Search for pair-produced long-lived neutral particles decaying to jets in the ATLAS hadronic calorimeter in pp collisions at s√ = 8 TeV

HAL Id: in2p3-01104690

http://hal.in2p3.fr/in2p3-01104690

Submitted on 2 Apr 2015

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Search for pair-produced long-lived neutral particles decaying to jets in the ATLAS hadronic calorimeter in

pp collisions at s √ = 8 TeV

G. Aad, S. Albrand, J. Brown, J. Collot, S. Crépé-Renaudin, B. Dechenaux, P.A. Delsart, C. Gabaldon, M.H. Genest, J.Y. Hostachy, et al.

To cite this version:

G. Aad, S. Albrand, J. Brown, J. Collot, S. Crépé-Renaudin, et al.. Search for pair-produced long-lived neutral particles decaying to jets in the ATLAS hadronic calorimeter in pp collisions at s √ = 8 TeV.

Physics Letters B, Elsevier, 2015, 743, pp.15-34. �10.1016/j.physletb.2015.02.015�. �in2p3-01104690�

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Search for pair-produced long-lived neutral particles decaying to jets in the ATLAS hadronic calorimeter in pp collisions at √

s = ^{8 TeV}

.ATLAS Collaboration

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

Articlehistory:

Received16January2015

Receivedinrevisedform4February2015 Accepted5February2015

Availableonline11February2015 Editor: W.-D.Schlatter

Keywords:

High-energycolliderexperiment Long-livedneutralparticle Newphysics

The ATLASdetector attheLargeHadronCollider atCERNisused to searchforthe decayofascalar bosontoapairoflong-livedparticles,neutralundertheStandardModelgaugegroup,in20.3 fb⁻¹of datacollectedinproton–protoncollisionsat√

s=^{8 TeV.Thissearchissensitivetolong-livedparticles} thatdecaytoStandardModel particlesproducing jetsattheouteredgeofthe ATLASelectromagnetic calorimeter orinsidethe hadroniccalorimeter.No significantexcessofevents isobserved.Limitsare reportedontheproductofthescalarbosonproductioncrosssectiontimesbranchingratiointolong-lived neutralparticlesasafunctionoftheproperlifetimeoftheparticles.Limitsarereportedforbosonmasses from100GeV to900GeV,andalong-livedneutralparticlemassfrom10GeV to150GeV.

PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

1. Introduction

ThediscoveryoftheHiggsboson[1–3]bytheATLASandCMS experiments [4,5] in 2012 identified the last piece ofthe highly successfulStandardModel(SM).Subsequentmeasurementsofthe Higgsbosonbranchingratiosandcouplings,whileconsistentwith theSMexpectations,allowforasubstantialbranchingratiotoex- oticparticles.ThisletterdescribesasearchfordecaysoftheHiggs bosonandotherscalarbosonstonon-SMstatesthatinturndecay toSMparticles.

A number of extensions of the SM involve a hidden sector that isweakly coupled to theSM, with thetwo connected via a communicatorparticle. Thisletter considers models containing a hiddensectorwithaconfininggaugeinteractionthatisotherwise invisibletotheSM.ThecommunicatorischosentobeaSM-sector scalar boson,

[6–9]. The communicator mixes with a hidden- sectorscalarboson,

_hs,whichdecaysintodetectableSMparticles.

Thissearchconsiderscommunicatormassesbetween100 GeV and 900 GeV.A

massclosetothemassofthediscoveredHiggsbo- sonisincludedtosearchforexoticdecaysoftheHiggsboson.

A Hidden Valley (HV) model [8,9] is used as the benchmark model.The lightest HV particles forman isospintriplet of pseu- doscalar particles which are called valley pions (

π

v) because of theirsimilaritytotheSMtriplet.The

π

varepair-produced(

_hs

→ π

v

π

v)andeach decaysto apair ofSM fermions.The

π

v possess Yukawacouplingstofermionsandthereforepreferentiallydecayto accessibleheavyfermions,primarilybb,ccand

τ

⁺

τ

⁻.

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

Thelifetimeofthe

π

visunconstrainedandcouldbequitelong.

A long-lived

π

v can resultin signatures that traditionalsearches fail todetect.Ifa

π

v decaysintheinner detectorormuonspec- trometer, itcan be reconstructed asa displacedvertex. However, standard vertex-finding algorithms [10] are not likely to recon- struct it without modification. Likewise, a

π

v decay deep inside the calorimeteris reconstructed asa jet withan unusual energy signaturethatmosttraditionalsearchesrejectashavingpoordata quality. Thissearch focusses onfinal states whereboth

π

v decay inthehadroniccalorimeterorneartheouteredgeoftheelectro- magnetic calorimeter. Each heavy fermion pair from a

π

v decay is reconstructed asa single calorimeterjet withthree character- isticproperties:anarrowradius,notracksfromchargedparticles matchedto thejet,andlittleornoenergydeposited intheelec- tromagneticcalorimeter.

Scalar boson masses ranging from 100 GeV to 900 GeV are considered in addition to the Higgs boson’s mass (generated at m_H

=

^{126 GeV)and}

π

v massesbetween10 GeV and150 GeV are studied.Othersearchesforpairsofdisplacedverticesgeneratedby pair-produced neutral, long-livedparticleswere performedinAT- LAS [11] and CMS[12] at the LHC andin D0 [13] andCDF [14]

atthe Tevatron.The Tevatronexperiments andCMS searchedfor displaced vertices in their tracking system only, which results in a corresponding proper decay length range of a few meters.

CMSalso lookedatthe multi-leptondecaychannel, anotherpos- sible decay of HV particles. The previous ATLAS analysis, based on 7 TeV data,used the muon spectrometer and is sensitive to proper decay lengths between 0.5 m and 27 m, depending on thebenchmarkmodel.No evidenceofphysicsbeyondtheSMwas found.

http://dx.doi.org/10.1016/j.physletb.2015.02.015

0370-2693/PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

2. TheATLASdetector

TheATLASdetector[15]isamulti-purposedetectorattheLHC, consistingofseveralsub-detectors.Fromtheinteractionpoint(IP) outwards there are an inner detector (ID), electromagnetic and hadroniccalorimeters,andamuonspectrometer(MS).TheID,im- mersed in a 2 T axial magnetic field, provides tracking and ver- tex information for charged particles within the pseudorapidity¹ (

η

) region

| η _| <

2

.

5. It consistsof three differenttracking detec- tors.From smallradii outwards,thesearea siliconpixeldetector, a siliconmicrostriptracker(SCT)andatransitionradiationtracker (TRT).

The calorimeterprovides coverage overthe range

| η _| <

4

.

9.It consistsof a lead/liquid-argon electromagneticcalorimeter (ECal) at smaller radii surrounded by a hadronic calorimeter (HCal) at larger radii comprising a steel and scintillator-tile system in the barrel region (

| η _| <

1

.

7) and a liquid-argon system with copper absorbers in the endcaps (1

.

5

< | η | <

3

.

2). The ECal spans the range1

.

5 m

<

r

<

2

.

0 m in thebarreland 3

.

6 m

< |

^z

| <

4

.

25 m intheendcaps.TheHCalcovers2

.

25 m

<

r

<

4

.

25 m inthebarrel and4

.

3 m

< |

^z

| <

6

.

05 m intheendcaps.Thereisalsoa forward calorimeter(FCal), withcoverage between 3

.

1

< | η | <

4

.

9,which usescopperabsorbersinthefirstlayer,andtungstenabsorbersin thesecondandthirdlayers,andliquid-argonastheactivemedium inalllayers.Muonidentificationandmomentummeasurementare provided by the MS, which extends to

| η _{| =}

2

.

7. It consistsof a three-layer system of gas-filled precision-tracking chambers. The region

| η | <

2

.

4 isalsocoveredbyseparatetriggerchambers.

A sequential three-level trigger system selects events to be recorded for offline analysis. The first level consists of custom hardware that implements selection on jets, electrons, photons,

τ

leptons,muons,andmissingtransversemomentumorlargetotal transverseenergy.Thesecondandthirdlevelsaddchargedparticle trackfindingandrefinethefirst-levelselectionswithprogressively moredetailedalgorithms.

3. Dataandsimulationsamples

All data used inthis analysis were collected during the 2012 LHC proton–protonrun ata centre-of-mass energy of 8 TeV. Af- terdataqualityrequirementsareapplied,thesamplecorresponds to an integrated luminosity of 20

.

3 fb⁻¹. The HV Monte Carlo (MC)samples aregeneratedwithPYTHIA8.165[16] andthePDF MSTW2008[17]tosimulategluon fusiongg

→

productionand the

hs decay

hs

→ π

v

π

v for different

and

π

v masses (Ta- ble 1).

massesbelow300 GeV areconsideredlow-masssamples andtherestareconsideredhigh-masssamples.The

π

v lifetimeis fixed ineach sample to ensuredecays throughouttheATLAS de- tector.The

issimulatedinPYTHIAbyreplacingtheHiggsboson withthe

andhavingthe

decayto

π

v 100% ofthetime.The

samples are produced with cross sections calculated at next- to-next-to-leading-logarithmic accuracy in QCD processes and at next-to-leading-order in electro-weak processes assuming the

ateachmasshasthesamepropertiesastheSMHiggsboson[18].

Aftergenerationtheeventsarepassed throughadetailedsimula- tionofthe detectorresponsewithGEANT4 [19,20]andthesame reconstructionalgorithmsasareusedonthedata.GEANT4needed nomodificationtosimulatethesignalasalldecayparticlesareSM

1 ATLASusesaright-handed coordinatesystemwith itsoriginat thenominal interactionpoint(IP)inthecentreofthedetectorandthez-axisalongthebeam pipe.Thex-axispointsfromtheIPtothecentreoftheLHCring,andthe y-axis pointsupward.Cylindricalcoordinates (r,φ)areusedinthe transverseplane,φ beingtheazimuthalanglearoundthebeampipe.Thepseudorapidityisdefinedin termsofthepolarangleθasη= −^{ln tan}(θ/2).

Table 1

The massorHiggsbosonmass, gluonfusionproductioncrosssection,and πvmassofeachbenchmarkHiddenValleymodelgenerated.Thecross-sectionsare based onthe assumptioninthebenchmarked modelthatthe bosonproduc- tionmechanismisthesameastheHiggsbosonproductionmechanism.Thedecay branchingratiosoftheπvasafunctionoftheπvmassarelistedinthesecond tableasdeterminedinthesimulationsamples.

mH[GeV] σ[pb] πvMass [GeV]

126 19.0 10, 25, 40

Mass [GeV] σ [pb] πvMass [GeV]

100 29.7 10, 25

140 15.4 10, 20, 40

300 3.59 50

600 0.52 50, 150

900 0.06 50, 150

πvMass [GeV] BRbb[%] BRτ⁺τ⁻[%] BRcc[%]

10 70.0 16.4 13.4

20 86.3 8.0 5.6

25 86.6 8.1 5.3

40 86.5 8.5 5.0

50 86.2 8.8 4.9

150 84.8 10.2 4.8

particles.AllMCsamplesarereweightedtoreproducethenumber ofinteractionsperbunchcrossingobservedinthedata.

4. Triggerandeventselection

Candidateeventsarecollected usingadedicatedtrigger,called the CalRatio trigger [21], which looks specifically for long-lived neutral particles that decay near theouter radius of theECal or within the HCal.The trigger is tuned tolook forevents contain- ingatleastonenarrowjetwithlittleenergydepositedintheECal andno chargedtracks pointingtowards thejet. At thefirst level thetriggerselectsonlynarrowjetsbyrequiringatleast40 GeV of transverseenergy(ET)inthecalorimeterina0

.

2

×

⁰

.

2 (

η _× φ

) region usingtopologicaljets[15,22],incontrasttothedefaultal- gorithminwhichtheenergyina0

.

4

×

⁰

.

4 regionissummed.The 40 GeV ETthresholdrequirementisfullyefficientatan offlinejet E_T of 60 GeV.To selectjets witha highfraction oftheir energy in theHCalthe secondlevel ofthetrigger requiresthesenarrow jetstohavelog₁₀

(

E_H

/

E_EM

) >

1

.

2,whereE_H

/

E_EMistheratioofthe energydepositedintheHCal(EH)totheenergydepositedinthe ECal (EEM). The trigger also requires no tracks with pT

>

1 GeV in theregion 0

.

2

×

⁰

.

2 (

η × φ

)around thejet axis.The third levelofthetriggerusestheslowerbutmoreaccurateanti-kt algo- rithm[23]withR

=

⁰

.

4 toreconstructthejetandrequiresthejet tohaveaminimumof35 GeV oftransverseenergy.

The probability(

ε

_πv) forasingle

π

v tofire thetriggerinsim- ulatedeventsisshowninFig. 1,forthe(a)barreland(b)endcap region of thecalorimeter inseveral differentsignal samples. The averageprobabilityforthelow(high)scalarbosonmassesisabout 20% (55%) for

π

v decays occurring at radii between 2.0 m and 3.5 m in the barrel, andabout 6% (30%) for

π

v decays with

|

^z

|

between4.0 m and5.5 mintheendcaps.Theturn-ontakesplace before the inner edge of the HCAL asthe log₁₀

(

EH

/

EEM

)

cut al- lows for a small amount of energy in the ECal. The probability decreases towards the outer region of the HCalwhere too much of the energy escapes the HCal to pass the jet ET requirement.

The efficiencyislower intheendcapsbecauseeventstendto not satisfy the isolation criteriaduetothe increasedoccupancyfrom extracollisioneventsinthesamebunchcrossingasahard-scatter interaction(pile-up).

Events also contain a reconstructed primary vertex with at leastthree trackswith p_T

>

1 GeV.Events arerejectedifanyre-

Fig. 1.Theprobability(επv)forasingleπvtopassthetriggerasafunctionofthe πv(a)radialdecaylengthinthebarreland(b)thezpositionofthedecayvertex intheendcapsforseveralandπvmasses.

constructed jets show evidence of being caused by a beam-halo interaction [21]. A missing transverse momentum requirement, E^miss_T

<

50 GeV, is appliedto reject non-collision events,such as cosmicraysorbeam-halointeractions.

Inthe offline selection, jets are reconstructed with an anti-k_t algorithm with R

=

⁰

.

4, starting from calorimeter energy clus- terscalibratedusingthelocalclusterweighting method[24]. Jets arethencalibratedusinganenergy- and

η

-dependentsimulation- basedcalibration scheme. Jets are rejectedif they do not satisfy thestandardATLASgood-jetcriteriawiththeexceptionofrequire- mentsthat reject jetswithsmall electromagneticenergyfraction (EMF)[25].At least one jet must havefired the CalRatio trigger.

The jet matchingthe trigger mustpass an ET

>

60 GeV require- mentwhileasecondjetmustsatisfyanET

>

40 GeV requirement.

IfmorethanonejetfiredtheCalRatiotriggerthenonlytheleading jetisrequiredtohaveET

>

60 GeV.

Individually, all jets must satisfy

| η _| <

2

.

5, have log₁₀

(

EH

/

EEM

) >

1

.

2, andhaveno good tracks inthe ID with pT

>

1 GeV inaregion

R

<

0

.

2² centredonthejet axis.Agoodtrackmust haveatleasttwo hitsinthepixeldetectorandatotalofatleast nine hits in the pixel and SCT detectors. Fig. 2(a) compares the distribution of the number of good tracks associated with each jet in the multi-jet sample (described in the next section) with that in jets resulting from simulated

π

v decays in the HCAL or ID. Fig. 2(b) makes the same comparison forthe distribution of

2 R=

(η)²+(φ)².

Fig. 2.Distributionof(a)thenumberofgoodtracks(ntracks)withpT>1 GeV and R<0.2 aroundthejet axisand(b)thedistributionofjetlog10(EH/EEM)with jet |η|<2.5, pT>40 GeV.Thedashed histogramisforπv jetsdecayinginthe hadroniccalorimeter, andthedottedhistogramisforπvjetsdecayingintheID.

BotharefromthemH=^{126 GeV,m}πv=10 GeV sample.Thefilledhistogramisthe multi-jetdatasampleusedtoevaluatethemulti-jetcontributiontothebackground.

EventsarerequiredtosatisfyE^miss_T <50 GeV.

log₁₀

(

EH

/

EEM

)

ofeach jet.Themulti-jet data was gatheredusing aprescaled,single-jettriggerwitha15 GeV requirement.

Jetscausedbycosmicraysandbeam-halointeractionsareoften out-of-time. The jet timing is calculated by making an energy- weighted average ofthe timing foreach cellin thejet. Each cell isdefinedtohaveatimeof0 nsifitsenergyisrecordedatatime consistentwiththearrivalofa

β =

^{1 particlefromtheIP.Thetim-} ing ofeach jet isrequired tosatisfy

−

¹

<

t

<

5 ns.This cut will impact the efficiencyforlow

β π

v.Due to the requirementofa high-ETjetinthisanalysisthe

β

-distributionispeakednear1for lowmass

samples.Forthehighmass

samplesthedifference betweenm andmπv results in a large boost for the

π

v at the generatedlifetimes.As aresult,theinefficiencyintroducedby the timingcutisatworst1.5%fortheconsideredsamples.

Theanalysisrequiresthatexactlytwojetssatisfytheserequire- ments. The second jet requirement significantly reduces the SM multi-jet background contribution.Table 3 lists the final number ofexpectedeventsineachsignalMCsample.Thefinalnumberof eventsselectedindatais24.

5. Backgroundestimation

The largest contribution to the expected background comes fromSM multi-jetevents.Cosmic-ray interactionscontribute ata much lower level, and beam-halo interactions make a negligible contribution.

Toestimate the multi-jet backgroundcontribution, amulti-jet datasampleisusedtoderivetheprobability thatajetpassesthe

trigger and analysis selection. To obtain a raw background pre- diction, these jet probabilities are applied to a data sample that representsthemulti-jetbackgroundbeforeapplicationofjet-level analysisselection.Acorrectiontoaccount fortwo-jet correlations isappliedto thisrawpredictiontoyieldthe finalmulti-jet back- groundestimate.

The multi-jetdata sample contains events that pass single-jet triggers with an ET threshold of 15 GeV or higher. These trig- gerswere prescaledin2012andtheireffectiveluminositiesrange between 14

.

8 nb⁻¹ and 454

.

1 nb⁻¹. The dataset is representa- tive of thefull 20

.

3 fb⁻¹ ofdata collected in 2012and contains events from throughout the data collection period. The events are required to pass the analysis E^miss_T requirement and have at least two back-to-back (

φ >

2

.

0) jets with ET

>

40 GeV and

−

²

.

5

< η <

2

.

5.One jetisrequiredto satisfythemodified ATLAS good-jetcriteriausedbytheanalysis. Thesecondisusedtomea- suretwoprobabilities: one, called P,fora jetto passthetrigger andtheET

>

60 GeV jetrequirementandtheother,calledQ,fora jettosatisfytherequirement ET

>

40 GeV.Forboth P and Q the jetmustalsopassthelog₁₀

(

E_H

/

E_EM

)

,trackisolation,andallother analysisjet selection requirements includingthe modified ATLAS good-jetcriteria.Theprobabilitiesaredeterminedasafunctionof jet E_T and

η

.Thejet E_T and

η

dependenceiscalculatedindepen- dentlybecausethesampleisnotlarge.Asystematicuncertaintyto accountforanypotentialcorrelationisincludedintheanalysis.To calculatethissystematicuncertaintythechangeinthemeanE_T as afunctionof

η

andthechangeinthemean

η

asafunctionofET ismeasuredinthemulti-jetdatasample. Themaximumvariation is2%forboth E_Tand

η

.TheE_Tor

η

ofeachjetaresystematically shifted by thisamountas P and Q are recalculated.The new P andQ distributionsareusedtoestimatethemulti-jetbackground asdescribedbelow,andthemaximumvariationintheresult(6%) isusedasthesystematicuncertainty.

Binned fits of the probabilities as a function of ET are made to a Landau function andan exponential function for P and Q, respectively.The E_T requirementis ignoredwhen fittingtoallow thecurve tobestmatchthedistribution’sshape.Thefiterrorsare propagatedthroughtothesystematicinthemulti-jetbackground.

The

η

dependenceis strongly correlated withthe distribution of materialinthecalorimetersandcannotbe well describedby any simple functional form. Thus the probability is obtaineddirectly fromthe distribution. The P

(

E_T

)

parameterisation is additionally splitintoleadingjetandsub-leadingjetsamplesbecausetheprob- ability is different for the two types of jets. This effect is also presentforQ;however,itisaccountedforbythecorrectionforjet correlationsdiscussedbelow.Plotsof P

(

E_T

)

and Q

(

E_T

)

areshown in Fig. 3. The peak present in P

(

ET

)

is the resultof the trigger turn-onfor the full trigger chain.The trigger jetsare dominated byleadingjets,andsodominatedbytheleadingjet P.

Theprobability P isverifiedusingthe CalRatio-triggereddata.

TheCalRatio-triggeredeventsarerequiredtopassthesameevent selectionused toderivethe single-jetprobabilitiesaswell asthe requirements for calculating P. The CalRatio-triggered data con- tains501387 eventsthatfiredtheunprescaledCalRatiotriggerand passedtherequiredselection,andthesingle-jetprobabilitiespre- dict513000

±

⁹⁴000 (statisticalerroronly)events.

Tocalculatetherawmulti-jetbackgroundpredictiontheprob- abilities P and Q are applied to jets in events selected by the 15 GeV single-jet trigger. These single-jet probabilities are com- bined into an event probability using a combinatoric calculation thatrequiresatleastonejetintheeventtofirethetriggerandex- actlytwojetstopassallthejetselectioncriteria.Theeventprob- abilityisscaledtoaccountforsingle-jettriggerprescales,yielding aweightforeachevent.Thesumofallweightsinthedatasample yieldsarawbackgroundpredictionof13

.

2

±

²

.

9 (statistical

⊕

^sys-

Fig. 3.Theprobability,P,thatajetfromthemulti-jetdatasamplepassesthetrig- gerandalljetrequirementsincludingtheET>60 GeV requirementisshownin (a).

A Landau functionis fittedtothe leadingand sub-leadingjet distributionssep- arately(solidand dashedlines).Theprobability, Q,topassalljetrequirements includingthe ET>40 GeV requirementasafunctionofjetETisshownin (b).An exponentialfunctionisfittedtothedistribution(solidline).TheETrequirementis ignoredwhenfittingtoallowthecurvetobestmatchtheshapeofthedata,butis usedexplicitlywhenPandQareappliedtoajet.Themulti-jet datawasgathered usingarangeofprescaled,single-jettriggers.

tematic error)events.The uncertainty isdominated by thesmall numberofjetsfiringtheCalRatiotriggerinmulti-jetevents.

In multi-jet eventsthe log₁₀

(

E_H

/

E_EM

)

and trackisolation val- ues of one jet are correlated withthose of the second jet. Ifan event contains one jet of high log₁₀

(

E_H

/

E_EM

)

, the second jet is morelikelytohavehighlog₁₀

(

E_H

/

E_EM

)

aswell.Likewise,ifonejet has notracksassociated withit,the other ismorelikely tohave no associatedtracksaswell. Thesingle-jetprobabilitiesabove ig- nore thiscorrelation because each is calculatedindependently of the log₁₀

(

E_H

/

E_EM

)

andn_track of other jets inthe event. As a re- sult, Q islower thanifitwere calculatedonlyineventswithan accompanyinglown_track,highlog₁₀

(

E_H

/

E_EM

)

jet.

Ascale factortoaccountforthecorrelation iscalculatedfrom the multi-jetdata sampleandthe CalRatio-triggereddatasample by examiningnumbersofeventsinregions inthelog₁₀

(

E_H

/

E_EM

)

andnumber-of-tracks(n_track)planethat areoutsidethesignal re- gion(log₁₀

(

E_H

/

E_EM

) >

1

.

2 andn_track

=

^0).Thelog10

(

E_H

/

E_EM

)

bin- ningischosensuchthatbinningisuniforminEMF.³ Arangefrom 0to7wasusedforn_track.Theregionsoutsidethesignalregionare expectedtohaveverylittlesignalcontamination.

Ineachregiontheratioofthenumberofeventsobservedinthe CalRatio-triggereddatatotherawpredictioniscalculated.Twose- ries ofratios are calculated, one as a function of log₁₀

(

E_H

/

E_EM

)

andone asa functionofn_track.Todeterminethetrendinthera-

3 log₁₀(EH/EEM)=^log10((1−^EMF)/EMF).

tioasa function oflog₁₀

(

EH

/

EEM

)

the n_track requirementis held constant:ajetisrequiredtohave5or6tracks.Theratioisthen determined forseveralnon-overlapping ranges of log₁₀

(

EH

/

EEM

)

. The same procedure is used for n_track by requiring jets to have 0

.

55

<

EMF

<

0

.

65.Because theratioistakenwithrespecttothe observeddata,thisratiowill correctforanynormalisation errors inP orQ.

Bothsetsofratiosarefittedtoallowextrapolationintothesig- nalregion.Theproductofthetworatiosinthesignalregionyields a scale factor to correct for the correlation betweenjets. A sys- tematicerrorisaddedtoaccountfortheassumptionthatthetwo ratiosareuncorrelated.Thecalculatedscalefactoris1

.

8

±

⁰

.

5.The uncertaintyonthescalefactorisduetothelimitedsamplesize.

Toverifytheprocedureeightotherbins onthelog₁₀

(

EH

/

EEM

)

and ntrack plane were chosen and the full background predic- tion method was applied. Because signal contamination is negli- gibleoutsideofthesignal region,thepredictednumberofevents can be directly compared to the number of events in the same log₁₀

(

EH

/

EEM

)

–ntrack region in the CalRatio-triggered data. In all casesthepredictionisconsistentwithdatatowithinonestandard deviation.

The final multi-jet prediction is 23

.

2

±

⁸

.

0 (statistical

⊕

^sys- tematicerror)eventsinthesignalregion.Theuncertaintyisdom- inatedby the statisticaluncertainty, which isin turn dominated bythe smallnumberofjetsmatching theCalRatio trigger inthe multi-jetdatasample.Thesystematiccontributioncomesfromthe correlation between ET and

η

aswell asfromthe inclusionof a requirementon

φ

(notusedinthesignalselection)inthedeter- minationof P and Q.

Particlesfroma cosmic-ray showermaypass throughandde- posit energy in the calorimeter without passing through the ID.

Theseenergydeposits can bereconstructed astrackless jets. The overallcontributiontotheexpectedbackgroundisreducedbythe jet-timingandE^miss_T requirements.

Thecosmic-raybackgroundwasstudiedusingatriggersimilar to the CalRatio trigger, but active only during an empty cross- ing.Eachproton beamisdividedup into buckets,mostofwhich arefilledwithprotons.Anempty crossingoccurswhenan empty bucketineach beamcoincides in thecentre ofthedetector, and fivebucketsoneithersideineachbeamarealsoempty.Datagath- eredfrom these empty crossings are used to study backgrounds thatarenotbeamrelated.

Theanalysisselection, withtheexception ofthejet-timingre- quirement and the good-vertex requirement, are applied to all events triggered in empty crossings. The

−

¹

<

t

<

5 ns timing requirementisremovedtoretainmoreeventstogiveamoreaccu- ratedeterminationofthebackground.Asimplescalingcanbeused to predict the expected cosmic-ray eventrate within the timing windowbecausethearrivaltimeofcosmic-raymuonsisuniformly distributed.It is found that about 5% ofcosmic-ray eventsfiring thetriggerandcontainingtwojetsareeventswherebothjetssat- isfythe

−

¹

<

t

<

5 ns requirement.

Twoadditional corrections are applied to determine the final backgroundpredictionduetocosmic-rayevents.Thefirstaccounts forthedifferent live-times ofthe triggers.The number ofempty crossingsis2.9timessmallerthanthenumberusedtocollectthe fulldataof20

.

3 fb⁻¹.Thesecondcorrectionweightseacheventto accountforsofttracksduetopile-upandunderlying-eventeffects thatwouldhavecausedthejet tofail thetrackisolation require- menthaditoccurredinacollisionenvironment.Todeterminethe weights a trigger that selects random collision events is used to determine the probability as a function of

η

that a track with pT

>

1 GeV is presentin a

R

<

0

.

2 cone anywhere in the de- tectorasafunctionof

η

.Thisprobabilityisappliedtoeachjetin eacheventtodetermineaneventweight.Theeventweightsrange

from0.55to0.63. Combining allthe correctionsresultsinapre- dictednumberofcosmic-rayeventsof0

.

3

±

⁰

.

2 (statisticalerror).

Anotherpossiblebackgroundcontributioncomesfromabeam- halomuonthatundergoesbremsstrahlungintheHCal.Twoselec- tioncriteriareducethistype ofbackground.Ajet-timingrequire- mentisimposedbecausemostofthejetsproducedbybeam-halo interactions arenot coincidentintime withjetsfrompp interac- tions.Inaddition,eventsare rejectedwhentracksegments inthe endcapmuonchambers,fromtheenteringbeam-halomuon,align in

φ

witha jet. These two requirements reduce the background considerablywithnodiscernibleeffectonthesignal.

Unpaired isolated crossings, i.e. crossings where only protons froma single beamarepresentandatleastthree bucketson ei- thersideoftheemptybeam’sbucketarealsoempty, canbeused to studybeam-halo events.Toestimate thisbackground,artificial eventsare created bysamplingtwo jetsfrom acollection ofjets passingbothaCalRatiotriggeractiveonlyduringunpairedisolated crossingsandtheleadingjet requirementsfromunpairedisolated crossings.AllpossiblepairsofjetsareusedandtheE^miss_T

<

50 GeV requirementisapplied toeach constructedevent. The numberof jets passing the jet analysis selection and the fraction of con- structed eventssatisfying the E^miss_T requirementare combined to estimate the background. This method, which also accounts for cosmic-ray muoncontamination, predicts 0

.

07

±

⁰

.

07 events.The largeuncertaintyisdueprimarilytothesmallnumberofjetspass- ingallrequiredcuts.

Backgrounds from combinations of these non-beam interac- tions,i.e.abeam-halojetplusamulti-jet,orabeam-halojetplus ajetduetoacosmic-raymuon,werefoundtobenegligible.

6. Systematicuncertainties

Table 2 presentsa summary ofsystematic uncertainties asso- ciated with the signal sample. The overall uncertainty, taken as the sum in quadrature ofall positive and negative contributions respectively, is listed inthe last column. The MC signal samples’

statisticaluncertaintyisshowninTable 3anditisaccountedforin thestatisticalanalysis.Theoverallnormalisation uncertaintyofthe integratedluminosityis2

.

8% obtainedfollowingthesamemethod- ology asthat detailed inRef. [26] froma preliminary calibration of the luminosity scale derived from beam-separation scans per- formed in November 2012. The uncertainties on the Higgs bo- son production cross-sectionsat

√

s

=

^{8 TeV, which are equal to} the uncertainties on the

production cross-sections, are about 10%[18].

The uncertainty onthe signal MC samples dueto parton dis- tribution functions(PDF) iscalculated by reweighting each event usingthreedifferentPDF sets(MSTW2008nlo68cl[17],CT10[27], and NNPDF2.3 [28]) and their associated error sets. The RMS change inacceptanceforthe errorsets ofeach PDF is calculated andcombined withthedifference in acceptancesforeach of the threePDFs.

Pile-up primarilyaffectstheacceptanceby addingextra tracks and degrading the track isolation of a jet. All MC samples are reweighted to reproduce the observed distribution of the num- ber of interactions per bunch crossing inthe data. To determine ifpile-upissimulatedproperlyintheMC samples,adirectcom- parisonofdataandMCmulti-jetsamplesisperformed.ThejetET, EMF,

η

,

φ

,associatedtracksandtimingdistributionsasafunction ofthemeannumberofpile-upinteractionsare comparedindata and MC simulation. A 10% systematic uncertainty is assigned to theacceptancecoversalltheobserveddifferences.

Thejetenergyscale(JES)uncertaintyisevaluatedasafunction ofthe jetEMF and

η

,followingthesame strategyused inthein situ jet energy intercalibration [24]. The JES is rederived forlow

EMFjetsforthisanalysis.The relativejet calorimeterresponseis studied by balancing the transverse momenta of dijets. The sys- tematic uncertainty is obtained by comparing the pT-balance in datato the p_T-balancein MC samples.Thisdifference is usedto calculatea differenceintheJESindataandMCsimulationandis propagated to the signal MC samplesto get a systematic uncer- taintyontheacceptance.Thisstudyalsoprovidesausefulperfor- mancecomparisonbetweendataandMCjetsthatresemblesignal jets.

The E_T^miss uncertaintyaccountsforvariations inmissingtrans- versemomentumscaleandresolution[29].Thetimingsystematic accountsformismodelingofjettimingbetweenMCanddata.Both oftheseuncertaintiesweredeterminedbysmearingtheassociated cuttodeterminetheimpactontheacceptance.

Thesimulationofthetriggerisverifiedbycomparingtheper- formanceofthetriggerindatawiththeperformanceinMCsimu- lationonthesamemulti-jetsampleusedtoevaluatethemulti-jet background.Each triggerrequirement isstudied individually: the jet ET,the log₁₀

(

EH

/

EEM

)

, and the track isolation. Each require- mentisadjustedtomaketheperformance matchindataandMC events,andtheresultingdifferencesinacceptancefromthenom- inalacceptance,foreachrequirement, areaddedinquadratureto determinethesystematicuncertainty.

The simulation of initial state radiation (ISR) cannot be di- rectly verified because it is difficult to uniquely identify ISR jets indata [30].An incorrectISR ratein thesimulation impacts the acceptance by altering the number of jets in the event and by altering the boost of the

boson. Each of these is studied in- dependently. The ISR jet population is alteredevent by eventso thatthenumberofISRjetsishalvedordoubled(jetsinMCsam- ples are labelled as containing ISR if they contain a gluon with pT

>

2 GeV).Thepopulation of

π

v jetsisnotalteredbythispro- cess,butanaddedISRjetmayoverlapwithoneofthe

π

vjets.The effectofa boostcausedbyan ISRjetisstudiedby exploitingthe correlationbetweenthe

π

v jetET andthe

boost.FromRef.[30], the

p_Tspectrumhasanuncertaintyof5%,whichdirectlycorre- lateswitha5% uncertainty inthe

π

v jetenergy.Tocalculatethe systematicuncertaintyassociatedwiththeboost,thepT ofISRjets isconservativelyvariedby5%andthechangeinacceptanceisob- served.

ThechangesinacceptancefrombothsourcesofISRuncertainty aretakenascorrelatedsystematicerrorsandaddedtogetthetotal systematicforISRsimulation.

Anincorrectsimulationoffinalstateradiation(FSR)hasaneg- ligibleeffectontheanalysis’acceptance.FSRcanoccurinaprompt ordisplacedjet.Butevenifdisplaced,theextrajetcannotdegrade trackisolation or depositextra energy in the ECalif the

π

v has decayedintheHCal.

7. Resultsandexclusionlimits

Theglobalacceptanceoftheselectedeventtopologyinthesig- nal MC samples is a function ofm,mπv and the proper decay length of the

π

v. At a proper decay length of 1.5 m the accep- tancerangesfrom0.07%to0.61%.Themain efficiencyloss isdue to the low probability that both

π

v decay inside the calorime- ter. High mass samples suffer further efficiency loss due to the E^miss_T requirement. Table 3 lists the expected number of events from all signal MC samples and the background expectation in 20

.

3 fb⁻¹. The mH

=

126 GeV mass samples use the SM Higgs boson cross-sections of

σ

SM

=

¹⁹

.

0 pb for the gluon fusion pro- cess:other production modesare ignored.The numberof events observedin data,24,isalso shownforcomparison. No excessof events is observed since the expected background is 23

.

5

±

⁸

.

0.

The CL_s method [31] is used to derive an upper limit on the

Table 2

SummaryofsystematicuncertaintiesfortheandHiggsbosonproductioncross- section,jetenergyscale,trigger,missingtransversemomentum,andtherequire- mentonjettimingasapercentageofthesignalyield.Systematicerrorsthathave commonvaluesacrosssamplesarenotlisted(pile-upat10%,ISRat⁺₋²₁^._.⁹₂%,andPDF at2.1%).Thelastcolumnreportsthetotalsystematicuncertainty(includingthelu- minosityandcommonsystematicerrors).

Sample mH,mπv

[GeV]

Hσ [%]

JES [%]

Trigger [%]

E^miss_T [%]

Time Cut [%]

Total [%]

126, 10 ⁺₋¹⁰₁₀^._.⁴₄ ⁺₋²₂^._.²₇ ±1.1 ⁺₋⁵₂^._.⁵₄ ⁺₋¹₆^._.⁶₆ ⁺₋¹⁶₁₆^._.⁴₇ 126, 25 ⁺₋¹⁰₁₀^._.⁴₄ ⁺₋¹₁^._.⁵₆ ±¹.3 ⁺₋³₁^._.¹₈ ⁺₋⁰₃^._.⁸₃ ⁺₋¹⁵₁₅^._.⁶₅ 126, 40 ⁺₋¹⁰₁₀^._.⁴₄ ⁺₋²₆^._.⁶₂ ±¹.1 ⁺₋⁷₄^._.⁷₆ ⁺₋¹₅^._.⁹₉ ⁺₋¹⁸₁₆^._.²₉ Sample

m,mπv

[GeV]

σ

[%]

JES [%]

Trigger [%]

E^miss_T [%]

Time Cut [%]

Total [%]

100, 10 ⁺₋¹¹₁₀^._.¹₆ ⁺₋²₄^._.³₀ ±⁰.1 ⁺₋⁴₃^._.⁶₄ ⁺₋²₉^._.⁷₅ ⁺₋¹⁶₁₈^._.⁷₅ 100, 25 ⁺₋¹¹₁₀^._.¹₆ ⁺₋⁵₃^._.⁵₇ ±¹.2 ⁺₋³₂^._.⁴₅ ⁺₋¹₀^._.⁷₇ ⁺₋¹⁷₁₅^._.⁰₈ 140, 10 ⁺₋¹⁰₁₀^._.¹₃ ⁺₋⁰₁^._.⁶₁ ±⁰.5 ⁺₋⁴₅^._.⁰₆ ⁺₋¹₆^._.⁹₆ ⁺₋¹⁵₁₇^._.⁶₂ 140, 20 ⁺₋¹⁰₁₀^._.¹₃ ⁺₋¹₁^._.²₆ ±1.0 ⁺₋⁴₃^._.⁰₉ ⁺₋⁰₅^._.⁴₀ ⁺₋¹⁵₁₆^._.⁵₂ 140, 40 ⁺₋¹⁰₁₀^._.¹₃ ⁺₋¹₁^._.³₆ ±1.5 ⁺₋⁶₄^._.³₆ ⁺₋¹₂^._.⁸₄ ⁺₋¹⁶₁₅^._.⁵₈ 300, 50 ⁺₋⁹₁₀^.6_.0 ⁺₋⁰₀^._.¹₃ ±0.3 ⁺₋⁹₇^._.⁰₄ ⁺₋⁰₃^._.⁵₀ ⁺₋¹³₁₃^._.⁹₃ 600, 50 ⁺₋¹¹₁₀^._.²₁ ⁺₋⁰₀^._.⁰₁ ±0.2 ⁺₋¹¹₁₁^._.⁷₃ ⁺₋²₄^._.²₄ ⁺₋¹⁷₁₆^._.⁰₂ 600, 150 ⁺₋¹¹₁₀^._.²₁ ⁺₋⁰₀^._.²₂ ±0.3 ⁺₋¹¹₁₀^._.⁵₂ ⁺₋²₅^._.⁷₃ ⁺₋¹⁷₁₅^._.⁵₁ 900, 50 ⁺₋¹²₁₁^._.⁸₅ ⁺₋⁰₀^._.⁰₁ ±0.1 ⁺₋¹²_9.7^{.6 +}₋¹₃^._.⁰₇ ⁺₋¹⁸₁₅^._.⁵₉ 900, 150 ⁺₋¹²₁₁^._.⁸₅ ⁺₋⁰₀^._.²₃ ±⁰.2 ⁺₋¹¹₁₀^._.⁸₉ ⁺₋⁰₂^._.⁹₅ ⁺₋¹⁸₁₆^._.¹₃

σ () ×

^BR

( → π

v

π

v

)

. A profile likelihood ratio is used as the test statistic and a frequentest calculatoris used to generatetoy data. The likelihood includes a Poisson probability term describ- ing the total number of observed events. Systematic uncertain- ties areincorporated asnuisanceparameters through their effect on the mean of the Poisson functions and through convolution with their assumed Gaussian distributions. The number of ex- pected events in signal MC samples, together with the estimate ofexpectedbackground,theobservedcollision eventsandall the systematic uncertainties areprovided asinput for computingthe CL_s value, which represents the probability for the given obser- vation to be compatible with the signal

+

background hypothe- sis.

The acceptanceisa functionof the

mass,the

π

v massand

π

v properdecaylength.Toextrapolatetothenumberofexpected eventsatdifferentproperdecaylengths,alargesampleof

π

v de- cays is generated in a range from 0 to 50 m and an efficiency map asafunctionof

π

v boostisusedtodeterminetheefficiency ateachdecaylength.Theresultingefficienciesarethenconverted intothe finalnumberofexpectedeventsshowninFig. 4.Finally, Figs. 5 and 6 show the observed limit distribution for the three 126 GeV Higgs samplesandfortheother

samplesrespectively.

The derived 95% confidencelevel (CL) excluded ranges of proper decaylengthare listedinTable 4forthemH

=

126 GeV samples, under the alternative assumptions of a 30% BRor a 10% BR for H

→ π

v

π

v.

8. Summaryandconclusions

Asearchforthedecayofascalarbosoninthemassrangefrom 100 GeV to900 GeV,includingasearch foranexoticdecayofthe Higgs boson,to a pair oflong-lived neutralparticles decaying in the ATLAS hadroniccalorimeterhas beenpresented.The analysis is basedon20

.

3 fb⁻¹ of pp collisionsat

√

s

=

8 TeV collectedin 2012bytheATLASexperimentattheLHC.

Table 3

Summaryofexpectednumberofsignalevents,expectedbackgroundpresentinthe datasample,andtheobservednumberofeventsin20.3 fb⁻¹.Theglobalacceptance isalsogiven.Theerroronthesignalsamplesisstatisticalonly,theerroronthe expectedbackgroundisstatistical ⊕systematic.All resultsarenormalised fora properdecaylengthoftheπvof1.5 m.A100%branchingratioforhs→πvπvis assumed.

Sample (mH,mπv[GeV])

Expected yields

Global acceptance (%)

126, 10 536±^{23 0}.139±⁰.006

126, 25 941±44 0.244±0.011

126, 40 365±31 0.095±0.008

Sample (mH,mπv[GeV])

Expected yields

Global acceptance (%)

100, 10 440±^{29 0}.073±⁰.005

100, 25 424±^{37 0}.070±⁰.006

140, 10 525±^{20 0}.168±⁰.006

140, 20 900±^{37 0}.287±⁰.012

140, 40 641±30 0.205±0.010

300, 50 444±11 0.609±0.015

600, 50 35±1 0.330±0.010

600, 150 41±^{2 0}.386±⁰.015

900, 50 3.5±⁰.1 0.304±⁰.011

900, 150 4.6±⁰.2 0.397±⁰.016

Background Expected events

SM Multi-jets 23.2±⁸.0

Cosmic rays 0.3±0.2

Total Expected Background 23.5±8.0

Data 24

Table 4

Rangesofπvproperdecaylengthsexcludedat95%CLassuminga30%anda10%

BRforamH=126 GeV.

MCsample mH,mπv

[GeV]

Excludedrange 30% BRH→πvπv [m]

Excludedrange 10% BRH→πvπv [m]

126, 10 0.10–6.08 0.14–3.13

126, 25 0.30–14.99 0.41–7.57

126, 40 0.68–18.50 1.03–8.32

Nosignificantexcessofeventsisobservedoverthebackground estimate. Limits are set on the

π

v proper decay lengths for dif- ferent scalar boson and

π

v mass combinations. For a SM Higgs decayingto

π

v properdecaylengthsbetween0.10 mand18.50 m assuminga30%BRareruledout,andbetween0.14 mand8.32 m assuming a BR of 10%. Results for low mass

(100 GeV and 140 GeV)andhighmass

(300 GeV,600 GeV,and900 GeV)have alsobeenpresentedasafunctionofproperdecaylength.

Acknowledgements

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

We acknowledge the support of ANPCyT, Argentina; YerPhI, Armenia; ARC,Australia; BMWFW andFWF,Austria; ANAS, Azer- baijan; SSTC, Belarus; CNPq and FAPESP, Brazil; NSERC, NRC and CFI, Canada; CERN; CONICYT, Chile; CAS, MOST and NSFC, China;COLCIENCIAS, Colombia; MSMT CR, MPO CRand VSC CR, Czech Republic; DNRF, DNSRC and Lundbeck Foundation, Den- mark; EPLANET, ERC and NSRF, European Union; IN2P3-CNRS, CEA-DSM/IRFU,France;GNSF,Georgia;BMBF,DFG,HGF,MPG and AvHFoundation,Germany;GSRTandNSRF,Greece;ISF,MINERVA,

Fig. 4.Numberofeventsexpectedtopasstheanalysisselectionin20.3 fb⁻¹asa functionoftheπvproperdecaylengthfor(a)threelow-massdatasetsand(b)three high-massdatasets.100%branchingratiosforhs→πvπvareassumed.

Fig. 5.Observed95%CLlimitsonσ/σSMformH=^{126 GeV asafunctionofthe} πvproperdecaylength:thesolidlineisformπv=^{10 GeV,the}short-dashedline isformπv=^{25 GeV,the}long-dashedlineisformπv=^{40 GeV.The}σSMistaken tobe19.0 pb.ThehorizontalsolidlinecorrespondstoBR=^{30% andthehorizontal} dashedlinetoBR=10%.

GIF, I-CORE and Benoziyo Center, Israel; INFN, Italy; MEXT and JSPS, Japan; CNRST, Morocco; FOM and NWO, Netherlands; BRF and RCN, Norway; MNiSW and NCN, Poland; GRICES and FCT, Portugal; MNE/IFA, Romania; MES of Russia andROSATOM, Rus- sian Federation; JINR; MSTD, Serbia; MSSR, Slovakia; ARRS and MIZŠ, Slovenia; DST/NRF, South Africa; MINECO, Spain; SRC and Wallenberg Foundation, Sweden;SER, SNSF and Cantons of Bern and Geneva, Switzerland; NSC, Taiwan; TAEK, Turkey; STFC, the

Fig. 6.Observed 95%CLlimitsonσ×BR [pb]for (a)low and(b)high-mass samples.

Royal Society and Leverhulme Trust, United Kingdom; DOE and NSF,UnitedStatesofAmerica.

The crucial computingsupport fromall WLCG partners is ac- knowledged gratefully, in particular from CERN and the ATLAS Tier-1 facilities at TRIUMF (Canada), NDGF (Denmark, Norway, Sweden), CC-IN2P3 (France), KIT/GridKA (Germany), INFN-CNAF

(Italy), NL-T1(Netherlands),PIC(Spain), ASGC(Taiwan),RAL (UK) andBNL(USA)andintheTier-2facilitiesworldwide.

References

[1]F.Englert,R.Brout,Phys.Rev.Lett.13(1964)321.

[2]P.Higgs,Phys.Lett.12(1964)132.

[3]G.S.Guralnik,C.R.Hagen,T.W.B.Kibble,Phys.Rev.Lett.13(1964)585.

[4]ATLASCollaboration,Phys.Lett.B716(2012)1,arXiv:1207.7214[hep-ex].

[5]CMSCollaboration,Phys.Lett.B716(2012)30,arXiv:1207.7235[hep-ex].

[6]S.Chang,P.J.Fox,N.Weiner,JHEP0608(2006)068,arXiv:hep-ph/0511250.

[7]S.Chang,R.Dermisek,J.F.Gunion,N.Weiner,Annu.Rev.Nucl. Part.Sci.58 (2008)75–98,arXiv:0801.4554[hep-ph].

[8]M.J. Strassler,K.M. Zurek,Phys. Lett. B651(2007) 374–379,arXiv:hep-ph/

0604261.

[9]M.J. Strassler, K.M.Zurek,Phys. Lett. B 661(2008) 263–267, arXiv:hep-ph/

0605193.

[10]ATLASCollaboration,NewJ.Phys.13(2011)053033,arXiv:1012.5104[hep-ex].

[11]ATLASCollaboration,Phys.Rev.Lett.108(2012)251801,arXiv:1203.1303[hep- ex].

[12]CMSCollaboration,JHEP1302(2013)085,arXiv:1211.2472[hep-ex].

[13]D0Collaboration,V.Abazov,etal.,Phys.Rev.Lett.103(2009)071801,arXiv:

0906.1787[hep-ex].

[14]CDFCollaboration,T.Aaltonen,etal.,Phys. Rev.D85(2012)012007,arXiv:

1109.3136[hep-ex].

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

[16]T. Sjostrand, S. Mrenna, P. Skands, Comput. Phys. Commun. 178 (2008) 852–867,arXiv:0710.3820[hep-ph].

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

[18]S.Dittmaier,C.Mariotti,G.Passarino,R.Tanaka(Eds.),LHCHiggsCrossSection WorkingGroup,CERN,Geneva,2011,CERN-2011-002,arXiv:1307.1347[hep- ex].

[19]ATLAS Collaboration, Eur. Phys. J. C 70 (2010) 823–874, arXiv:1005.4568 [physics.ins-det].

[20]GEANT4 Collaboration, S. Agostinelli, et al., Nucl. Instrum. MethodsA 506 (2003)250–303.

[21]ATLASCollaboration,J.Instrum.8(2013)P07015,arXiv:1305.2284[hep-ex].

[22]R.Achenbach,etal.,J.Instrum.3(2008)P03001.

[23]M.Cacciari,G.P.Salam,G.Soyez,JHEP0804(2008)0633,arXiv:0802.1189[hep- ph].

[24]ATLASCollaboration,Eur.Phys.J.C73(2013)2304,arXiv:1112.6426[hep-ex].

[25] ATLAS Collaboration, ATLAS-CONF-2010-038, https://cdsweb.cern.ch/record/

1277678.

[26]ATLASCollaboration,Eur.Phys.J.C73(2013)2518,arXiv:1302.4393[hep-ex].

[27]J.Gao,etal.,Phys.Rev.D89(2014)033009,arXiv:1302.6246[hep-ex].

[28]R.D.Ball,etal.,Nucl.Phys.B867(2013)244,arXiv:1207.1303[hep-ex].

[29] ATLASCollaboration,ATLAS-CONF-2013-082,http://cds.cern.ch/record/1570993.

[30]D.deFlorian,G.Ferrera,M.Grazzini,D.Tommasini,JHEP1111(2011)064, arXiv:1109.2109[hep-ex].

[31]A.Read,J.Phys.G28(2002)2693.

ATLASCollaboration

G. Aad ⁸⁴ , B. Abbott ¹¹² , J. Abdallah ¹⁵² , S. Abdel Khalek ¹¹⁶ , O. Abdinov ¹¹ , R. Aben ¹⁰⁶ , B. Abi ¹¹³ , M. Abolins ⁸⁹ , O.S. AbouZeid ¹⁵⁹ , H. Abramowicz ¹⁵⁴ , H. Abreu ¹⁵³ , R. Abreu ³⁰ , Y. Abulaiti ^{147a , 147b} , B.S. Acharya ^{165a , 165b ,}

^a

, L. Adamczyk ^38a , D.L. Adams ²⁵ , J. Adelman ¹⁷⁷ , S. Adomeit ⁹⁹ , T. Adye ¹³⁰ , T. Agatonovic-Jovin ^13a , J.A. Aguilar-Saavedra ^{125a , 125f} , M. Agustoni ¹⁷ , S.P. Ahlen ²² , F. Ahmadov ^{64 ,}

^b

, G. Aielli ^{134a , 134b} , H. Akerstedt ^{147a , 147b} , T.P.A. Åkesson ⁸⁰ , G. Akimoto ¹⁵⁶ , A.V. Akimov ⁹⁵ ,

G.L. Alberghi ^{20a , 20b} , J. Albert ¹⁷⁰ , S. Albrand ⁵⁵ , M.J. Alconada Verzini ⁷⁰ , M. Aleksa ³⁰ , I.N. Aleksandrov ⁶⁴ , C. Alexa ^26a , G. Alexander ¹⁵⁴ , G. Alexandre ⁴⁹ , T. Alexopoulos ¹⁰ , M. Alhroob ^{165a , 165c} , G. Alimonti ^90a , L. Alio ⁸⁴ , J. Alison ³¹ , B.M.M. Allbrooke ¹⁸ , L.J. Allison ⁷¹ , P.P. Allport ⁷³ , A. Aloisio ^{103a , 103b} , A. Alonso ³⁶ , F. Alonso ⁷⁰ , C. Alpigiani ⁷⁵ , A. Altheimer ³⁵ , B. Alvarez Gonzalez ⁸⁹ , M.G. Alviggi ^{103a , 103b} , K. Amako ⁶⁵ , Y. Amaral Coutinho ^24a , C. Amelung ²³ , D. Amidei ⁸⁸ , S.P. Amor Dos Santos ^{125a , 125c} , A. Amorim ^{125a , 125b} , S. Amoroso ⁴⁸ , N. Amram ¹⁵⁴ , G. Amundsen ²³ , C. Anastopoulos ¹⁴⁰ , L.S. Ancu ⁴⁹ , N. Andari ³⁰ ,

T. Andeen ³⁵ , C.F. Anders ^58b , G. Anders ³⁰ , K.J. Anderson ³¹ , A. Andreazza ^{90a , 90b} , V. Andrei ^58a , X.S. Anduaga ⁷⁰ , S. Angelidakis ⁹ , I. Angelozzi ¹⁰⁶ , P. Anger ⁴⁴ , A. Angerami ³⁵ , F. Anghinolfi ³⁰ ,

A.V. Anisenkov ^{108 ,}

^c

, N. Anjos ¹² , A. Annovi ⁴⁷ , A. Antonaki ⁹ , M. Antonelli ⁴⁷ , A. Antonov ⁹⁷ , J. Antos ^145b ,