s = 13 TeV using an ionisation measurement with the ATLAS detector

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Search for heavy charged long-lived particles in proton–proton collisions at √

s = 13 TeV using an ionisation measurement with the ATLAS detector

.TheATLAS Collaboration

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

Articlehistory:

Received14August2018

Receivedinrevisedform3October2018 Accepted29October2018

Availableonline1November2018 Editor: M.Doser

ThisLetterpresentsasearchforheavychargedlong-livedparticlesproducedinproton–protoncollisions at√

s=^{13 TeV attheLHCusingadatasample}correspondingtoanintegratedluminosityof36.1 fb⁻¹ collected by the ATLAS experiment in2015 and 2016. Theseparticles are expected to travel with a velocitysignificantlybelowthespeedoflight,andthereforehaveaspecificionisationhigherthanany high-momentum Standard Model particle ofunit charge. The pixel subsystem of the ATLAS detector is used in this search to measure the ionisation energy loss of all reconstructed charged particles whichtraversethepixeldetector.ResultsareinterpretedassumingthepairproductionofR-hadronsas compositecolourlessstatesofalong-livedgluinoandStandardModelpartons.Nosignificantdeviation from Standard Model background expectations is observed, and lifetime-dependent upper limits on R-hadron productioncross-sectionsand gluino massesare set, assumingthe gluino alwaysdecays to twoquarksanda100 GeVstableneutralino.R-hadronswithlifetimesabove1.0 nsareexcludedatthe 95% confidencelevel,withlowerlimitsonthegluinomassrangingbetween1290 GeVand2060 GeV.In thecaseofstableR-hadrons,thelowerlimitonthegluinomassatthe95% confidencelevelis1890 GeV.

©^{2018TheAuthor.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense} (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP³.

1. Introduction

AwiderangeofphysicsmodelsthatextendtheStandardModel (SM) predict the existence of new, massive, long-lived particles (LLPs).Theseparticles appear inproposed solutions to thegauge hierarchy problem [1], including supersymmetric (SUSY) models thateitherviolate [2–4] orconserve [5–12] R-parity. R-parityisa quantumnumberdefinedas(−¹)^3(B−L)+2S whereSistheparticle spinandL andB are,respectively,its leptonandbaryon number.

Within SUSY models, sparticles, including gluinos, may be long- lived,withlifetimesdepending,forinstance,onthemasshierarchy parameters,oronthesizeofany R-parity-violatingcoupling [13].

The study in this Letter is sensitive to many different mod- els of new physics, in particular those that predict the produc- tion of massive particles with lifetimes exceeding 1 ns at LHC energies,such asmini-split SUSY [10,14,15] oranomaly-mediated supersymmetry-breaking(AMSB) models [16,17]. Results arepre- sentedassumingtheproductionofR-hadronsascompositecolour- lessstatesofagluinotogetherwithSMquarksorgluons [18].

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

Due to their large mass, LLPs are expectedto be slow (βγ ≤ 0.9¹ inalargefractionofcases)andtherefore,ifcharged,tohave a specificionisationlarger thananySM particleofunit chargeat high momentum. The pixel subsystem [19] of the ATLAS detec- tor [20] providesmeasurementsofionisationenergyloss(dE/dx) for charged particles withsufficient accuracy to distinguish such highlyionisingparticlesfromSMparticles.InthisLetter,thedE/dx information is used to search for LLPs using a data sample of proton–proton (pp) collisions corresponding to an integrated lu- minosityof 36.1 fb⁻¹ collectedat√

s=^{13 TeV.This extendsthe} reachbeyondthatofapreviousstudy [21],thankstoatenfoldin- crease ofthe integrated luminosity andto several improvements to the analysis. It also extends the reach beyond that of simi- lar studies by CMS [22] andATLAS [23] carriedout atthe same centre-of-mass energy and dedicated to the search for LLPs not decayinginsidethedetector.

1 Hereβisthespeedoftheparticlerelativetothespeedoflightinvacuumand γ=_1−β¹ 2.

https://doi.org/10.1016/j.physletb.2018.10.055

0370-2693/©^{2018TheAuthor.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense}(http://creativecommons.org/licenses/by/4.0/).Fundedby SCOAP³.

2. ATLASdetectorandionisationmeasurement

The ATLAS detector² is a general-purpose detector with a forward–backward symmetric cylindrical symmetry described in detailinRef. [20].Itconsistsofatrackerformeasuringthetrajec- toriesof chargedparticles inside a2 T solenoidal magnetic field, followedbycalorimetersformeasuringtheenergyofparticlesthat interactelectromagnetically orhadronically.A muonspectrometer immersedinatoroidalmagneticfield surroundsthecalorimeters, and provides tracking for muons. A two-level trigger system is usedtoselectevents [24].Thefirst-leveltriggerisimplementedin hardwareandusesasubsetofthedetectorinformation.Thisisfol- lowedbythesoftware-basedhigh-leveltrigger,whichrunsoffline reconstruction and calibration software, reducing the event rate toabout1 kHz.The detectorishermetic andcan thereforemea- surethemagnitudeof themissingtransverse momentum (E^miss_T ) associated with each event. The tracker is made of three detec- torsystemsorganisedinconcentriclayers. Theoutermostlayeris madeofdenselypackedproportionalgas-filleddetectors [25],the radialregion fromroughly 30 cmto 55 cmis equippedwithsil- iconmicrostrip detectors [26] and theinnermostlayer iscovered byasiliconpixeldetector [19],whichisdescribedbelowinsome detailasithasacrucialroleinthisanalysis.

The pixel detector typically provides four precision measure- mentsforeachtrackintheregion|η_|<2.5 atradialdistancesof 33 mm,50 mm,88 mmand122 mmfromtheLHCbeamline.The innermostpixellayerisnamedtheinsertableB-layer(IBL) [27] and was designedto maintain efficient operation of the pixelsystem above2×^{1034 cm−2s−1} luminosity,whenthenext-to-innermost pixellayerbeginstolosedetectionefficiency.Thehitefficiencyof thepixeldetectorinthedatasampleusedforthisanalysisstillex- ceeds99% inalllayers.Foreachpixelhitthelengthoftimewith signalabovethreshold,knownastimeoverthreshold(ToT), isdigi- tisedandrecorded. The ToTis approximatelyproportional to the ionisationchargeandallowsthecalculationofthespecificionisa- tionenergy, dE/dx, ofa track.The ToT measurement isdigitised withfour bits in the IBL andeight bits in all other pixel layers.

If the dynamic range is exceeded for a particular hit in the IBL an overflow bit is set, while for the other layers the hit is not recorded.

Thechargereleasedbyamovingchargedparticleisrarelycon- tainedwithin justone pixel; neighbouring pixels registering hits arejoinedtogether usingaconnectedcomponentanalysis [28,29]

toformclusters.Thechargeofaclusteriscalculatedbysumming the charge of all pixels belonging to the cluster after calibration corrections.Toavoidlossofcharge,only clusterscompletelycon- tainedinsensorfiducial regions areused (e.g.clusterscannot be incontactwithpixelsonthesensoredge).ThedE/dxforeachre- constructedtrackiscalculatedusingtheaverageoftheindividual cluster ionisation measurements (charge collected in the cluster perunittracklengthinthesensor),fortheclustersassociatedwith atrack.ToreducetheimpactofthetailsoftheLandaudistribution, whichis expectedto describe theenergydeposition distribution, thetrackdE/dxisevaluatedusingatruncated-meanmethod.The averageiscalculatedafterremoving thehighest-dE/dxcluster,or thetwo highest-dE/dxclustersintherelativelyrarecaseofmore

2 ATLASusesaright-handedcoordinatesystemwithitsoriginatthenominalin- teractionpoint(IP)inthecentreofthe detectorandthe z-axisalongthebeam pipe.Thex-axispointsfromtheIPtothecentreoftheLHCring,andthe y-axis pointsupward.Cylindricalcoordinates(r,φ)areusedinthetransverseplane,φbe- ingtheazimuthalanglearoundthez-axis.Thepseudorapidityisdefinedinterms ofthepolarangleθasη= −^{ln tan}(θ/2),andangulardistanceismeasuredinunits ofR≡

(η)²+(φ)².

than four clusters associated withthe track.More details of the calculationofdE/dxmaybefoundinRef. [21].

3. Analysisoverview

Thesearchstrategyconsistsoflookingforexcessesinthemass distributionofreconstructed trackswithhightransversemomen- tum, p_T,and large dE/dx. The mass value is determined froma parameterisation oftheBethe–Blochrelationanddependson the momentumanddE/dxofselectedtracks.

Twosignalregionsareconsidered,andtheselectionisdetailed in Section 6. The first region targets metastable R-hadrons with lifetimes such that the majority of their decays occur inside the detector. In this region, charged particles that reach the muon spectrometer are removed and the selections are optimised for R-hadronswithlifetimesfromaround 1 ns toseveraltens of ns.

A secondsignal regiontargetsstable R-hadronswhichdonotde- cay within the detector. Inthis region, nomuon veto is applied, since some of the stable R-hadrons that pass throughthe muon spectrometerarereconstructedasmuons.

Events are selected using the lowest-threshold unprescaled calorimetricE^miss_T trigger.InmetastableR-hadronevents,themea- sured E^miss_T largely originatesfromneutralinoswhich carry away unmeasured momenta.In stable R-hadron events,the R-hadrons leaveonlymodestenergydepositionsinthecalorimeters [30] and onlyafractionarereconstructedasmuonsduetotheirlatearrival timeinthemuonspectrometer.Therefore,mostofthemomentaof R-hadronsarenotaccountedforinthemeasurementofE^miss_T ,and only QCD initial-state radiation (ISR) provides a visible contribu- tionthatresultsinameasuredimbalance.Duetotheneutralinos, the E^miss_T triggerefficiencyishigherformetastablethanforstable R-hadrons. Thetrackreconstructionefficiencyis,onthe contrary, higherforthe stable R-hadronsandpenalises particleswithlife- times shorter than 10 ns, which may not have crossed enough detectorlayers.Thesearchesforstableandmetastable R-hadrons requireslightlydifferentoptimisations.

The background is estimated with a data-driven approach, as described in Section 7. Data control samples are usedto param- eterise the momentum and dE/dx distributions and their inter- dependence, and then to generate pseudo data which predicts thebackgrounddistribution.The potentialsignalcontaminationis minimisedinthesebackgroundsamplesbyinvertingsome ofthe selectioncriteria.

4. Dataandsimulation

This search usesdata from pp collisions at √

s=13 TeV pro- videdbytheLHC in2015and2016.The integratedluminosity of thedatasampleis36.1 fb⁻¹,afterrequirementsondetectorstatus anddata quality have beenapplied. Further detector-level clean- ing selectionsare appliedto thedatatoreject eventsaffectedby calorimeternoiseanddatacorruption.

An additional data sample, collected in a dedicated low- luminosity run in 2016, isused forthe calibration ofdE/dx and mass;it consistsofrandomlytriggeredevents inbunchcrossings wherecollisionsareexpectedandamountstoabout0.4 nb⁻¹.

Simulation samples are used to determine the efficiency and associated uncertainty for selecting signal events. To model sig- nal events, the pair production of gluinos with masses between 400 GeV and 3000 GeV was simulated in Pythia 6.4.27 [31] at leading order with the AUET2B [32] set oftuned parameters for the underlying event and the CTEQ6L1 [33] parton distribution function(PDF)set.Dedicatedroutines [34] wereusedtohadronise the gluinos; after hadronisation, about 2/3 of the eventscontain atleastonechargedR-hadron.Allsparticlesexceptthegluinoand

thelightestneutralinoaredecoupled.TheMonteCarlo(MC)signal samplesincludeamodellingofpile-up,addingtheexpectednum- ber of minimum-bias pp interactions fromthe same andnearby bunchcrossings.

InordertomoreaccuratelymodelISRinthesignalevents,ad- ditionalsamplesofgluinos were generatedat leadingorder with up to two additional partons using MadGraph5_aMC@NLO [35], interfaced to the Pythia 8.186 [36] parton shower model. The NNPDF2.3LO [37] PDF set is used along with the A14 [38] set of tuned parameters. The distribution ofthe transverse momen- tumofthegluino–gluinosystemsimulatedwithPythia6.4.27was reweightedtomatchthatobtainedinthesamplessimulatedwith MadGraph5_aMC@NLO.

Simulated events undergo full detector simulation [39] based on a Geant4 [40] framework; the hadronic interactions of R- hadronswiththedetectorwerehandledbydedicatedGeant4rou- tinesbasedonthemodeldescribedinRefs. [30,34,41].Signalsam- plesweregeneratedbothfornon-decayinggluinos,andforgluinos withaset oflifetimesranging from1.0 nsto50 nswhich decay into SM quarks anda 100 GeV stable neutralino via the process

˜

g→^qq¯χ˜₁⁰.ThedecayoftheR-hadronsandthefragmentationand hadronisationoftheresultingquarkswereperformedwithamod- ifiedversionofPythia6.4.27.

To normalise the number of expected signal events, gluino pair production cross-sections are calculated at next-to-leading order in the strong coupling constant, including the resumma- tion of soft-gluino emissions at next-to-leading-logarithm accu- racy [42–46]. The nominal cross-section valuesand uncertainties aretakenfromanenvelopeofcross-sectionpredictionsusingdif- ferent PDF sets and factorisation and renormalisation scales, as describedinRef. [47].

5. dE/dxcorrectionsandmasscalculation

ATLAS hasused the measured dE/dx tosearch for R-hadrons inseveralpreviousanalyses [21,48,49].Thismethodhasbeencon- stantly improved to take into account the evolution of the pixel detectorandtheexperimental conditions.Detailedimprovements relatedtothemeasurementofdE/dxandmassintroducedinthis analysis,include:

• Correctionshavebeenmadeforluminosity- andtime-depen- dent variations of the measured valuesof dE/dx. The varia- tions are due to changes in theoperation parameters ofthe pixelsystemandto lossofcharge collectiondueto radiation damagecaused bythe luminositydelivered. The dE/dx mea- suredindataisscaledbyaper-runfactorderivedtokeepthe most probablevalue of the energy loss (MPVdE/dx) constant versustime.TheMPV_dE/dx variationwithintegratedluminos- itybeforecorrectionsisshowninFig.1.

• ^Alow-momentumcorrection forkaonsandprotonshasbeen added.Allparticlesaretreatedaspionsinthereconstruction program,but, below 500 MeV,the effectofmultiple scatter- ingon thetrajectoriesofkaonsandprotons isdifferentfrom theeffect on a pionandtheir momentaare underestimated.

Tocorrectforthiseffect,thedifferencebetweenthegenerated andthereconstructedmomentumofprotonandkaontracksin simulationsamplesisfittedasafunctionofmomentum.This parameterisedcorrectionisthenappliedtothemomentumof protons andkaons indata, where these particles are identi- fiedbymeansoftheirdE/dxandmomentum.Thisprocedure hassimplifiedthedE/dxcalibration,whichisperformedwith low-βγ SMparticles.

• ^{There is a small dependence of the dE}/dx on the traversed thickness [50]. For that reason the dE/dx calculated in this

Fig. 1.ThemostprobablevalueofthetrackdE/dx(MPVdE/dx)versustheintegrated luminositydeliveredtoATLASisreportedforeachdata-takingrunusedinthisanal- ysisbeforecorrectionsareapplied.Theluminosityplottedhereisbeforedetector efficiencyanddataqualitycriteriaareimposed.TheMPVdE/dxiscalculatedforall trackswithpT>400 MeV.Thepointstotheleftofthedashedlinerepresentthe datarecordedduring2015,duringwhichavariationoftheMPVdE/dxduetotheToT driftoftheIBLelectronicsispronounced.Indatarecordedduring2016,adropof MPVdE/dxoverintegratedluminosityisobservedduetochargecollectionefficiency losses.Smalllocalfluctuationsarealsovisible.Thesearecausedbythechangeof theexperimentalenvironmentandofthedetectorconditions.Inthisanalysis,the measurementofdE/dxiscorrectedtoaccountforthevariationasafunctionof data-takingrun.

analysis takes into account its small (<10%) η-dependence.

Afterthiscorrection, thedE/dxdependsonlyon theparticle momentum and mass,which simplifies thebackground esti- mation(seeSection7).

•^{As the simulation does not include the effects of radiation} damage to the pixeldetector sensors,a scale factor of0.886 isappliedtothemeasurementofdE/dxinsimulationtoalign the MPV_dE/dx of the minimum-ionising particles in MC sim- ulation withdata,aftertherun-dependent correctionsto the datadE/dxhavebeenapplied.

The βγ of a particle, and therefore its mass if the momen- tumisknown,canbecalculatedfromthedE/dxofitstrackusing therelationship betweenβγ anddE/dx.Aβγ valuecanonly be measured intherange0.3< βγ <0.9.Onaverage,particleswith βγ <0.3 have a dE/dx such that the ToT dynamic range is ex- ceeded. Particleswithβγ >0.9 have a dE/dxwhich istooclose totheionisationplateauofrelativisticSMparticlesforanefficient discrimination. This range overlaps well withthe expected aver- ageβγ ofR-hadronsproducedattheLHC,whichdecreasesfrom around 0.8for a gluino with mass 600 GeV to around 0.4 for a 2000 GeVgluino.

The mass of a charged particle can be derived from a fit of the specific energyloss and the momentum measurement to an empiricalfunctionmotivatedbythelow-β behaviouroftheBethe–

Bloch distribution. After applying the low-momentum correction for kaons and protons, it is possible to fit the function relating dE/dxtoβγ withonlythreeparameters(insteadoffiveasinthe previous analysis [21]), asshown inFig. 2. The parametric func- tion describingtherelationship betweenthemostprobablevalue oftheenergyloss(MPV_dE/dx)andβγ is:

MPV_dE/dx=^A/(βγ)^C+^{B (1)}

The A, B and C calibration constants were measured using low- momentum pions, kaons and protons reconstructed by ATLAS in low-luminosity runs where all reconstructed tracks with p_T >