Measurement of the dependence of transverse energy production at large pseudorapidity on the hard-scattering kinematics of proton–proton collisions at $\sqrt{s}=2.76$ TeV with ATLAS

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Measurement of the dependence of transverse energy production at large pseudorapidity on the hard-scattering kinematics of

proton–proton collisions at √

s = ² . 76 TeV with ATLAS

.ATLASCollaboration^

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

Articlehistory:

Received2December2015

Receivedinrevisedform22February2016 Accepted24February2016

Availableonline2March2016 Editor:W.-D.Schlatter

The relationship betweenjet productionin the central regionand the underlying-eventactivity ina pseudorapidity-separated region is studied in 4.0 pb⁻¹ of √

s=².76 TeV pp collision data recorded with the ATLASdetector atthe LHC. Theunderlying event is characterisedthrough measurementsof theaveragevalueofthesumofthetransverseenergyatlargepseudorapiditydownstreamofoneofthe protons,whicharereportedhereasafunctionofhard-scatteringkinematicvariables.Thehardscattering ischaracterisedbytheaveragetransversemomentumandpseudorapidityofthetwohighesttransverse momentumjetsintheevent.Thedijetkinematicsareusedtoestimate,onanevent-by-eventbasis,the scaled longitudinal momentaof the hard-scatteredpartonsin thetarget and projectile beam-protons moving towardandaway fromtheregionmeasuringtransverseenergy,respectively.Transverseenergy productionatlargepseudorapidityisobservedtodecreasewithalineardependenceonthelongitudinal momentumfraction inthe targetprotonand todepend onlyweaklyonthat intheprojectileproton.

TheresultsarecomparedtothepredictionsofvariousMonteCarloeventgenerators,whichqualitatively reproducethetrendsobservedindatabutgenerallyunderpredicttheoverallleveloftransverseenergy atforwardpseudorapidity.

©^{2016CERNforthebenefitoftheATLAS}Collaboration.PublishedbyElsevierB.V.Thisisanopen accessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

1. Introduction

Propertiesofthe underlyingeventatlarge rapidity inproton–

proton (pp) collisions in the presence of a hard parton–parton scatteringaresensitive tomanyfeatures ofhadronicinteractions.

Previousstudiesoftheunderlyingeventmainlyfocused onprob- ing theregion transverseto final-statejetsat mid-rapidity[1–4].

This Letter presents a study of the transverse energy produced atsmallangles withrespectto theproton beam,a region where particleproductionmaybeparticularlysensitivetothecolourcon- nectionsbetweenthehard partonsandthebeamremnants. Such measurements are neededto constrain particle productionmod- els,whichsystematically underpredictthetotaltransverse energy atforwardrapiditiesinhard-scatteringevents[4].

Measurementsoftransverse energyproduction atlarge rapid- ityare also neededto aid in the interpretation of recent results onjetproductioninproton–lead(p+^{Pb)collisions[5,6].Inthese} collisions, hard scattering rates are expected to grow with the increasing degree of geometric overlap between the proton and the nucleus. Simultaneously, the level of overlap is traditionally

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

thoughttobereflectedintherateofsoftparticleproduction,par- ticularlyatlargepseudorapidityinthenucleus-goingdirection.The recent resultsfound that single anddijet productionratesin the proton-going (forward, or projectile) direction are relatedto the underlying-event activityin the nucleus-going (backward,or tar- get)directioninawaythatcontradictsthemodelsofhowjetand underlying-event productionshould correlate.Specifically,the av- erage transverse energyproduced in the backward directionwas found to systematically decrease, relative to that for low-energy jetevents,withincreasingjetenergy.Thisdecreaseresultedinan apparentenhancementofthejetrateinlow-activity,orperipheral, eventsandasuppressionofthejetrateinhigh-activity,orcentral, events.

Theseresultshaveseveralcompetinginterpretations.Forexam- ple, theyare takenasevidencethat protonconfigurations witha parton carryinga large fractionx of the protonlongitudinal mo- mentuminteractwithnucleonsinthenucleuswithasignificantly smallerthanaveragecross-section[7].Alternatively,otherauthors have argued that in the constituentnucleon–nucleon (N N)colli- sions,energyproductionatbackwardrapiditiesnaturallydecreases withincreasing x intheforward-goingproton,eitherthroughthe suppression of soft gluons available for particle production [8]

or froma rapidity-separatedenergy-momentumconservation be-

http://dx.doi.org/10.1016/j.physletb.2016.02.056

0370-2693/©^{2016CERNforthebenefitoftheATLAS}Collaboration.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

tween the hard process and soft production [9]. More generally, themodificationofsoftparticleproductioninN N collisionsinthe presenceof a hardprocess is expectedto affectestimates ofthe collisiongeometryofp+Pb collisionswithahardscatter[10–12].

Thus a control measurement in pp collisions to determine how soft particle production at negative pseudorapidities varies with thex in theprojectile (target) beam-protonheaded towards pos- itive(negative) rapidity canprovide insight intothe relevance of thesevariousscenarios.

ThisLetter presentsameasurementoftheaverage ofthe sum ofthe transverse energy at large pseudorapidity,¹ 

E_T , down- stream of one of the protons in pp collisions, as a function of thehard-scatteringkinematicsindijetevents.Foreachkinematic selection, 

E_T

is the average of the 

E_T distribution in the selected events. The 

ET measurement was deliberately made in only one of the two forward calorimeter modules on either side ofthe interaction point. This was done in analogy withthe centrality definition in p+Pb collisions [5,13], which is charac- terisedbythe

ETintheforwardcalorimetermodulesituatedat

−⁴.9<η<−³.2, inthe nucleus-goingdirection. In pp collisions theasymmetricchoice ofthe

ET-measuring regionmeans that thetarget protonplays therole ofone ofthenucleons inthePb nucleus.

Thevalue of

ET was measured by summing the transverse energyintheforwardcalorimetercells andcorrectingforthede- tectorresponse. Theaverage value, 

ET

,isreportedasa func- tion of the average dijet transverse momentum, p^avg_T = (^pT,1+ pT,2)/2,andpseudorapidity, η^dijet= (η1+η2)/2. Inthesequanti- ties,pT,1and η1arethetransversemomentumandpseudorapidity oftheleading(highest-p_T)jetintheevent,whilep_T,2 and η2 are thosefor thesubleading (secondhighest-pT) jet. Resultsare also reportedasafunction oftwo kinematicquantities xproj andxtarg definedby

x_proj=^pavg_T (e^+η1+^e+η2)/√

s, (1)

x_targ=^pavg_T (e^−η1+^e−η2)/√

s. (2)

In a perturbative approach, at leading order, x_proj (xtarg) cor- responds approximately to the Bjorken-x of the hard-scattered parton in the beam-proton with positive (negative) rapidity. Es- timates of the initial parton–parton kinematics through jet-level variableshavebeenusedpreviouslyindijetmeasurementsatthe CERNSppS collider¯ [14,15]andinmeasurements ofdihadronsin d+Au collisions atRHIC [16]. Finally, to better reveal the rela- tive dependenceof 

ET

 on the hard-scattering kinematics,re- sults are also reported as a ratio to a reference value 

ET

ref , whichisthe

E_T

evaluatedatafixedchoiceofdijetkinematics, 50 GeV<p^avg_T <63 GeV and|η^dijet| <⁰.3.

Fig. 1 schematically illustrates the meaning of the kinematic variables utilised in this measurement. The top panel in Fig. 1 showstheconventionusedinp+Pb collisionsatATLAS,inwhich theprotonbeamisthe“projectile”andhaspositiverapidity,while the nuclear beam is the “target” and has negative rapidity. The centralityofthe p+Pb collision, anexperimental quantitysensi- tivetothecollisiongeometry,ischaracterisedbythe

E_Tinthe forward calorimeter situated in the nucleus-going direction. The middlepanelinFig. 1illustratesthemeasurementinpp collisions

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

Thex-axispointsfromtheIPtothecentreoftheLHCring,andthe y-axispoints upward.Cylindricalcoordinates(r,φ)areusedinthetransverseplane,φbeingthe azimuthalanglearoundthebeampipe.Thepseudorapidityisdefinedintermsof thepolarangleθasη= −^{ln tan}(θ/2).

Fig. 1. Schematicillustrationofthekinematicvariablesinthemeasurement.Panel (a)illustratestheconventioninp+Pb collisions.Panels(b)and(c)illustratehow asinglepp eventprovidesameasurementof

ETattwovaluesofη^dijet= (η1+ η2)/2,inthiscaseforη^dijet= +^{1 andfor}η^dijet= −^1,respectively.

reportedinthisLetter,inwhichtheprotonbeamwithpositivera- pidityisconsideredtobe theanalogueoftheprojectileprotonin p+Pb collisions,whilethetargetprotonwithnegative rapidityis the analogueof a single nucleon within the Pbnucleus, andthe

E_T is measured intheforward calorimeterdownstreamof the target proton.Duetothe symmetricnatureof pp collisions,each eventcan alsobe interpreted byexchanging the rolesofthe tar- get and projectile between the two protons, and measuring the

ET in the opposite forward calorimeter module. To keep the same convention in this case, the z-axis (and thus the pseudo- rapidity) is inverted and the kinematic variables are determined within thisnewcoordinatesystemasshowninthebottompanel of Fig. 1. The full analysis was performed separately using each forwardcalorimeterside,oneatatime,andthefinalresultswere obtained by averaging the 

E_T

measurements fromeach side.

Thisincreased thenumberof 

E_T measurements by a factorof two and alsoprovided an importantcross-check on the detector energyscale.Forsimplicity,all η valuesintheselectioncutsand

η^dijet values inthe results described beloware always presented accordingtothe conventionwhere

ET is measuredatnegative pseudorapidity.

Thedatasetusedinthismeasurementwascollectedduringthe

√s=².76 TeV pp collision data-taking in February 2013 at the LargeHadronCollider,withanintegratedluminositycorresponding to4.0 pb⁻¹.Duringdata-taking,the meannumberof pp interac- tions per bunch crossing varied from 0.1 to 0.5. This dataset is particularlysuitable forthemeasurementbecausethesmallmean interactionratepercrossingallowsrejectionofdijet-producing pp eventswithadditionalpp interactionsinthesamebunchcrossing (pileup)withgoodsystematiccontrolwhilesimultaneouslyhaving enough integratedluminosity to measure dijetproduction overa widekinematicrangewithgoodstatisticalprecision.

2. Experimentalsetup

TheATLASdetectorisdescribedindetailinRef.[17].Thisanal- ysisusesprimarilythetrackingdetectors,thecalorimeter,andthe trigger system. Charged-particle tracks were measured over the range |η| <².5 using the inner detector, which is composed of silicon pixel detectors in the innermost layers, silicon microstrip detectors,andastraw-tube transition-radiationtracker(|η| <².0) intheouterlayer,allimmersedina2 Taxialmagneticfield.The calorimeter system consists of a liquid argon (LAr) electromag- neticcalorimeter(|η| <³.2),asteel/scintillatorsamplinghadronic calorimeter (|η_{| <}1.7), a LAr hadronic calorimeter (1.5<|η_{| <}

3.2), and a forward calorimeter (3.2<|η| <⁴.9). The forward calorimeteriscomposedoftwomodulessituatedatoppositesides oftheinteractionregionandprovidesthe

ETmeasurement.The modulesconsistoftungstenandcopperabsorberswithLArasthe active medium,which togetherprovide teninteraction lengthsof material, and are segmented into one electromagnetic and two hadronicsections longitudinal inthe shower direction. The 1782 cells in each forward calorimeter module are aligned parallel to the beam axis andtherefore are not projective, but have a seg- mentationcorrespondingtoapproximately0.2×⁰.2 in η andφ.

Data were acquiredforthis analysisusinga series ofcentral- jet triggers covering |η_{| <}3.2 with different (increasing) jet-pT thresholds,rangingfrom40 GeVto75 GeV[18].Eachtriggerwas prescaled,meaningthatonlyafractionofeventspassingthetrig- ger criteria were ultimately selected, and these fractions varied withtimetoaccommodatetheevolutionoftheluminositywithin anLHC fill.Thisfractionincreasedfortriggers withincreasing jet p_T threshold andthe highest-threshold trigger, which dominates thekinematic rangestudied in thisLetter, sampledthe full inte- gratedluminosity.

3. MonteCarlosimulation

Monte Carlo (MC) simulations of √

s= ².76 TeV pp hard- scatteringeventswereusedtounderstandtheperformanceofthe ATLASdetector,tocorrectthemeasured 

E_T anddijetkinematic variablesfordetectoreffects,andtodeterminethesystematicun- certainties in the measurement. Three MC programs were used to generate event samples with the leading-jet p_T in the range from20 GeVto1 TeV: the Pythia 6generator [19]withparame- tervalueschosen toreproducedataaccordingto theAUET2Bset oftunedparameters(tune)[20]andCTEQ6L1partondistribution function(PDF)set[21];the Pythia 8generator[22]withtheAU2 tune [23] and CT10 PDF set [24]; and the Herwig++ generator [25]withtheUE-EE-3tune[26]andCTEQ6L1PDFset.Thegen- eratedeventswerepassed throughafull Geant 4simulation[27, 28]oftheATLASdetectorunderthesameconditionspresentdur- ingdata-taking.Thesimulatedeventsincludedcontributionsfrom pileupsimilartothatindata.

At the particle level, jets are defined by applying the anti-kt algorithm [29] with radius parameter R of0.4 to primary parti- cles² within |η_{| <}4.9, excluding muons and neutrinos. 

ET is definedatthe particlelevel asthe sumofthe transverseenergy ofallprimary particleswithin−⁴.9<η<−³.2,includingmuons andneutrinos,andwithnoadditionalkinematicselection.

4. Eventreconstructionandcalibration

The vertex reconstruction, jet reconstruction and calibration, and

ET measurement andcalibrationprocedures are described

2 Primary particlesare defined as final-state particleswith a proper lifetime greaterthan30 ps.

inthissection. Theywereapplied identicallytotheexperimental dataandthesimulatedevents.

4.1. Trackandvertexreconstruction

In the offline analysis, charged-particle tracks were recon- structed in the inner detector with an algorithm used in previ- ous measurementsofcharged-particle multiplicitiesinminimum- bias pp interactions [30]. Analysed eventswere requiredto con- tain a reconstructed vertex, formed by at least two tracks with pT>0.1 GeV[31].The contributionfrompileupinteractions was suppressed by rejecting events containing more than one re- constructed vertexwith five or more associated charged-particle tracks.Thisrequirementrejectedapproximately8%ofevents.

4.2. Jetreconstructionandcalibration

The jet reconstruction and associated background determina- tion procedures closelyfollow those developed within ATLAS for jet measurements in heavy-ion and pp collisions [5,32–34]. This procedureissummarisedinthefollowingandisdescribedinmore detailinRef.[32].Jetswere reconstructedbyapplyingtheanti-kt algorithm with R=⁰.4 to calorimeter cells groupedinto towers of size η_{× φ =}0.1×⁰.1. The procedure providedan η- and samplinglayer-dependentestimateofthesmallenergydensityde- posited by the soft underlying eventfrom pileup interactions in each crossing.Theenergies ofthecellsineachjetwere corrected for this estimate of the soft pileup contribution. The p_T of the resulting jetswas corrected for the calorimeter energy response throughasimulation-derivedcalibration,withanadditionalinsitu correction, typically at the percent level, derived through com- parisons of boson–jet and dijet pT balance in collision data and simulation[35].

4.3. Forwardtransverseenergymeasurementandcalibration The

E_T quantitywasevaluatedbymeasuringthesumofthe transverse energy in the cells in one forward calorimeter mod- ule (

E_T^raw). The energysignals fromthecells were included in thesumwithoutanyenergythresholdrequirement. Thisquantity wascorrectedevent-by-eventtoaccountforthedetectorresponse, usingacalibrationprocedurederivedinsimulation,togiveanes- timate of the full energy depositedin the calorimeter (

E^calib_T ).

Pythia 8 was found togive the bestoverall description of  ET productionandofitsdependenceondijetkinematicsindata.Thus, a subset of Pythia 8 events with good kinematic overlap with the data anda wide rangeof 

ET values was used tocalibrate

E^raw_T . The calibration was derived by requiring that for each subset ofsimulated events witha narrow rangeof particle-level

ET values(

E^gen_T ),themeanvalueofthe

E^calib_T distribution in thoseeventscorrespondedto themeanvalue of 

E^gen_T .First, to determine the averageoffset inthe response (), the average

E^raw_T asa functionof

E^gen_T wasextrapolatedwithalinearfit to zero 

E_T^gen.Thisadditive offset, which described the average neteffectofenergyinflowfromoutsideandenergyoutflowfrom inside thefiducialpseudorapidity acceptanceof−⁴.9<η<−³.2, was found to be approximately ≈ −⁰.7 GeV. It also reflected the residual contribution from pileup interactions and the aver- agedistortionofthesignalfromenergydepositedby collisionsin previous bunch crossings. Second, the average response (C ) was determined by the ratio ofthe mean offset-corrected 

E^raw_T to each corresponding value of 

E^gen_T , C= 

E^raw_T −  / E^gen_T . C wasfoundtobeapproximately0.7 andvariedonlyweaklywith

E^gen_T after the offset correction. This residual dependencewas modelled by evaluating C in narrow bins of 

E^raw_T and fitting