. In these selections, the high scalar sum of jet transverse momenta, the presence of high-transverse-momentum large-radius jets, as well as—in the case of the single-production selections—the presence of forward jets are used. No significant excess over the background-only hypothesis is found and exclusion limits at 95% confidence level allow masses of vector-like quarks of () and () in the singlet (doublet) model. In the case of 100% branching ratio for (), the limits are (). Limits at 95% confidence level are also set on the coupling to Standard Model quarks for given vector-like quark masses.

1 Introduction

IntheStandardModel(SM),theelectromagneticandweakinteractionsarisefromagaugesymmetrythatisspontaneouslybrokenbytheEnglert–Brout–Higgsmechanism.Measurementsatcolliderexperimentsaresofarconsistentwithitspredictions.However,itisbelievedtobeonlyalow-energyapproximationofamorefundamentaltheorybecauseseveralquestionsremainunansweredintheSM.Forexample,itcannotexplainthematter–antimatterasymmetryintheuniverseandtheoriginofdarkmatter.WhentheSMisextrapolatedtohighenergies,fine-tuningisrequiredduetodivergentcorrectionstotheHiggsbosonself-energy [Naturalness].Solutionstothisso-called“hierarchyproblem”areproposedinseveralbeyond-the-StandardModel(BSM)theories,whichcanbeconsideredafirststeptowardsamorefundamentaltheoryofparticlephysics.Sincealargecontributiontothefine-tuningoriginatesfromtop-quarkloopcorrections,thehierarchyproblemcanbereducedinmodelspredictingtop-quarkpartnersthatmitigatethetopquark’scontribution:whileascalartop-quarkpartnerappearsinsupersymmetryasthebosonicsuperpartnerofthetopquark,fermionictop-quarkpartnersappearintheorieswithanewbrokenglobalsymmetry,inwhichtheHiggsbosonisinterpretedasapseudoNambu–Goldstoneboson [StrongEWSB],forexampleinLittleHiggs [LittleHiggs, LittleHiggsRev]andCompositeHiggs [CompHiggs1, CompHiggs2]models.Inthesemodels,thenewsymmetrycorrespondstoanewstronginteraction,whoseboundstatesincludevector-likequarks(VLQ).Thesearecolor-tripletspin-fermions,butincontrasttothechiralSMquarkstheirleft-andright-handedcomponentshavethesamepropertiesundertransformations.OnlyalimitedsetofpossibilitiesexistforthequantumnumbersoftheVLQsifgaugeinvarianceisrequiredtobepreserved [delAguila, delAguila:2000rc].Theirelectricchargecouldbe(quark),(quark),(quark)or(quark),whereistheelementarycharge,andtheycouldappearinelectroweaksinglets,()or(),electroweakdoublets,( ),( ),or( ),orelectroweaktriplets,( )or( ).Thispaperfocusessolelyonthesearchforandquarks,whichcouldcoupletoSMquarksbymixing [VLQmixing].AlthoughcouplingsofVLQstofirst-andsecond-generationSMquarksarenotexcluded [Atre:2008iu, Atre:2011ae],thispapersearchesforVLQsthatcoupleexclusivelytothird-generationSMquarks.Thecouplingsofandquarkscanbedescribedintermsofand [Aguilar-Saavedra:2013qpa],whereandarethemixingangleswiththetopquarkandthe-quark,respectively,ortheycanbedescribedintermsofgeneralizedcouplingsandoftheorquarktothird-generationSMquarks [Buchkremer:2013bha, Matsedonskyi:2014mna].SearchstrategiesforVLQshavebeenproposed [ContinoServant, JA_TP, TP_guide, Aguilar-Saavedra:2013qpa, Backovic:2015bca]thatfocuseitheronthesearchforVLQpairproductionviathestronginteractionoronsingleproductionviatheelectroweakinteraction.Thedecayofandquarkscaneitherhappenviathechargedcurrent,i.e. and,111Throughoutthisdocument,decaysthatarewritteninashortform,forexampleor,alsorefertothecorrespondingantiparticledecays,i.e. ,andareunderstoodtoincludetheproperbosonchargeandantifermionnotation,i.e. and.orviaflavor-changingneutralcurrents [delAguila2],i.e. ,,,and.Decaysincludingnon-SMparticlesarenotexcluded [Chala:2017xgc],butarenotconsideredinthispaper,sothatforandquarksthebranchingratios(BR)tothethreedecaymodesadduptounity.Whilethecrosssectionforpairproductionisgivenbyquantumchromodynamics,thesingle-productioncrosssectionexplicitlydependsonthecouplingoftheVLQtoSMquarks.

Pair-productionchannels Single-productionchannels
Dileptonwith\ljet Dileptonwith\ljets Trilepton Dilepton Trilepton
(PP 2 0-1J) (PP 2 2J) (PP 3) (SP 2) (SP 3)
Leptons
\btagged jets
\Ljets (top-tagged)
Forwardjets
\ptll
Additionaloptimizedkinematicrequirementsforeachchannel
Table 1: OverviewoftherequirementsusedineachchanneltosearchforpairandsingleproductionofVLQs.

TheATLASandCMSCollaborationshavesearchedforpairproductionofandquarksthatdecayintothird-generationquarksincollisionsat [Aad:2015kqa, Aad:2015gdg, Aad:2014efa, Aad:2015mba, Khachatryan:2015gza, Khachatryan:2015oba]inallthreepossibledecaymodesofeachoftheVLQs.Currentsearchesathaveusedsingle-leptonfinalstatestosearchforthedecaywiththebosondecayinginvisibly [Aaboud:2017qpr, Aaboud:2018xuw], [Aaboud:2017zfn, Sirunyan:2017pks], [Aaboud:2018xuw],and [Aaboud:2017zfn, WtX],aswellasgeneralsingle-leptonfinalstateswithboostedandHiggsbosons [Sirunyan:2017usq].TheCMSCollaborationhasalsosearchedforpairproductionofandquarksinacombinationofsingle-leptonfinalstates,dileptonfinalstateswiththesameelectricchargeandtrileptonfinalstates [Sirunyan:2018omb]at.Thesesearcheshavesetupperlimitsat95%confidencelevel(CL)ontheVLQpair-productioncrosssection,alsointerpretedaslowerlimitsontheVLQmass,,dependingontheVLQBRsassumed.Themoststringentlimitsinthecaseoftheandsingletsare1.20 \TeV[Sirunyan:2018omb]and1.17 \TeV[WtX, Sirunyan:2018omb],respectively.Inthecaseof100%BRsoftoandto,themoststringentlimitsare1.30 \TeV[Sirunyan:2018omb]and0.96 \TeV[Sirunyan:2018omb],respectively.Thesearchesataresignificantlymoresensitivethanthesearchesatduetothelargerexpectedpair-productioncrosssectionsatthehighercenter-of-massenergy.Thispaperincludessearchesforpair-producedVLQatinfinalstateswithmorethanoneleptonwhichareparticularlysensitivetothedecaysand.Atlarge,thecrosssectionforthesingleproductionofVLQsmaybelargerthanthepair-productioncrosssectionbecauseofthelargeravailablephasespace,eventhoughsingleproductionismediatedbytheweakinteraction.However,thecomparisonofsingle-andpair-productioncrosssectionsdependsontheassumedcouplingtotheSMquarks.Singleproductionwassearchedforat [Aad:2015voa, Aad:2014efa, Aad:2016qpo]bytheATLASandCMSCollaborations.At,theCMSCollaborationhassearchedforthedecays [Sirunyan:2017tfc], [Sirunyan:2016ipo, Khachatryan:2016vph], [Sirunyan:2017ynj, Sirunyan:2017ezy], [Sirunyan:2018fjh],and [Sirunyan:2017ezy].Inthesesearches,upperlimitsweresetonthesingle-productioncrosssection,whichwerealsointerpretedasupperlimitsonthecouplingtoSMquarksasafunctionof.Similarlytothecaseofpairproduction,theexpectedsingle-productioncrosssectionsaremuchlargeratthanat,sothatthesearchesatthehighercenter-of-massenergyaremoresensitive.Searchesforsingle-VLQproductionatwerenotperformedbeforebytheATLASCollaboration.AsinthesearchforVLQpairproduction,finalstateswithmorethanoneleptonareused,whichareparticularlysensitivetothedecay.

[]
[]
[]

Figure 1: Sketchesoftheprocessessearchedforinthepair-productionchannelsin(a)dileptonfinalstateswithatmostone\ljet (PP 2 0-1J),(b)dileptonfinalstateswithatleasttwo\ljets (PP 2 2J),and(c)finalstateswithatleastthreeleptons(PP 3),andsketchesoftheprocessessearchedforinthesingle-productionchannelsin(d)thedileptonfinalstate(SP 2),and(e)finalstateswithatleastthreeleptons(SP 3).

Theanalysiswasperformedwithdatacollectedincollisionsat,searchingforthepairproductionofandquarksandforthesingleproductionofquarksinfinalstateswithatleastoneboson.Inthecaseofsingleproduction,thequarkishenceexpectedtodecayinto.Inthecaseofpairproduction,thesearchtargetsonlyoneVLQdecayintoabosonandathird-generationquarkexplicitly,sothatitisparticularlysensitivetoalldecaysthatincludeatleastonebosoninthefinalstate,i.e. notonlyand,butalso,,,and.Theoverallanalysisstrategyisbasedonasearchthatwasperformedwithdata [Aad:2014efa],whichexploitedtheleptonicbosondecaysand.Severalimprovementshavebeenmade,inparticularaddingnewchannelsandoptimizingtheanalysisforthehigherandalargerdataset.Fiveanalysischannelsaredefined;threeforthesearchforandpair-production,andtwoforthesearchforsingle-production,asshownin\Tab1.Aneventpreselectionthatiscommontoallchannelsisused,inwhichalleventsarerequiredtoincludeabosoncandidate,reconstructedfromtwosame-flavorleptons(,)withoppositeelectriccharge.Theeventselectionineachchannelwasthenoptimizedforaparticularfinalstate,asshownin\Fig1.First,thesearchesweresplitintopair-andsingle-productioncategoriesandthenfurtherintodileptonchannels—requiringnoleptoninadditiontotheleptonsthatareusedtoreconstructthebosoncandidate—andtrileptonchannels,inwhichatleastthreeleptonsarerequired.SincetheVLQsareassumedtodecayintothird-generationSMquarks,thepresenceof-taggedjetsisexploitedinordertodiscriminatethesignalfromSMbackgroundprocesses.Sincethesignalprocessincludeshigh-energyhadronicallydecayingmassiveresonances,\ljets areusedinthedileptonchannels,furtherenhancingthesensitivityofthesearch.Inthedileptonsingle-productionchannel,top-taggingisusedinordertoidentify\ljets originatingfromthehadronicdecaysofhigh-energytopquarks.Inbothsingle-productionchannels,thepresenceofaforwardjetfromthe-channelproductionisusedtosuppressthebackground.Duetothelargeexpectedvaluesof,thetransversemomentum222ATLASusesaright-handedcoordinatesystemwithitsoriginatthenominalinteractionpoint(IP)inthecenterofthedetectorandthe-axisalongthebeampipe.The-axispointsfromtheIPtothecenteroftheLHCring,andthe-axispointsupwards.Cylindricalcoordinatesareusedinthetransverseplane,beingtheazimuthalanglearoundthe-axis.Thepseudorapidityisdefinedintermsofthepolarangleas.Angulardistanceismeasuredinunitsof.Thetransversemomentumisdefinedas,andthetransverseenergy,\et,isdefinedanalogously.oftheboson,\ptll,isexpectedtobemuchlargerinsignalthaninbackgroundevents.Morerequirements,inparticularontheeventkinematics,wereoptimizedineachchannel,asdiscussedin\Sect5.Inthefollowing,thethreepair-productionchannelsarereferredtoasthedileptonchannelwithatmostone\ljet (PP 2 0-1J),thedileptonchannelwithatleasttwo\ljets (PP 2 2J),andthetrileptonchannel(PP 3).Thetwosingle-productionchannelsarereferredtoasthedileptonchannel(SP 2),andthetrileptonchannel(SP 3).

2 TheATLASdetector

TheATLASdetector [PERF-2007-01]attheLHCcoversnearlytheentiresolidanglearoundthecollisionpoint.Itconsistsofaninnertrackingdetectorsurroundedbyathinsuperconductingsolenoid,electromagneticandhadroniccalorimeters,andamuonspectrometerincorporatingthreelargesuperconductingtoroidalmagnets.Theinner-detectorsystem(ID)isimmersedinaaxialmagneticfieldandprovidescharged-particletrackingintherange.Thehigh-granularitysiliconpixeldetectorcoversthevertexregionandtypicallyprovidesfourmeasurementspertrack,thefirsthitbeingnormallyintheinnermostlayer,theinsertableB-layer [ATLAS-TDR-19].Itisfollowedbythesiliconmicrostriptrackerwhichusuallyprovidesfourtwo-dimensionalmeasurementpointspertrack.Thesesilicondetectorsarecomplementedbythetransitionradiationtracker,whichenablesradiallyextendedtrackreconstructionupto.Thetransitionradiationtrackeralsoprovideselectronidentificationinformationbasedonthefractionofhits(typically30intotal)aboveahigherenergy-depositthresholdcorrespondingtotransitionradiation.Thecalorimetersystemcoversthepseudorapidityrange.Withintheregion,electromagneticcalorimetryisprovidedbybarrelandendcaphigh-granularitylead/liquid-argon(LAr)electromagneticcalorimeters,withanadditionalthinLArpresamplercovering,tocorrectforenergylossinmaterialupstreamofthecalorimeters.Hadroniccalorimetryisprovidedbythesteel/scintillator-tilecalorimeter,segmentedintothreebarrelstructureswithin,andtwocopper/LArhadronicendcapcalorimeters.Thesolidanglecoverageiscompletedwithforwardcopper/LArandtungsten/LArcalorimetermodulesoptimizedforelectromagneticandhadronicmeasurementsrespectively.Themuonspectrometer(MS)comprisesseparatetriggerandhigh-precisiontrackingchambersmeasuringthedeflectionofmuonsinamagneticfieldgeneratedbysuperconductingair-coretoroidalmagnets.Thefieldintegralofthetoroidalmagnetsrangesbetweenandacrossmostofthedetector.Asetofprecisionchamberscoverstheregionwiththreelayersofmonitoreddrifttubes,complementedbycathodestripchambersintheforwardregion,wherethebackgroundishighest.Themuontriggersystemcoverstherangewithresistiveplatechambersinthebarrel,andthingapchambersintheendcapregions.Atwo-leveltriggersystemisusedinordertoselectinterestingevents [Aaboud:2016leb].Thefirst-leveltriggerisimplementedinhardwareandusesasubsetofdetectorinformationtoreducetheeventratetoadesignvalueofatmost.Thisisfollowedbyasoftware-basedtriggerwhichreducestheeventratetoabout.

3 DataandMonteCarlosamples

Forthissearch,collisiondatacollectedwiththeATLASdetectorduring2015and2016atwereused,correspondingtoanintegratedluminosityof\lumi.Onlyeventstakenduringstablebeamconditions,andforwhichallrelevantcomponentsofthedetectorwereoperational,areconsidered.Eventsarerequiredtohaveaprimaryvertexwithatleasttwotrackswithaminimum\pt of400 .Ifseveralsuchverticesexists,thevertexwiththehighestischosenasthehard-scattervertex [ATL-PHYS-PUB-2015-026].Eventsarerejectediftheysatisfythecriteria [ATLAS-CONF-2015-029]designedtorejectbeam-inducedbackgroundandbackgroundsfromcosmic-rayshowersandcalorimeternoise.Severalsingle-leptontriggerswithdifferent\pt thresholdswereusedforelectronsandmuonsdependingonthedata-takingperiod.Fordatacollectedin2015,thethresholdsare24,60and120 \GeV forelectronsand20and50 \GeV formuons,whereleptonisolationrequirementsareappliedtothelowest-\pt triggerstoreducetheirrate.Forthehighest-\pt electrontrigger,theidentificationcriteriaarerelaxed.Fordata-takingin2016,thethresholdswereraisedslightlyto26,60and140 \GeV forelectronsand26and50 \GeV formuons.Themainsourcesofbackgroundinthissearchare\zjets and\ttbar productioninthecaseofthedileptonchannelsanddiboson(,,)andproductioninthecaseofthetrileptonchannels,whereisdominatedby\ttbar productionwithassociatedvectorbosons(,or)butalsoincludesandproduction.Smallersourcesofbackgroundarealsoconsidered,333Inthefiguresinthispaperthesmallerbackgroundsaregroupedtogetherandaredenotedby“Other.”whichincludesingle-topandtribosonproduction(,,,).Thebackgroundcontributionfromproductionwasfoundtobenegligibleandisnotconsideredinthissearch.Forallbackgroundandsignalprocesses,MonteCarlo(MC)samplesweregeneratedandthedetectorresponsewassimulatedinGEANT4 [Agostinelli:2002hh]withafullmodeloftheATLASdetector [SOFT-2010-01],unlessstatedotherwise.Thesimulationsincludethecontributionsfromadditionalcollisionsinthesameoranadjacentbunchcrossing(pileup).Correctionsfortriggerandobject-identificationefficiencies,andfor-taggingmisidentificationefficiencies,aswellasforenergyandmomentumscalesandresolutionsoftheobjectswereappliedtothesimulatedsamples,basedonthedifferencesobservedbetweendataandMCsamplesinreferenceprocesses.Asummaryofthebackgroundsamplesusedinthispaperisshownin\Tab2.The\zjets processwassimulatedwith\SHERPAV2.2.1 [Gleisberg:2008ta, Hoeche:2009rj, Gleisberg:2008fv, Schumann:2007mg]usingtheNNPDF3.0 [Ball:2014uwa]next-to-next-to-leading-order(NNLO)setofpartondistributionfunctions(PDFs),andnormalizedtotheNNLOcrosssectioninQCD444TheorderinperturbationtheoryreferstoQCDthroughoutthispaper.calculatedwithFEWZ [Anastasiou:2003ds]andtheMSTW2008 [Martin:2009iq, Martin:2009bu, Martin:2010db]NNLOPDFset.The\ttbar processwassimulatedwiththePOWHEGmethod [Nason:2004rx, Frixione:2007vw]implementedin\POWHEGBOXVv2 [Alioli:2010xd, Campbell:2014kua]usingtheNNPDF3.0NNLOPDFset.\POWHEGBOXwasinterfacedwith\PYTHIAV8 [Sjostrand:2007gs]withtheA14setoftunedparameters [ATL-PHYS-PUB-2014-021]andtheNNPDF2.3LOPDFset [Ball:2012cx]forpartonshoweringandhadronization.Theparameter555TheparametercontrolsthetransversemomentumofthefirstadditionalgluonemissionbeyondtheBornconfiguration.Themaineffectofchoosingistoregulatethehigh-\pT emissionagainstwhichthe\ttbar systemrecoils.in\POWHEGBOXwassetto [ATL-PHYS-PUB-2017-007],where.ThesamplewasnormalizedtotheNNLOcrosssectionincludingresummationofnext-to-next-to-leadinglogarithmic(NNLL)softgluontermswithTOP++ [Czakon:2011xx, Czakon:2013goa, Beneke:2011mq, Cacciari:2011hy, Czakon:2012pz, Czakon:2012zr, Baernreuther:2012ws].ThePDFanduncertaintieswerecalculatedusingthePDF4LHCprescription [Botje:2011sn]withtheMSTW2008NNLO,CT10NNLO [Lai:2010vv, Gao:2013xoa]andNNPDF2.35fFFNPDFsets,addedinquadraturetothescaleuncertainty.Thedibosonprocessesweresimulatedwith\SHERPAV2.2.1foruptooneadditionalpartonatnext-to-leadingorder(NLO)anduptothreeadditionalpartonsatleadingorder(LO)usingComix [Gleisberg:2008fv]andOpenLoops [Cascioli:2011va],andmergedwiththe\SHERPA partonshower [Schumann:2007mg]accordingtotheME+PS@NLOprescripton [Hoeche:2012yf].TheNNPDF3.0NNLOPDFsetwasusedandthesampleswerenormalizedtotheNLOcrosssectionscalculatedwith\SHERPA.Theprocessesweresimulatedwith\MGMCatNLO[Alwall:2014hca]usingtheNNPDF3.0NLOPDFset.\MGMCatNLOwasinterfacedwith\PYTHIAV8withtheA14setoftunedparametersandtheNNPDF2.3LOPDFsetforpartonshoweringandhadronization.ThesampleswerenormalizedtotheNLOcrosssectioncalculatedwith\MGMCatNLO.Thesingle-topprocessesweresimulatedwith\POWHEGBOXVv1 [Alioli:2009je, Re:2010bp]usingtheCT10PDFset.\POWHEGBOXwasinterfacedto\PYTHIAV6 [Sjostrand:2006za]withthePerugia2012 [Skands:2010ak]setoftunedparametersandtheCTEQ6L1PDFset [Pumplin:2002vw].Thesingle-topsampleswerenormalizedtoNLOcrosssectionswithadditionalNNLLsoftgluonterms [Kidonakis:2011wy, Kidonakis:2010tc, Kidonakis:2010ux].Thetribosonprocessesweresimulatedusing\SHERPAV2.1usingtheCT10PDFset,andnormalizedtotheNLOcrosssectionscalculatedwith\SHERPA.Theandprocessesweresimulatedwith\MADGRAPHV5and\PYTHIAV8usingtheNNPDF2.3LOPDFsetandtheA14setoftunedparameters,andwerenormalizedtotheNLOcrosssectioncalculatedwith\MGMCatNLO.AdditionalMCsamplesweregeneratedfortheevaluationofsystematicuncertaintiesduetothechoiceoffactorizationandrenormalizationscales,generator,andpartonshowerprogramforthe\zjets,\ttbar,anddibosonbackgroundprocesses.Thesesamplesaredescribedin\Sect6.ThepairproductionofVLQswassimulatedatLOwith\PROTOS[PROTOS]usingtheNNPDF2.3LOPDFset.\PROTOS wasinterfacedto\PYTHIAV8withtheA14setoftunedparameters.Sampleswereproducedforintherangeof500to1400 \GeV.Stepsof50 \GeV wereusedintherangefrom700to1200 \GeV,andstepsof100 \GeV otherwise.Thesamplesweregeneratedinthesingletmodelsforandquarks,butsamplesatof700,900and1200 \GeV werealsogeneratedinthe( )doubletmodelinordertotestkinematicdifferencesbetweensingletanddoubletmodels.Inthesingletmodels,theBRsareindependentofthemixinganglesbetweenVLQandSMquarksforsmallvaluesofthemixinganglesandhenceonlyafunctionof.Withthisassumption,forlarge,theBRsapproachtherelativeproportionsof50:25:25forthe::decaymodesinthesingletmodelforthequarkaswellasforthequark.Inthe( )doubletand( )doubletmodels,theBRsapproachtherelativeproportionsof50:50forthe:decaysofthequarkandquark,respectively.Thesameholdsforthe( )doubletmodelifthetopquarkmixesmuchmorestronglywithitsVLQpartnerthanthebottomquark,anaturalscenariofortheSMYukawacouplings [JA_TP].However,kinematicdifferencesmayexistbetweenthesingletanddoubletmodels.Thesamplesgeneratedforthe( )doubletwereusedtoverifythatsuchkinematicdifferenceshaveasmallimpactontheanalysis,andthereforethedifferencebetweenthetwocasesisonlyachangeintheBRs.Thus,thesingletmodelsampleswerealsousedforthedoubletcase,reweightingtheyieldsforeachdecaymodetoobtaintheexpectedobservablesforanygivenBR.Thepair-productioncrosssectionswerecalculatedwithTOP++atNNLO+NNLLusingtheMSTW2008NNLOPDFset.Thesingleproductionofquarkswassimulatedusing\MADGRAPHV5withthe“VLQ”UFOmodel [UFO],whichimplementstheLagrangiandescribedinRef [Buchkremer:2013bha],usingtheNNPDF2.3LOPDFset.\MADGRAPH wasinterfacedto\PYTHIAV8withtheA14setoftunedparameters.Onlythedecaywasconsidered.Samplesweregeneratedwithaquarkproducedviaandalsoviainteractions.Sinceproductionviatheinteractionissuppressedduetotherequiredtopquarkintheinitialstate,single-VLQproductionreferstoproductionviatheinteractionintheremainderofthispaper,unlessstatedotherwise.Samplesweregeneratedforintherangefrom700to2000 \GeV withabenchmarkcouplingoffortheandinteractions,respectively.Additionalsamplesweregeneratedwithalternativevaluesofandinordertostudytheeffectofavarying-quarkwidthonkinematicdistributions.Thesingle-productioncrosssectionswerecalculated [Matsedonskyi:2014mna]atNLOandinnarrow-widthapproximationfor,withthecouplingdefinedinRef. [Matsedonskyi:2014mna]andcorrespondingtouptonumericalconstants.Inordertopredictthecrosssectionfordifferentvaluesof,theyaremultipliedby.Itwasshowninthecontextofthisanalysisthatthechiralityofthecouplinghasanegligibleimpactonthesensitivityoftheanalysisandhenceistakenasthesuminquadratureoftheleft-andright-handedcouplingsand,i.e. .Thecrosssectionisadditionallycorrectedforwidtheffectscalculatedwith\MADGRAPHV5,assumingthattheratioofNLOandLOcrosssectionsremainsapproximatelythesameforanon-vanishing-quarkwidth.ThecrosssectionisthenmultipliedbytheBRtointhesingletmodel,whichis%intherangeofVLQmassesinvestigatedinthisanalysis.Thebenchmarkcouplingofcorrespondstoacouplingofthequarktotheboson,.

Generator Showerprogram PDFset(ME) Crosssection
andtune
\zjets \SHERPAV2.2.1 \SHERPAV2.2.1 NNPDF3.0NNLO NNLO
\ttbar \POWHEGBOXVv2 \PYTHIAV8,A14 NNPDF3.0NNLO NNLO+NNLL
Diboson \SHERPAV2.2.1 \SHERPAV2.2.1 NNPDF3.0NNLO NLO
(/) \MGMCatNLO \PYTHIAV8,A14 NNPDF3.0NLO NLO
\MADGRAPHV5 \PYTHIAV8,A14 NNPDF2.3LO NLO
\MADGRAPHV5 \PYTHIAV8,A14 NNPDF2.3LO NLO
Singletop \POWHEGBOXVv1 \PYTHIA8, CT10 NLO+NNLL
Perugia2012
Triboson \SHERPAV2.1 \SHERPAV2.1 CT10 NLO
Table 2: ListofbackgroundMonteCarlosamplesused,givinginformationaboutthematrix-elementgenerator,thepartonshowerprogramtowhichitisinterfacedanditssetoftunedparameters(“tune”,ifapplicable),thePDFsetsusedinthematrixelement(ME),andtheorderinQCDofthecross-sectioncalculation.

4 Objectreconstruction

Reconstructedelectrons,muonsandjetsareused.Jetsarereconstructedwiththeanti-algorithm [Cacciari:2008gp]witharadiusparameterof0.4(\sjets)andwithaparameterof1.0(\ljets).A-taggingalgorithmisappliedto\sjets,andatop-taggingalgorithmisappliedto\ljets.Moreover,missingtransversemomentum(\met)isusedforthedefinitionofonesignal-enrichedregionandonebackground-enrichedregion.Forelectrons,muonsandjets,anoverlap-removalprocedurebasedontheirproximityinspaceisused,asdescribedattheendofthissection.Electronsarereconstructed [ATLAS-CONF-2016-024]fromenergyclustersintheelectromagneticcalorimeterwithIDtracksmatchedtothem.Theirenergyiscalibrated [PERF-2013-05, ATL-PHYS-PUB-2016-015],andtheyarerequiredtofulfillthe“tightlikelihood”identificationcriteria [ATLAS-CONF-2016-024].Electronsarerequiredtohaveaminimumtransverseenergy,\et,ofatleast28 \GeV andtobewithinthefiducialregion,excludingthebarrel–endcaptransitionregion,.Electrontracksmustpointtotheprimaryvertex,whichisensuredbyrequiringthatthetrack’simpactparametersignificanceissmallerthan5,andthatissmallerthan0.5 mm,whereisthedistancealongthe-axisbetweentheprimaryvertexandthetrack’spointofclosestapproach.Inordertosuppressbackgroundfromelectronsoriginatingfromhadrondecaysandfromhadronsthataremisidentifiedaselectrons,anisolationcriterionisappliedthatrequiresthescalarsumofthe\pt ofthetrackswhichpointtotheprimaryvertexwithinaconearoundtheelectron(butexcludingitstrack)belessthan6%ofits\et.Avariableconesize [Rehermann:2010vq]ofisused.Muonsarereconstructed [PERF-2015-10]fromcombinedtracksintheMSandtheID.Theirtransversemomentum,\pt,iscalibrated [PERF-2015-10],andtheyarerequiredtofulfillthe“medium”identificationcriteria [PERF-2015-10].Muonsmusthaveaminimum\pt of28 \GeV andtheymustbewithinthefiducialregion.Muontracksmustpointtotheprimaryvertex,whichisensuredbyrequiringthatthetrack’simpactparametersignificanceissmallerthan3,andthatissmallerthan0.5 mm.Inordertosuppressbackgroundfrommuonsoriginatingfromhadrondecays,anisolationcriterionsimilartothatforelectronsisapplied:thescalarsumofthe\pt ofthetracksaroundthemuonwhichpointtotheprimaryvertex,excludingthemuontrack,mustbesmallerthan6%ofits\pt,usingavariableconesizeof.\Sjets arereconstructedfromtopologicalclustersofcalorimetercells [PERF-2014-07, ATL-PHYS-PUB-2015-036]withtheanti-algorithmusingFastJet [Cacciari:2011ma]witharadiusparameterof0.4.\Sjets arecalibratedtothejetenergyscale(JES)atparticlelevel [PERF-2016-04]andarerequiredtobewithinthefiducialvolume.\Sjets withmusthaveaminimum\pt of25 \GeV andforwardjets,,musthaveaminimum\pt of35 \GeV toreducecontributionsfrompileup.For\sjets withand,pileupcontributionsaresuppressedbytheuseofthejetvertextagger [PERF-2014-03].\Sjets withinare\btagged usingtheMV2c10algorithm [ATL-PHYS-PUB-2016-012],forwhichseveralbasic-tagging-algorithms [PERF-2012-04]arecombinedinaboosteddecisiontree.TheMV2c10algorithmisusedsuchthatitprovidesa-taggingefficiencyoffor-jets,666Jetsoriginatingfromthehadronizationofgluonsandlightquarks(-,-,-and-quarks)arecalledlightjetsinthisdocument.Jetsoriginatingfromthehadronizationof-quarksarecalled-jets.andarejectionfactoroffor-jetsandforotherlightjets,basedonsimulated\ttbar events.\Ljets arealsoreconstructedfromtopologicalclusterswiththeanti-algorithm,butwitharadiusparameterof1.0.Incontrasttothe\sjet calibration,thetopologicalclustersthatareusedasinputstothe\ljet reconstructiontakeintoaccountcorrectionsforthecalorimeter’sresponsetohadronsandothereffects [Barillari:2009zza].Contributionsto\ljets frompileupandtheunderlyingeventareremovedbyapplyingtrimming [Krohn:2009th]withparametersthatwereoptimizedforseparating\ljets thatoriginatefromhadronicdecaysofhigh-energymassiveresonances [PERF-2015-03, ATL-PHYS-PUB-2015-033, ATL-PHYS-PUB-2015-053]fromthosethatoriginatefrom-quarks,lightquarksorgluons.\Ljets arecalibratedtotheJESatparticlelevel [PERF-2012-02].Theyarerequiredtohaveaminimum\pt of200 \GeV andtobewithinthefiducialregion.Themassof\ljets iscalculatedfromacombinationofcalorimeterandtrackinginformation [ATLAS-CONF-2016-035].Itiscalibrated [ATLAS-CONF-2016-035]andrequiredtobeatleast50 \GeV,whichsuppressescontributionsfrom-jetsandlightjetsinfavorof\ljets thatoriginatefromhadronicdecaysofhigh-energybosons,bosons,Higgsbosonsandtopquarks.IntheSP 2 channel(\Sect5.4),top-taggingisusedtoidentifyhadronicdecaysofhigh-energytopquarks.Itisbasedonacombination [ATL-PHYS-PUB-2015-053]ofthe\ljet massandthe-subjettiness [Thaler:2010tr, Thaler:2011gf]ratio,calculatedinthe“winner-take-all”mode [Larkoski:2014uqa].Thistop-taggerprovidesanefficiencyofforhadronicallydecayingtopquarkswitha\pt ofatleast200 \GeV withavaryingbackgroundrejectionofatthatdecreasestoat,asestimatedwithsimulateddijetevents.Inordertoavoiddouble-countingoftracksorenergydepositsandinordertoimprovetheidentificationofthedifferentreconstructedobjects,asequentialoverlap-removalprocedureisused.Inthefirststep,electronsthatshareatrackwithamuonareremoved.Inthesecondstep,any\sjet isremovedthathasatoanelectronthatissmallerthan0.2,andinthethirdstep,electronsareremovediftheyarecloserthan0.4toanyremaining\sjet.Finally,\sjets thathaveatoamuonareremovediftheyhaveatmosttwoassociatedtrackswith,otherwisethemuonisremoved.\Sjets and\ljets arenotsubjecttoanoverlap-removalprocedure,becausetheanalysisstrategiesinallchannelsaredesignedsuchthattheenergydepositsin\ljets and\sjets arenotcountedtwice,asexplainedinthefollowinglines:inthetrileptonchannels,\ljets arenotused(\Sect5.3and5.5);inthedileptonpair-productionchannels,\ljets areonlyusedfortheclassificationofevents(\Sect5.1and5.2);inthedileptonsingle-productionchannel,\sjets areonlyusedfortheclassificationofevents,butnotforthecalculationofthediscriminatingvariable(\Sect5.4).Missingtransversemomentumisonlyusedforthereductionofthecontributionfrom\TTbar pair-productioninonesearchregionforsingle-production(\Sect5.4)andforthedefinitionofonebackground-enrichedregion(\Sect5.2),anditiscalculatedfromthevectorialsumofthetransversemomentaofreconstructedandcalibratedleptonsand\sjets[Aaboud:2018tkc],withtheoverlapbetweentheseobjectsremoved.ThecalculationalsoincludesthecontributionsfromtracksintheIDthatarematchedtotheprimaryvertexbutarenotassociatedwithanyofthereconstructedobjects.

5 Eventselectionandbackgroundcontrolregions

Fivedifferentchannelsareanalyzed,eachsearchingforeitherpairproductionorsingleproductionofVLQs,asintroducedin\Sect1andvisualizedin\Fig1.Ineachchannel,event-selectioncriteriawereoptimizedformaximumsensitivitytobenchmarkprocessesbystudyingexpected95%CLexclusionlimits.Inthepair-productionchannels,themassreachforandquarksinthesingletanddoubletmodelswasmaximized.WhilethesearchfocusesonthedecayofoneVLQtoabosonandathird-generationSMquark,ahighsensitivitytoallthree-and-quarkdecaymodesisensuredbychoosingthesebenchmarkmodels,becausethesecondVLQisnotonlyallowedtodecayintoaboson,butalsointoabosonoraHiggsbosoninassociationwithathird-generationSMquark.Inthesingle-productionchannels,thesensitivitytosingle--quarkproductionviatheexchangeofabosonwithwasoptimized.

[]
[]

Figure 2: Distributionsofthesumofallbackgroundprocesses(solidarea)andofbenchmarksignalprocesses(lines),basedonMCsimulationsafterpreselectionandrequiring:(a)thenumberofleptons,(b)thenumberof\btagged jets,(c)thenumberof\ljets,and(d)thenumberofforwardjets.Thesignalprocessesshownare-and-quarkpairproductioninthesingletmodelandsingle--quarkproductionwithacouplingof,eachwithamassof.Alldistributionsarenormalizedtounitarea.Thelastbincontainstheoverflow.

Apreselectioncommontothechannelswasusedasthebasisfortheseoptimizations.Thispreselectionrequiresthepresenceofabosoncandidatethatisconstructedfromtwoleptonswithopposite-signelectriccharge.Inallevents,atleasttwoleptonsofthesameflavorwithandwithopposite-signelectricchargearerequired.Outofallsuchleptonpairsinanevent,abosoncandidateisdefinedbythepairwithinvariantmassclosesttothemassoftheboson.Eventsinwhichthisinvariantmassislargerthan400 \GeV areremovedbecausetheyareveryunlikelytooccurinanyoftheconsideredsignalprocesses.Inaddition,atleasttwo\sjets withmustbepresent.IntheSP 2 channel,thislastcriterionisreplacedbyarequirementonthepresenceofatleastone\ljet withand.In\Fig2,normalizeddistributionsafterpreselectionareshownforthesumofallbackgroundprocesses,whichareestimatedfromMCsimulations,aswellasforbenchmarksignalmodelsforpairandsingleVLQproduction.In\Fig2,thedistributionofthenumberofleptonsisshown.Byselectingeventswithexactlytwoleptons,ahighsignalefficiencyisachieved.Ineventswithatleastthreeleptons,however,thesignal-to-backgroundratioissignificantlyimproved.Thesearchesforpairandsingleproductionarehencesplitintocomplementarydileptonandtrileptonchannels.Thedistributionofthenumberof\btagged jetsisshownin\Fig2.Ahighernumberof\btagged jetsischaracteristicofthesignalprocesses,andatleastoneortwo\btagged jetsarerequiredintheeventselection,dependingonthechannel.Thedistributionofthenumberof\ljets isshownin\Fig2.Signaleventsshowahighernumberof\ljets thanbackgroundevents,whichinsignalmostlyoriginatefromthehadronicdecaysofboostedtopquarks,bosons,bosonsorHiggsbosons.Thepresenceof\ljets isusedinthedileptonchannelstosuppressbackgroundsandhenceimprovethesensitivitytothesignal.Inordertoachieveahighsignalefficiency,inthepairproductioncase,twocomplementarydileptonchannelsaredefined,oneforeventswithatmostone\ljet andoneforeventswithatleasttwo\ljets.Inthetrileptonchannels,\ljet requirementsarenotusedbecausethepresenceofatleastthreeleptonssuppressesthebackgroundsefficiently.In\Fig2,theforward-jetmultiplicityisshown.Thesingle-productionprocessoftenfeaturesaforwardjetfrom-channelproduction.Thepresenceofaforwardjetishenceusedinthesingle-productionsearchestoseparatethesignalfromthebackground.TheeventselectioncriteriainthedifferentchannelsaredefinedinSections 5.15.5.Ineachchannel,thesesignalregions(SR)arecomplementedbyasetofcontrolregions(CR),whichareenrichedinthemainbackgroundprocesses.TheCRsareusedtocheckthemodelingofthebackgroundandtoimprovethebackgroundpredictionintheSRsbyacombinedfitofCRsandSRs(\Sect7).InthedesignoftheCRs,notonlyahighpurityoftherespectivebackgroundprocesseswasaimedfor,butalsoalargenumberofbackgroundevents,aswellaskinematicpropertiesofthebackgroundeventsthatresemblethoseoftheeventsintheSRs.EachCRwascheckedtoensurethatitwasnotsensitivetoanysignalprocess.AllSRsandCRsdefinedinthethreepair-productionchannels(Sections 5.15.3)areorthogonal,sothattheresultsinthesechannelscanbecombined(\Sect7).ThesameholdsforallSRsandCRsinthesingle-productionchannels(Sections 5.45.5),whicharealsocombined(\Sect7).Orthogonalityisnotensuredbetweenpair-andsingle-productionregions.However,single-productionchannelsincluderequirementsdesignedtosuppressthepair-productionsignalintheirSRs.

5.1 Searchstrategy:PP 2 0-1j

Twoorthogonalchannelsaredefinedforthepair-productionsearchindileptonfinalstates,onewithatleasttwo\ljets,describedin\Sect5.2(PP 2 2J),andonewithatmostone\ljet (PP 2 0-1J),describedinthissection.WhileinthePP 2 2J channelbackgroundprocessesarestronglysuppressed,thesignalefficiencyisalsoreducedsothatacomplementarychanneloptimizedforeventswithatmostone\ljet providesadditionalsensitivitytothesignal.ThedefinitionsoftheSRsinthePP 2 0-1J channelaresummarizedin\Tab3.TwoSRsaredefined,forwhichthepreselectionandthepresenceofexactlytwoleptonsarerequired.Themassofthebosoncandidate,builtfromthetwoleptons,\mll,mustbewithina10 \GeV windowaroundthebosonmass,.Atleasttwo\btagged jetsmustbepresent,whichstronglyreducesthebackgroundcontributionfromtheproductionofabosoninassociationwithlightjets.ThesensitivityofthechannelisimprovedbydefiningtwoSRs,oneforeventswithoutany\ljet andoneforeventswithexactlyone\ljet.SinceinthesignalprocessthebosonisproducedinthedecayofamassiveVLQ,the\pt ofthebosoncandidate,\ptll,isonaveragemuchlargerthaninthebackgroundprocesses,so\ptll isrequiredtobelargerthan250 \GeV forbothSRs.Moreover,thescalarsumofthetransversemomentaofall\sjets intheevent,\htj,isonaveragemuchlargerforsignaleventsthanforbackgroundevents,becausethequarksfromthedecaychainofthemassiveVLQresultinhigh-\pt jets.Therefore,the\htj distributionisusedinthestatisticalanalysis(\Sect7)tosearchforanexcessofdataoverthebackgroundprediction,withasignalexpectedtoresultinanexcessforlargevaluesof\htj.Inaddition,aminimum\htj valueof800 \GeV isrequiredforbothSRs.

CR +jetsCR 0-\ljet SR 1-\ljet SR
Preselection
leptons
and
\btagged jets
\ljet \ljets \ljet
Table 3: DefinitionofthecontrolandsignalregionsforthePP 2 0-1J channel.

Themainbackgroundprocessesarefrom+jetsproductioncontainingtwojetswhichoriginatefromthehadronizationof-quarksand\ttbar productionwithadileptonicfinalstate.Thebackgroundfrom\ttbar productionisstronglysuppressedbyrequiring\mll tobecloseto\mz.Inbothmainbackgroundprocesses,nohadronicallydecayingmassiveresonancesarepresent,sothattheSRwithexactlyone\ljet hasahighersignal-to-backgroundratiothantheSRwithouta\ljet.Thecontributionsfromallbackgroundprocessesarestronglyreducedbytherequirementson\ptll and\htj.Inordertovalidatethemodelingofthemainbackgroundprocesses,CRsaredefinedforthe+jetsand\ttbar processes.AsummaryoftheCRdefinitionsisgivenin\Tab3.The+jetsCRisdefinedbythesamecriteriaastheSRs,exceptforthe\ljets and\htj criteria.Eventswithno\ljets andeventswithexactlyone\ljet areconsideredtogetherand\htj isrequiredtobeintherange200–800 \GeV,ensuringthattheCRisalmostfreeofapotentialsignal.TheresultingCRsampleisexpectedtobe+jetsevents.The\ttbar CRisdefinedbyrequiringthesamepreselection,leptonmultiplicity,and\btagged-jetmultiplicitycriteriaasintheSRs.However,themassofthebosoncandidate,\mll,mustbeoutsideofa10 \GeV windowaroundthebosonmass,\mz.Inaddition,\mll isrequiredtobelargerthan50 \GeV,becauseeventswithlower\mll donotstemmainlyfrom\ttbar production,butfromDrell–Yanproductioninassociationwithjets.Alsointhe\ttbar CR,eventswithout\ljets andeventswithexactlyone\ljet areconsideredtogether.IncontrasttothedefinitionoftheSRs,the\pt ofthebosoncandidateisrequiredtobelessthan600 \GeV inordertoensurethattheCRdoesnotcontainsignalcontributionsfrompotentialVLQpairproductionwithtwoleptonsthatdonotstemfromthedecayofaboson,suchas.Morever,thelowerboundon\htj isloweredto200 \GeV inordertoincreasethenumberofeventsintheCRandtotestthemodelingofthefull\htj distribution.TheresultingCRsampleisexpectedtobe\ttbar events.

5.2 Searchstrategy:PP 2 2j

InadditiontothePP 2 0-1J channel,aseconddileptonchannelwasoptimizedforeventswithatleasttwo\ljets (PP 2 2J)inordertoexploitthepresenceofhighlyboosted,hadronicallydecayingmassiveresonancesinthesignalprocesses.Allsuch\ljets arerequiredtohavea\pt ofatleast200 \GeV andamassofatleast50 \GeV aftertrimming.Duetothemassrequirement,hadronicdecaysofboostedtopquarks,andof,,andHiggsbosonsareefficientlyselectedandjetsthatoriginatefromthehadronizationofhigh-\pt lightquarks,-quarksorgluonsaresuppressed.ThedefinitionoftheSRinthePP 2 2J channelissummarizedin\Tab4.ThesamerequirementsasinthePP 2 0-1J channelareimposed:thepreselection,thepresenceofexactlytwoleptonswith\mll withina10 \GeV windowaround\mz,andthepresenceofatleasttwo\btagged jets.Inaddition,atleasttwo\ljets arerequiredineachevent.Alsointhischannel,thelargeexpectedvaluesfor\ptll and\htj areexploitedtodiscriminatethesignalfromthebackgroundprocesses.Theoptimizedrequirementsareand.Inordertosearchforanexcessofdataoverthebackgroundprediction,theinvariantmassofthebosoncandidateandthehighest-\pt \btagged jet,,isusedasadiscriminatingvariable.Inthesearchfor\BBbar production,wouldshowaresonantstructurearoundifVLQswerepresent,becauseitoftencorrespondstothereconstructedmassoftheVLQ.Also,inthesearchfor\TTbar production,thisvariableshowsverygooddiscriminationbetweensignalandbackground,withthesignalresultinginlargervaluesofthanthebackground.

CR +jetsCR SR
Preselection
leptons
and
\btagged jets
\ljets
or
Table 4: DefinitionofthecontrolregionsandthesignalregionforthePP 2 2J channel.

Themainbackgroundprocessesare+jetsproductionwithtwojetsoriginatingfromthehadronizationof-quarks,and\ttbar productioninthedileptonicdecaymode.AsinthePP 2 0-1J channel,\ttbar productionisstronglysuppressedbyrequiring\mll tobeclosetothemassoftheboson,andthecontributionsfromallbackgroundprocessesaresignificantlyreducedbytherequirementson\ptll and\htj.Thecontributionsfrom+jetsproductionanddileptonic\ttbar decaysareefficientlyreducedbythepresenceoftwo\ljets,becausenomassivehadronicallydecayingresonanceispresentintheseprocesses.Forthetwomainbackgroundprocesses,+jetsand\ttbar production,CRsaredefined.AsummaryoftheCRdefinitionsisgivenin\Tab4.Similarlytothe\ttbar CRinthePP 2 0-1J channel,thedefinitionofthe\ttbar CRisbasedontherequirementthat\mll mustbeoutsidea10 \GeV windowaround\mz butmuststillfulfill.InordertosuppresspotentialsignalcontributionsintheCR,\ptll isrequiredtobesmallerthan600 \GeV.Therequirementon\htj isremoved,whichincreasesthenumberofeventsintheCR.Inaddition,\met isrequiredtobesmallerthan200 \GeV,whichreducespotentialsignalcontributionsfromVLQpairproductionwithtwoleptonsthatdonotstemfromthedecayofaboson,butforexamplefromthedecayofbosonsfromtheVLQdecaychain.Moreover,thebetweenthebosoncandidateandthehighest-\pt \ljet isrequiredtobesmallerthan2.0orlargerthan2.8,whichfurtherreducesthecontributionsfromapotentialsignalbecauseinsignaleventsthehighest-\pt \ljet andthebosoncandidatearetypicallynotback-to-backduetothepresenceofadditionalfinal-stateparticles.TheresultingCRsampleisexpectedtobe\ttbar events.TheCRforthe+jetsprocessisdefinedbythesamecriteriaasintheSR,buttherequirementon\htj isinvertedinordertoremovepotentialsignalcontributions,andtherequirementon\ptll isremovedinordertoincreasethenumberofeventsintheCR.TheresultingCRsampleisexpectedtobeonly+jetsevents,butalso\ttbar events.

5.3 Searchstrategy:PP 3

Thetrileptonpair-productionchannel(PP 3)issensitivetosignaleventsinwhichatleastoneleptonappearsinadditiontotheleptonsfromthebosondecaythatoriginatesfromor.AdditionalleptonscanoriginatefromthedecayoftheotherVLQ,suchasinor.In\TTbar production,anadditionalleptoncanalsooriginatefromthedecayitself,ifthetopquarkdecaysinto.ThedefinitionoftheSRissummarizedin\Tab5.Eventsmustpassthepreselection,andtheymusthaveatleastthreeleptonsincludingabosoncandidatewith\mll withina10 \GeV windowaround\mz.Onlyone\btagged jetisrequired,becausebackgroundcontributionsarealreadystronglyreducedbytherequirementofatleastoneadditionallepton.Relaxingthe-taggingrequirementcomparedtothedileptonchannelsimprovesthesensitivitytothesignalprocessesbecauseofthelargersignalefficiency.Asinthedileptonchannels,alargetransversemomentumofthebosoncandidateisrequired,.Inordertosearchforanexcessofdataoverthebackgroundprediction,thescalarsumofthe\sjet andleptontransversemomenta,\htjl,isused.Incontrasttotheuseof\htj inthePP 2 0-1J channel(\Sect5.1),theleptontransversemomentaareaddedtothediscriminatingvariable\htjl,whichexploitsthe\pt ofallleptonsinordertodiscriminatethesignalfromthebackgroundinadditiontotheuseof\ptll,whichisconstructedfromonlytwoleptons.

DibosonCR CR SR
Preselection
leptons
=0\btagged jets \btagged jets
Table 5: DefinitionofthecontrolregionsandthesignalregionforthePP 3 channel.

Themainbackgroundprocessesarediboson,inparticularand,production,andproduction(dominatedbyproduction),whichcanbothresultineventswiththreeleptons.Thedibosonbackgroundisstronglyreducedbythe-taggingrequirement,sothatonlydibosoneventswithadditional-jetsormis-taggedlightjetspasstheeventselection.Bothmainbackgroundsaresuppressedbytherequirementon\ptll,becauseinbackgroundeventsbosoncandidatesrarelyhavealargetransversemomentum.Forthetwomainbackgroundprocesses,dibosonandproduction,CRsaredefinedandsummarizedin\Tab5.ThedibosonCRisdefinedbythesamecriteriaastheSR,exceptforthe-taggingand\ptll requirements.No\btagged jetsareallowedinthedibosonCR,whichreducescontributionsfromproductionandfromapotentialVLQsignal.The\ptll requirementisremovedinordertofurtherincreasethenumberofdibosoneventsintheCR.TheresultingCRisexpectedtoconsistofdibosonevents,mainlyfromproduction.TheCRisdefinedbyinvertingonlythe\ptll requirement,whichremovescontributionsfromapotentialVLQsignal.TheresultingCRsampleisexpectedtoconsistmainlyofanddibosoneventsinsimilarproportions(and,respectively).

5.4 Searchstrategy:SP 2

Theproductionofasinglequarkresultsinasignaturewithfewerhigh-\pt objectsthanin\TTbar production.Asaresultitismoredifficulttoseparateitfromthebackground.However,aforward-jetfromthe-channelproductionisoftenpresent,whichcanbeexploitedtostronglyreducethecontributionsfrombackgroundprocesses.Thefinalstatefromthedecayofasinglewithaleptonicbosondecayconsistsofthetwoleptonsfromtheboson,aforwardjetandthedecayproductsofthetopquark.Whiletheleptonictop-quarkdecay,,isusedinthetrileptonsingle-productionchannel(SP 3),describedin\Sect5.5,thehadronicdecay,,isusedinthedileptonchannel(SP 2),describedinthissection.ThedefinitionoftheSRissummarizedin\Tab6.Eventsarerequiredtopassthepreselectionwiththeminimumrequirementoftwo\sjets replacedbythepresenceofatleastone\ljet.Eventsmusthaveexactlytwoleptonsthatformabosoncandidatewithaninvariantmasswithina10 \GeV windowaround\mz.Inthischannel,aminimum\pt ofthebosoncandidateisalsorequired,.Atleastone\btagged jetisrequiredintheevent.Althoughasecond-quarkfromgluonsplitting(\Fig1)ispresentinthesignal,onlyinafractionofsignaleventsisasecond\btagged jetfoundwithintheacceptanceoftheID.Thehadronicallydecayingtopquarkoriginatingfromthe-quarkdecayoftenhassuchalarge\pt thatthetop-quarkdecayproductsarecontainedwithinone\ljet.Top-taggingisusedtodiscriminate\ljets fromhadronictop-quarkdecaysinsingle--quarkproductionfromthemainbackgroundprocess,+jetsproduction,whichcanonlyfulfillthisrequirementifaquarkorgluonjetisfalselytop-tagged(mis-tags).Atleastoneforwardjetisrequiredineachevent,whichisacharacteristicpropertyofsingle--quarkproduction.Inordertosearchforanexcessofdataoverthebackgroundprediction,theinvariantmassofthebosoncandidateandthehighest-\pt top-tagged\ljet,,isused,which,ifVLQswerepresent,wouldshowaresonantstructurearound.Inordertofacilitatetheinterpretationofthesearchforsingle--quarkproduction,thepotentialsignalcontributionfrom\TTbar productionisreducedbyrequiring.Thisrequirementhasanefficiencyoffor\TTbar pair-productioninthesingletmodelinthemassrange800–1400 \GeV,whilemaintaininganefficiencyof90–95%forsingle--quarkproductionwithacrossthewholemassrangestudied.

0-\btagged-jetCR -\btagged-jetCR SR
Preselectionwith\ljet
leptons
\btagged jets \btagged jets
loose-not-tighttop-tagged\ljet top-tagged\ljet
forwardjet
Table 6: DefinitionofthecontrolregionsandthesignalregionfortheSP 2 channel.

Themainbackgroundprocessis+jetsproduction,whichmainlypassestheeventselectionintheSRifitcontainsjetsthatoriginatefromthehadronizationof-quarks.Thebosonismostlyproducedwithlowvaluesof\pt,sothatthe\ptll requirementstronglyreducesthisbackground.Inaddition,therequirementofatleastonetop-tagged\ljet efficientlysuppressesthecontributionfrom+jetsproduction,becauseitdoesnotcontaintopquarksandcanonlyfulfillthetop-taggingrequirementthroughmis-tags.Similarly,therequirementofatleastoneforwardjetreducesthe+jetsbackground,becauseforwardjetsarenotcharacteristicforthemainproductionmodeofthisprocess.Forthe+jetsproductionbackground,twoCRsaredefined.OneCR,called0-\btagged-jetCR,requiresthatno\btagged jetsbepresent,allowingtocorrectthemodelingof+jetsproductioninaregionthatiskinematicallyclosetotheSR.InasecondCR,called-\btagged-jetCR,themodelingof+jetsproductioninassociationwith\btagged jetsiscontrolled.Ifgooddata-MCagreementisobservedinbothCRsconsistently,thisprovidesconfidenceintheoverallmodelingof+jetsproduction.AsummaryoftheCRdefinitionsisgivenin\Tab6.BothCRsarebasedontheSRwithchangestothetop-tagging,-taggingandforward-jetrequirements.ForbothCRs,thetop-taggingrequirementischanged,sothattheremustbeatleastone\ljet thatfailsthetop-taggingrequirementsonbutfulfillsthetop-taggingrequirementsonthe\ljet mass.Outofthese\ljets,called“loose-not-tighttop-tagged”,the\ljet withthelargest\pt isusedinthecalculationofintheCRs.Thechangeinthetop-taggingrequirementenrichestheCRsin+jetsproductionincomparisonwithapotentialsignalcontribution.InbothCRs,theforward-jetrequirementisremoved,whichincreasesthenumberofeventsintheCRs.Finally,inthe0-\btagged-jetCR,no\btagged jetisallowed,whileinthe-\btagged-jetCRthesame-taggingrequirementasintheSRisused.TheresultingsamplesintheCRsareexpectedtobeand+jetsevents,respectively,andtocontainanegligibleamountofapotentialsingle--quarksignal.AstheCRsdonotcontainrequirementsonthenumberofforwardjetsandmakeuseofamodifiedtop-taggingrequirement(loose-not-tight),themodelingofthe+jetsbackgroundwascross-checkedinanotherregionwithno\btagged jets,butrequiringthepresenceofatleastoneforwardjetandusingthenominaltop-taggingalgorithm.ThemodelingofthedistributionsofkinematicpropertieswasfoundtobeconsistentbetweentheCRsandthecross-checkregionandasmalldifferenceobservedbetweentheoverallnumbersofeventswasassignedasasystematicuncertainty(\Sect6).

5.5 Searchstrategy:SP 3

Thesearchforsingle-productioninthetrileptonchannel(SP 3)issensitivetothedecay,featuringanadditionalleptonfromthetop-quarkdecay.ItishencecomplementarytotheSP 2 channel(\Sect5.4).ThedefinitionoftheSRissummarizedin\Tab7.Eventsmustpassthepreselection,andtheymusthaveatleastthreeleptonsincludingabosoncandidatewith\mll withina10 \GeV windowaround\mz.Inthischannel,aminimum\pt ofthebosoncandidateisalsorequired,.AsintheSP 2 channel(\Sect5.4),atleastone\btagged-jetandatleastoneforwardjetarerequired.Inordertosuppressbackgroundcontributionsinwhichleptonshavelower\pt onaveragethaninthesignal,thetransversemomentumofthehighest-\pt leptonineachevent,,mustbelargerthan200 \GeV.AsintheSP 2 channel,thepotentialsignalcontributionfrom\TTbar productionisreducedinthesearchforsingle--quarkproduction.IntheSP 3 channel,thisisachievedbyrequiringthat\htj multipliedbythenumberof\sjets intheeventissmallerthan6 \TeV.Thisrequirementhasanefficiencyof50–30%for\TTbar pair-productioninthesingletmodelinthemassrange800–1400 \GeV,whilemaintaininganefficiencyofforsingle--quarkproductionwithacrossthewholemassrangestudied.Inordertosearchforanexcessofdataoverthebackgroundprediction,\htjl isused,asinthePP 3 channel(\Sect5.3).

DibosonCR CR SR
Preselection
3leptons
=0\btagged jets \btagged jets
forwardjets forwardjets
<6 \TeV
Table 7: DefinitionofthecontrolregionsandthesignalregionfortheSP 3 channel.

Themainbackgroundprocessesaredibosonproductionwithadditional-quarksandproduction(dominatedbyproduction).Thecontributionsofthesebackgroundsarestronglyreducedbytherequirementson\ptll and,aswellasbyrequiringatleastoneforwardjet,becauseforwardjetsarenotcharacteristicfortheseprocesses.Forthetwomainbackgroundprocesses,dibosonandproduction,twoCRsaredefinedandsummarizedin\Tab7.ThedibosonCRisdefinedfollowingthecriteriaintheSR,buttherequirementson\ptll,andthepresenceofatleastoneforwardjetareremovedinordertoincreasethenumberofeventsintheCR.Inaddition,no\btagged jetisallowedinthedibosonCR.TheresultingCRsampleisexpectedtobedibosoneventsandtocontainanegligiblenumberofpotentialsignalevents.TheCRisbasedontheSRbyinvertingtherequirementonandbyrequiringthatnoforwardjetispresent.Thesechangesremovepotentialsignalcontributions.Inaddition,therequirementon\ptll isremovedinordertoincreasethenumberofeventsintheCR.TheresultingCRsampleisexpectedtoconsistmainlyofanddibosoneventsinsimilarproportions(and,respectively).

6 Systematicuncertainties

Systematicuncertaintiesaredividedintoexperimentaluncertainties,mostlyrelatedtotheuncertaintyinthemodelingofthedetectorresponseinthesimulation,andtheoreticaluncertainties,relatedtothetheoreticalmodelingofthebackgroundprocessesintheMCsimulation.Experimentaluncertaintiesonthesignalefficienciesandthesignalshapeofthediscriminatingvariablesarealsotakenintoaccount.Systematicuncertaintiesareevaluatedbyvaryingeachsourcebyofitsuncertainty.Asaresult,thepredictedbackgroundandsignaleventyieldsinthedifferentCRsandSRscanvaryaswellasthepredictedshapesofthediscriminatingvariablesintheseregions.Forsomesourcesonlyonesystematicvariationisdefined.Insuchcases,theeffectontheyieldsandshapesaresymmetrizedinordertoconstructthecorrespondingvariationintheotherdirection.Theuncertaintyintheintegratedluminosityoftheanalyzeddatasetis2.1%.ItisderivedfollowingamethodologysimilartothatinRef. [DAPR-2013-01]fromacalibrationoftheluminosityscaleusingbeam-separationscansinAugust2015andMay2016.Uncertaintiesinelectronandmuontrigger,reconstructionandidentificationefficienciesarederivedfromdatausingdecays [ATLAS-CONF-2016-024]anddecays [PERF-2015-10].Uncertaintiesintheelectron(muon)energy(momentum)calibrationandresolutionarealsoderivedusingevents [ATL-PHYS-PUB-2016-015, PERF-2015-10].Uncertaintiesinthe\sjet energyscaleareevaluatedfromMCsimulationsandfromdatausingmultijet,+jets,and+jetsevents [PERF-2016-04].Additional\sjet uncertaintiesarisefromthejetenergyresolution [ATL-PHYS-PUB-2015-015],whicharealsoderivedfrommultijet,+jetsand+jetseventsandfromthejetvertextagger.Uncertaintiesinthe-taggingefficiencyof\sjets arederivedfromdata [PERF-2012-04]for-jets,-jets,andotherlightjets.Forthederivationofthe-taggingefficiencyanditsuncertaintyfor-jets,dileptonic\ttbar eventsareused [Aaboud:2018xwy].AdditionaluncertaintiesarederivedusingMCsimulationsfortheextrapolationofthisefficiencybeyondthekinematicreachofthecalibration.Uncertaintiesinthe\ljet energyscale,massand-subjettinessratioarederivedfromacomparisonofthecalorimeter-to-track-jetratioindataandMCsimulations [PERF-2012-02, ATLAS-CONF-2017-063].Whiletheuncertaintyinthemassistakentobecorrelatedwiththeuncertaintyintheenergyscale,theuncertaintyistakentobeuncorrelatedwiththesetwo.Theuncertaintyintheresolutionsofthe\ljet energy,massandisestimatedbycomparingthepredictionfromthenominalMCsimulationswithsimulationswheretheresolutionis20%poorer.Theelectron,muonand\sjet uncertaintiesarepropagatedtothecalculationofthe\met.Additionaluncertaintiesareassignedtocontributionstothe\met calculationthatarisefromtrackswhicharematchedtotheprimaryvertexandnotassociatedwithanyobject [Aaboud:2018tkc].AllMCdistributionsarereweightedsothatthedistributionoftheaveragenumberofinteractionsperbunchcrossingcorrespondstothedistributionindata.Inordertoassesstheassociatedsystematicuncertainty,thereweightingisvariedwithinitsuncertainty.A5%uncertaintyisassignedtothecrosssectionfor\zjets production [ATL-PHYS-PUB-2016-002].Additionaluncertaintiesintheselectionefficiencyandintheshapeofthefinaldiscriminantduetothetheoreticalmodelingofthe\zjets processareevaluatedbycomparingthenominal\SHERPA samplewithalternativesamples,normalizedtothesamecrosssection.Anuncertaintyduetothechoiceofgeneratorandpartonshowerisassignedbycomparingthenominalsamplewithasamplegeneratedwith\MGMCatNLO andtheNNPDF3.0NLOPDFset,andshoweredwith\PYTHIAV8andusingtheA14setoftunedparameterswiththeNNPDF2.3LOPDFset.Anuncertaintyduetothescalechoiceisevaluatedbyvaryingtherenormalizationandfactorizationscalesinthenominalsampleindependentlybyfactorsof2and0.5.Theassigneduncertaintyisbasedonthelargestdeviationsfromthenominalsampleobserved.AnuncertaintyduetothechoiceofPDFsetisevaluatedbycomparingthenominal\SHERPA sampleusingtheNNPDF3.0NLOPDFsetwithsamplesusingtheMMHT2014NNLO [Harland-Lang:2014zoa]andCT14NNLOPDFsets [Dulat:2015mca].Thelargestobserveddeviationsfromthenominalsampleareusedtoassigntheuncertainty.Theuncertaintyinthecrosssectionfor\ttbar productionisassignedas/ [ATL-PHYS-PUB-2016-004].Alsofor\ttbar production,additionaluncertaintiesintheselectionefficiencyandintheshapeofthefinaldiscriminantareassignedbycomparingthenominalsamplewithalternativeMCsamples.Anuncertaintyduetothechoiceofgeneratorisevaluatedfromacomparisonofthenominal\POWHEGBOX samplewithasamplegeneratedwith\MGMCatNLO withtheNNPDF3.0NLOPDFset,andshoweredwith\PYTHIAV8usingtheA14setoftunedparametersandtheNNPDF2.3LOPDFset.Anuncertaintyduetothechoiceofshowermodelisassignedbycomparingthenominalsample,showeredby\PYTHIAV8,withanalternativesampleshoweredby\HERWIGV7 [Bahr:2008pv, Bellm:2015jjp]withtheH7-UE-MMHTsetoftunedparametersandtheMMHTPDFset.Theuncertaintiesduetothechoiceofrenormalizationandfactorizationscalesareevaluatedbyindependentlyvaryingthescalesbyfactorsof2and0.5.Thelargestdifferencesobservedareassignedasthesystematicuncertaintyforthesetwoscales.AnuncertaintyduetothechoiceofPDFsetisevaluatedbycomparingthenominalsamplewithsamplesgeneratedwiththeMMHT2014NLOandCT14NLOPDFsets.Thelargestobserveddeviationsfromthenominalsampleareusedtoassigntheuncertainty.Anuncertaintyof6%isassignedtothecrosssectionfordibosonproduction [ATL-PHYS-PUB-2016-002].Aswiththe\zjets and\ttbar processes,alternativeMCsamplesareusedtoassessadditionaluncertaintiesintheselectionefficiencyandintheshapeofthefinaldiscriminantofthedibosonprocesses.Inordertoassesstheuncertaintyduetothechoiceofrenormalizationandfactorizationscales,thenominal\SHERPA samplesarecomparedwithalternativesampleswiththescalesvariedindependentlybyfactorsof2and0.5andthelargestobserveddifferencesareassignedastheuncertainty.AnuncertaintyduetothechoiceofPDFsetisassessedbycomparingthenominalsamples,generatedwiththeNNPDF3.0NNLOPDFset,withsamplesgeneratedwiththeMMHT2014NNLOandCT14NNLOPDFsets.Thelargestdeviationsareusedtoassigntheuncertainty.Fortheprocesses,uncertaintiesof/areassignedfortheproductioncrosssectionandof/fortheproductioncrosssection [ATL-PHYS-PUB-2016-005].Fortheassessmentofadditionaluncertaintiesintheselectionefficiencyandintheshapeofthefinaldiscriminantoftheprocesses,thenominalsamplesarecomparedwithalternativeMCsamples.Anuncertaintyduetothechoiceofgeneratorisassignedbycomparingthenominalsamplewithasamplegeneratedwith\SHERPAV2.2andtheNNPDF3.0NLOPDFset.Forthesesamples,afastsimulationoftheATLASdetector [SOFT-2010-01]wasused,whichreliesonaparameterizationofthecalorimeterresponse [ATL-PHYS-PUB-2010-013].Thenominalsamplewasadditionallyproducedwiththefastsimulationconfigurationandtherelativedifferencesobservedinthecomparisonwiththesampleswithvariedscalesareassignedasthesystematicuncertainty.Anuncertaintyduetothepartonshowerisassignedbycomparingthenominalsamplewithsampleswithavariedamountofinitial-stateradiation.Thesealternativesampleswereproducedwithfastdetectorsimulationandtheproceduretoassignasystematicuncertaintyisagainbasedontherelativedifferenceobservedincomparisonwiththenominalsampleobtainedwithfastdetectorsimulation.Backgroundsduetomisidentifiedelectronsandmuonsplayaminorroleinthisanalysis,becausesuchleptonstypicallyhavelowtransversemomentumandarehencestronglysuppressedbytheSRrequirements,inparticularbythelowerthresholdsfor\ptll inthedifferentchannels.However,inthe\ttbar CRsinthePP 2 0-1J andPP 2 2J channelsandinthe\zjets CRinthePP 2 2J channel,low-\ptll eventsareincluded.Similarly,+jetsand\ttbar eventscouldcontributetotheCRsandSRsinthePP 3 andSP 3 channelsduetomisidentifiedleptons.ThemaximumobserveddifferencebetweendataandMCsimulationsinthelepton\pt spectraintheCRsis25%.Thisisassignedasanuncertaintyto+jetsand\ttbar eventsinthetrileptonchannelsandto\ttbar eventswithinthePP 2 0-1J andPP 2 2J channels.No-taggedjetareallowedinthedibosonCRsforthePP 3 andSP 3 channels(\Sect5.3and\Sect5.5).Whilethisrequirementensuresahighpurityindibosonprocesses,itdiffersfromtherequirementsintheSRs.Anuncertaintyof50%isassignedtotheproductionofdibosoneventsinassociationwith-quarks,motivatedbytheprecisionofmeasurementsof-and-bosonproductioninassociationwith-quarks [STDM-2012-11, STDM-2012-15].InordertoensurealargenumberofeventsintheCRsforthedileptonsingle-productionsearch,theSRforward-jetrequirementisremoved(\Sect5.4).Across-checkwasperformedinaregionthatonlydiffersfromtheSRbyavetoon-taggedjets.Whilethemodelingoftheshapesofkinematicvariablesinthisregionissatisfactory,the11%differenceintheoverallnumberofeventsbetweendataandbackgroundexpectationisassignedasanadditionaluncertaintyintheSRduetotheforward-jetrequirement.TheuncertaintiesonthereconstructedobjectsandtheluminosityalsoaffectthepredictionsforVLQpairandsingleproduction.Nofurtheruncertaintiesonthesignalprocesseswereconsidered.Asdiscussedin\Sect3,theMCsamplesforVLQpairproductionweregeneratedinthesingletmodelandalternativeBRhypothesesforandquarksareobtainedbyreweightingthesingletBRstothealternativeBRs.ThisprocedureisvalidatedbycomparingkinematicdistributionsofthenominalVLQpairproductionsampleswithalternativesamplesthatweregeneratedinthe( )doubletmodel.AfterreweightingbothtothesameBRs,nolargedifferenceswereobservedbetweenthesesamples.Hence,thereweightingprocedureisconsideredvalidatedandnosystematicuncertaintyisassigned.

7 Results

Ineachchannel,abinnedlikelihoodfitisperformedtothediscriminatingvariable.Controlandsignalregionsarefitsimultaneouslyandsystematicuncertaintiesareincludedinthefitasasetofnuisanceparameters(NP),.ThelikelihoodfunctionconsistsofPoissonprobabilitiesforeachbininthediscriminatingvariableineachregion,andaGaussianorlog-normaldistributionforeachNP.Inthelikelihoodfit,thesignalcrosssectionisparameterizedbymultiplyingthepredictedcrosssectionwithacorrectionfactor,calledthesignal-strengthfactor,whichisafreeparameterofthefit.Inabackground-onlyfit,andhence,issettozero.ForthecombinedcontrolandsignalregionfitthemodelingofthemainbackgroundprocesseswasadjustedduringthefitviachangesintheNPs,sothatthebackgroundpredictioninthesignalregionsisimproved.ThebinningofthediscriminatingvariableinthedifferentchannelswaschoseninordertoretainasmuchshapeinformationaboutthedistributionaspossiblegiventhenumberofbackgroundMCeventsineachbin.Theeffectofeachsinglesourceofsystematicuncertaintyistreatedascorrelatedacrossallregionsandprocesseswithtwoexceptions.Fortheuncertaintiesassociatedwithmisidentifiedleptons,separateNPsaredefinedforthedifferentCRsandSRsineachchannel;fortheuncertaintiesrelatedtothechoiceofMCgeneratorandhadronizationmodel,separateNPsaredefinedforeachchannel.Differentsourcesofsystematicuncertaintyaretreatedasuncorrelatedwitheachother,exceptforthecaseofthe\ljet scaleuncertaintiesaffectingthe\pt andmass,whicharetreatedas100%correlated.Inadditiontothesystematicuncertaintiesdiscussedin\Sect6,anadditionalNPisaddedforeachbininthediscriminatingvariableineachregionduetothestatisticaluncertaintyoftheMCsamples.

7.1 Results:PP 2 0-1j

TheobservednumberofeventsintheSRsandCRsandtheexpectednumberofeventsforthedifferentbackgroundcontributionsareshownin\Tab9forthePP 2 0-1J channel.Alsoshownistheexpectednumberofeventsfor\BBbar and\TTbar productioninthesingletmodelfor.Thesignalefficienciesforthesebenchmarksare0.060%(0.013%)for\BBbar (\TTbar)inthe0-\ljet SRand0.33%(0.16%)inthe1-\ljet SR,andincludethebranchingratiosoftheVLQaswellasofitsdecayproducts,includingthedecay.

CR +jetsCR 0-\ljet SR 1-\ljet SR
Singlet\BBbar(900 GeV)
Singlet\TTbar(900 GeV)
+jets
Singletop
Diboson
TotalBkg.
Data
Data/Bkg. 1.00 0.26 0.98 0.26 1.4 0.6 1.0 0.4
Table 9: Observednumberofeventsindataandpost-fitexpectednumberofbackgroundeventsinthecontrolandsignalregionsforthePP 2 0-1J channel,i.e. afterthefittothedata\htj distributionsunderthebackground-onlyhypothesis.Theuncertaintyintheexpectednumberofeventsisthefulluncertaintyfromthefit,fromwhichtheuncertaintyintheratiooftheobservedandexpectednumbersofeventsiscalculated.
CR +jetsCR 0-\ljet SR 1-\ljet SR
+jets
Singletop
Diboson
TotalBkg.
Data
Data/Bkg. 1.003 0.020 0.99 0.05 1.32 0.16 0.95 0.08
Table 8: Observednumberofeventsindataandpre-fitexpectednumberofsignalandbackgroundeventsinthecontrolandsignalregionsforthePP 2 0-1J channel,i.e. beforethefittodata.Forthesignal,theexpectednumberofeventsforthe\BBbar and\TTbar benchmarkprocesseswithisshownforthesingletmodel.StatisticaluncertaintiesfromthelimitedsizeofMCsamplesandsystematicuncertaintiesareaddedinquadrature.TheuncertaintyintheratiooftheobservedandexpectednumbersofeventscontainsthestatisticaluncertaintyofthepredictionfromPoissonfluctuations.

Afitofthebackgroundpredictiontothe\htj distributionsindatawasperformed.Thepost-fityieldsareshownin\Tab9.Theuncertaintyinthebackgroundpredictionissignificantlyreducedinallregionscomparedtothepre-fitvalue(\Tab9).Theoverall+jets(\ttbar)normalizationisadjustedbyafactorof\normfzjetszjcrtwolr (\normfttbarttcrtwolr)inthe+jets(\ttbar)CR.Theratiosofthepost-fitandpre-fitbackgroundyieldsareconsistentwithunityinallregions.

[]
[]

Figure 3: Comparisonofthedistributionofthetransversemomentumofthebosoncandidate,\ptll,betweendataandthebackgroundpredictionin(a)the\ttbar controlregion,(b)the+jetscontrolregion,(c)the0-\ljet-signalregion,and(d)the1-\ljet-signalregionofthepair-production(PP) 2 0-1J channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedata\htj distributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anupwardpointingtriangleintheratioplotindicatesthatthevalueoftheratioisbeyondscale.Anexampledistributionfora\BBbar signalinthesingletmodelwithisoverlaid.Forbettervisibility,itismultipliedbyafactoroffive.Thedataarecompatiblewiththebackground-onlyhypothesis.

[]
[]

Figure 4: Comparisonofthedistributionofthescalarsumof\sjet transversemomenta,\htj,betweendataandthebackgroundpredictionin(a)the\ttbar controlregion,(b)the+jetscontrolregion,(c)the0-\ljet-signalregion,and(d)the1-\ljet-signalregionofthepair-production(PP) 2 0-1J channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedata\htj distributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anupwardpointingtriangleintheratioplotindicatesthatthevalueoftheratioisbeyondscale.Anexampledistributionfora\BBbar signalinthesingletmodelwithisoverlaid.Forbettervisibility,itismultipliedbyafactoroffive.Thedataarecompatiblewiththebackground-onlyhypothesis.

ThemodelingofthemainbackgroundswasvalidatedbycomparingthedistributionsofkinematicvariablesandobjectmultiplicitiesbetweendataandbackgroundpredictionineachrespectiveCR.Asanexample,the\ptll distributionisshownin\Fig3inthetwoCRsandthetwoSRs.Thebackgroundpredictionisshownafterthefittothe\htj distribution.In\Fig4,the\htj distributionisshownintheCRsandSRsfordataandthebackgroundpredictionafterthefit.TheVLQpair-productionsignalwouldbeexpectedtoresultinanexcessofdataoverthebackgroundpredictionatlargevaluesof\htj,asshownin\Fig4and\Fig4.GoodagreementbetweendataandthebackgroundpredictionisobservedinbothkinematicvariablesintheCRs,validatingthebackgroundprediction.

7.2 Results:PP 2 2j

TheobservedandexpectedyieldsintheSRandtheCRsandtheexpectednumberofeventsforthedifferentbackgroundcontributionsareshownin\Tab11forthePP 2 2J channel.Alsoshownistheexpectednumberofeventsfor\BBbar and\TTbar productioninthesingletmodelfor.ThesignalefficiencyintheSRforthesebenchmarksis0.28%forboth\BBbar and\TTbarproduction,andincludesthebranchingratiosoftheVLQaswellasofitsdecayproducts,includingthedecay.

CR +jetsCR SR
Singlet\BBbar(900 GeV)
Singlet\TTbar(900 GeV)
+jets
Singletop
Diboson
TotalBkg.
Data
Data/Bkg. 1.2 0.8 0.98 0.26 0.7 0.4
Table 11: Observednumberofeventsindataandpost-fitexpectednumberofbackgroundeventsinthecontrolregionsandthesignalregionforthePP 2 2J channel,i.e. afterthefittothedatadistributionsunderthebackground-onlyhypothesis.Theuncertaintyintheexpectednumberofeventsisthefulluncertaintyfromthefit,fromwhichtheuncertaintyintheratiooftheobservedandexpectednumbersofeventsiscalculated.
CR +jetsCR SR
+jets
Singletop
Diboson
TotalBkg.
Data
Data/Bkg. 1.01 0.11 1.00 0.10 0.83 0.21
Table 10: Observednumberofeventsindataandpre-fitexpectednumberofsignalandbackgroundeventsinthecontrolregionsandthesignalregionforthePP 2 2J channel,i.e. beforethefittodata.Forthesignal,theexpectednumberofeventsforthe\BBbar and\TTbar benchmarkprocesseswithisshownforthesingletmodel.StatisticaluncertaintiesfromthelimitedsizeofMCsamplesandsystematicuncertaintiesareaddedinquadrature.TheuncertaintyintheratiooftheobservedandexpectednumbersofeventscontainsthestatisticaluncertaintyofthepredictionfromPoissonfluctuations.

Afitofthebackgroundpredictiontothedistributionsindatawasperformed.Thepost-fityieldsareshownin\Tab11.Theuncertaintyinthebackgroundpredictionwassignificantlyreducedinallregionscomparedtothepre-fitvalue(\Tab11).Theoverall+jets(\ttbar)normalizationwasadjustedbyafactorof\normfzjetszjcrtwolb (\normfttbarttcrtwolb)inthe+jets(\ttbar)CR.Theratiosofthepost-fitandpre-fitbackgroundyieldsareconsistentwithunityinallregions.

[]
[]

Figure 5: Comparisonofthedistributionofthescalarsumof\sjet transversemomenta,\htj,betweendataandthebackgroundpredictionin(a)the\ttbar controlregion,(b)the+jetscontrolregion,and(c)thesignalregionofthepair-production(PP) 2 2J channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedatadistributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anupwardordownwardpointingtriangleintheratioplotindicatesthatthevalueoftheratioisbeyondscale.Anexampledistributionfora\BBbar signalinthesingletmodelwithisoverlaid.Thedataarecompatiblewiththebackground-onlyhypothesis.

[]
[]

Figure 6: Comparisonofthedistributionoftheinvariantmassofthebosoncandidateandthehighest-\pt \btagged jet,,betweendataandthebackgroundpredictionin(a)the\ttbar controlregion,(b)the+jetscontrolregion,and(c)thesignalregionofthepair-production(PP) 2 2J channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedatadistributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anupwardordownwardpointingtriangleintheratioplotindicatesthatthevalueoftheratioisbeyondscale.Anexampledistributionfora\BBbar signalinthesingletmodelwithisoverlaid.Thedataarecompatiblewiththebackground-onlyhypothesis.

ThemodelingofthemainbackgroundswasvalidatedbycomparingthedistributionsofkinematicvariablesandobjectmultiplicitiesbetweendataandbackgroundpredictionintherespectiveCR.Asanexample,the\htj distributionisshownin\Fig5inthetwoCRsandintheSR.Thebackgroundpredictionisshownafterthefittothedistribution.In\Fig6,thedistributionisshownintheCRsandSRfordataandthebackgroundpredictionafterthefit.TheVLQpair-productionsignalwouldbeexpectedtoresultinanexcessofdataoverthebackgroundpredictionatlargevaluesof,asshownin\Fig6.GoodagreementbetweendataandthebackgroundpredictionisapparentinkinematicvariablesintheCRs,validatingthebackgroundprediction.

7.3 Results:PP 3

TheobservednumberofeventsintheSRandtheCRsandtheexpectednumberofeventsforthedifferentbackgroundcontributionsareshownin\Tab13forthePP 3 channel.Alsoshownistheexpectednumberofeventsfor\BBbar and\TTbar productioninthesingletmodelfor.ThesignalefficiencyintheSRforthesebenchmarksis0.31%for\BBbarand0.44%for\TTbarproduction,andincludesthebranchingratiosoftheVLQaswellasofitsdecayproducts,includingthedecay.

DibosonCR CR SR
Singlet\BBbar(900 GeV)
Singlet\TTbar(900 GeV)
+jets
Singletop
Diboson
Triboson
TotalBkg.
Data
Data/Bkg. 0.94 0.23 1.12 0.24 1.11 0.24
Table 13: Observednumberofeventsindataandpost-fitexpectednumberofbackgroundeventsinthecontrolregionsandthesignalregionforthePP 3 channel,i.e. afterthefittothedata\htjl distributionsunderthebackground-onlyhypothesis.Theuncertaintyintheexpectednumberofeventsisthefulluncertaintyfromthefit,fromwhichtheuncertaintyintheratiooftheobservedandexpectednumbersofeventsiscalculated.
DibosonCR CR SR
+jets
Singletop
Diboson
Triboson
TotalBkg.
Data
Data/Bkg. 1.00 0.04 1.02 0.07 1.03 0.07
Table 12: Observednumberofeventsindataandpre-fitexpectednumberofsignalandbackgroundeventsinthecontrolregionsandthesignalregionforthePP 3 channel,i.e.beforethefittodata.Forthesignal,theexpectednumberofeventsforthe\BBbar and\TTbar benchmarkprocesseswithisshownforthesingletmodel.StatisticaluncertaintiesfromthelimitedsizeofMCsamplesandsystematicuncertaintiesareaddedinquadrature.TheuncertaintyintheratiooftheobservedandexpectednumbersofeventscontainsthestatisticaluncertaintyofthepredictionfromPoissonfluctuations.

Afitofthebackgroundpredictiontothe\htjl distributionsindatawasperformedandthepost-fityieldsareshownin\Tab13.Theuncertaintyinthebackgroundpredictionwassignificantlyreducedinallregionscomparedtothepre-fitvalue(\Tab13).Theoveralldiboson()normalizationisadjustedbyafactorof\normfdibosonvvcrtril (\normfttvttvcrtril)inthediboson()CR.Theratiosofthepost-fitandpre-fitbackgroundyieldsareconsistentwithunityinallregions.

[]
[]

Figure 7: Comparisonofthedistributionofthetransversemomentumofthehighest-\pt lepton(leadinglepton),,betweendataandthebackgroundpredictionin(a)thedibosoncontrolregion,(b)thecontrolregion,and(c)thesignalregionofthepair-production(PP) 3 channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedata\htjl distributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anexampledistributionfora\BBbar signalinthesingletmodelwithisoverlaid.Forbettervisibility,itismultipliedbyafactoroffive.Thedataarecompatiblewiththebackground-onlyhypothesis.

[]
[]

Figure 8: Comparisonofthedistributionofthescalarsumof\sjet andleptontransversemomenta,\htjl,betweendataandthebackgroundpredictionin(a)thedibosoncontrolregion,(b)thecontrolregion,and(c)thesignalregionofthepair-production(PP) 3 channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedata\htjl distributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anupwardpointingtriangleintheratioplotindicatesthatthevalueoftheratioisbeyondscale.Anexampledistributionfora\BBbar signalinthesingletmodelwithisoverlaid.Forbettervisibility,itismultipliedbyafactoroffive.Thedataarecompatiblewiththebackground-onlyhypothesis.

ThemodelingofthemainbackgroundswasvalidatedbycomparingthedistributionsofkinematicvariablesandobjectmultiplicitiesbetweendataandbackgroundpredictionintherespectiveCR.Asanexample,thedistributionofthe\pt ofthehighest-\pt lepton,,intheeventisshownin\Fig7inthetwoCRsandintheSR.Thebackgroundpredictionisshownafterthefittothe\htjl distribution.In\Fig8,the\htjl distributionisshownintheCRsandSRfordataandthebackgroundpredictionafterthefit.TheVLQpair-productionsignalwouldbeexpectedtoresultinanexcessofdataoverthebackgroundpredictionatlargevaluesof\htjl,asshownin\Fig8.GoodagreementbetweendataandthebackgroundpredictionisobservedinbothkinematicvariablesintheCRs,validatingthebackgroundprediction.

7.4 Results:SP 2

TheobservednumberofeventsintheSRandtheCRsandtheexpectednumberofeventsforthedifferentbackgroundcontributionsareshownin\Tab15fortheSP 2 channel.Alsoshownistheexpectednumberofeventsforsingle-productionforand.ThesignalefficiencyintheSRisfordecaysproducedviaacoupling,andincludesthebranchingratiosoftheVLQaswellasofitsdecayproducts,includingthedecay.

0-\btagged-jetCR -\btagged-jetCR SR
Single-(900 GeV,)
+jets
Singletop
Diboson
Triboson
TotalBkg.
Data
Data/Bkg. 0.96 0.31 0.87 0.19 0.81 0.27
Table 15: Observednumberofeventsindataandpost-fitexpectednumberofbackgroundeventsinthecontrolregionsandthesignalregionfortheSP 2 channel,i.e. afterthefittothedatadistributionsunderthebackground-onlyhypothesis.Theuncertaintyintheexpectednumberofeventsisthefulluncertaintyfromthefit,fromwhichtheuncertaintyintheratiooftheobservedandexpectednumbersofeventsiscalculated.
0-\btagged-jetCR -\btagged-jetCR SR
+jets
Singletop
Diboson
Triboson
TotalBkg.
Data
Data/Bkg. 1.00 0.04 0.97 0.06 0.98 0.11
Table 14: Observednumberofeventsindataandpre-fitexpectednumberofsignalandbackgroundeventsinthecontrolregionsandthesignalregionfortheSP 2 channel,i.e. beforethefittodata.Forthesignal,theexpectednumberofeventsforthesingle--quarkbenchmarkprocesswithandisshown.StatisticaluncertaintiesfromthelimitedsizeofMCsamplesandsystematicuncertaintiesareaddedinquadrature.TheuncertaintyintheratiooftheobservedandexpectednumbersofeventscontainsthestatisticaluncertaintyofthepredictionfromPoissonfluctuations.

Afitofthebackgroundpredictiontothedistributionsindatawasperformedandthepost-fityieldsareshownin\Tab15.Theuncertaintyinthebackgroundpredictionwassignificantlyreducedinallregionscomparedtothepre-fitvalue(\Tab15).Theoverall+jetsnormalizationwasadjustedbyfactorsof\normfzjetsnotnobcrtwols and\normfzjetsnotbcrtwols inthe0-\btagged-jetCRandthe-\btagged-jetCR,respectively.Theratiosofthepost-fitandpre-fitbackgroundyieldsareconsistentwithunityinallregions.Theratiofor+jetsproductionintheSRis\normfzjetssrtwols,whichisconsistentwithunitywithinatmost.

[]
[]

Figure 9: Comparisonofthedistributionofthetransversemomentumofthehighest-\pt top-tagged\ljet betweendataandthebackgroundpredictionin(a)the0-\btagged-jetcontrolregion,(b)the-\btagged-jetcontrolregion,and(c)thesignalregionofthesingle-production(SP) 2 channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedatadistributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anupwardpointingtriangleintheratioplotindicatesthatthevalueoftheratioisbeyondscale.Anexampledistributionforasingle--quarksignalwithandisoverlaid.Forbettervisibility,itismultipliedbyafactorofthree.Thedataarecompatiblewiththebackground-onlyhypothesis.

[]
[]

Figure 10: Comparisonofthedistributionoftheinvariantmassofthebosoncandidateandthehighest-\pt top-tagged\ljet,,betweendataandthebackgroundpredictionin(a)the0-\btagged-jetcontrolregion,(b)the-\btagged-jetcontrolregion,and(c)thesignalregionofthesingle-production(SP) 2 channel.Thebackgroundpredictionisshownpost-fit,i.e. afterthefittothedatadistributionsunderthebackground-onlyhypothesis.Thelastbincontainstheoverflow.Anupwardpointingtriangleintheratioplotindicatesthatthevalueoftheratioisbeyondscale.Anexampledistributionforasingle--quarksignalwithandisoverlaid.Forbettervisibility,itismultipliedbyafactorofthree.Thedataarecompatiblewiththebackground-onlyhypothesis.

ThemodelingofthemainbackgroundwasvalidatedbycomparingthedistributionsofkinematicvariablesandobjectmultiplicitiesbetweendataandbackgroundpredictioninthetwoCRs.Asanexample,the\pt distributionofthehighest-\pt top-tagged\ljet intheeventisshownin\Fig9inthetwoCRsandtheSR.Thebackgroundpredictionisshownafterthefittothedistribution.ContributionsfromVLQsingleproductionwouldbeexpectedathighvaluesofthe\ljet \pt.In\Fig10,thedistributionisshownintheCRsandSRfordataandthebackgroundpredictionafterthefit.TheVLQsingle-productionsignalwouldbeexpectedtoresultinanexcessofdataoverthebackgroundpredictioninthedistribution,asshownin\Fig10.GoodagreementbetweendataandthebackgroundpredictionisobservedinbothkinematicvariablesintheCRs,validatingthebackgroundprediction.

7.5 Results:SP 3

TheobservednumberofeventsintheSRandtheCRsandtheexpectednumberofeventsforthedifferentbackgroundcontributionsareshownin\Tab17fortheSP 3 channel.Alsoshownistheexpectednumberofeventsforsingle--quarkproductionforand.DuetoalownumberofMCeventsforthesingle--quarksignalintheSP 3 channel,inthischannelthesignalefficiencywasinterpolatedasafunctionofwithathird-orderpolynomialdescribingtheefficienciesestimatedfromMCsimulationswithintheuncertainties.TheresultingsignalefficiencyintheSRisfordecaysproducedviaacoupling,andincludesthebranchingratiosoftheVLQaswellasofitsdecayproducts,includingthedecay.

DibosonCR CR SR
Single-(900 GeV,)
+jets
Singletop
Diboson
Triboson