Multiplicity and rapidity dependence of strange hadron production in pp, pPb, and PbPb collisions at the LHC

Physics Letters B 768 (2017) 103–129

Contents lists available atScienceDirect

Physics

Letters

B

www.elsevier.com/locate/physletb

Multiplicity

and

rapidity

dependence

of

strange

hadron

production

in pp,

pPb,

and

PbPb

collisions

at

the

LHC

.

The

CMS

Collaboration



CERN,Switzerland

a

r

t

i

c

l

e

i

n

f

o

a

b

s

t

r

a

c

t

Articlehistory: Received21May2016

Receivedinrevisedform10December2016 Accepted16January2017

Availableonline20February2017 Editor:M.Doser Keywords: CMS Physics Heavyion Spectra Radialflow

Measurements ofstrange hadron (K0

S,+,and−+

+_{)transverse momentumspectrain pp,pPb,} andPbPb collisionsarepresentedoverawiderangeofrapidityandeventcharged-particlemultiplicity. The data werecollectedwiththe CMS detectoratthe CERN LHCinpp collisions at√s=7TeV,pPb collisions at√sN N=5.02TeV,and PbPb collisions at√sN N=2.76TeV.Theaveragetransverse kinetic

energyisfoundtoincreasewithmultiplicity,atafasterrate forheavierstrange particlespeciesinall systems.Atsimilar multiplicities,thedifferenceinaveragetransverse kineticenergybetweendifferent particlespeciesisobservedtobelargerforpp andpPb eventsthanforPbPb events.InpPb collisions, theaveragetransversekineticenergyisfoundtobeslightlylargerinthePb-goingdirectionthaninthe p-goingdirectionforeventswithlarge multiplicity.Thespectraarecomparedtomodels motivatedby hydrodynamics.

©2017TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense (http://creativecommons.org/licenses/by/4.0/).FundedbySCOAP3.

1. Introduction

Studiesofstrange-particle productioninhighenergycollisions ofprotonsandheavy ionsprovideimportantmeanstoinvestigate thedynamics ofthe collision process. Earlier studies of relativis-ticheavy ion collisions at the BNL RHIC and CERN SPS colliders indicatedanenhancementofstrangenessproductionwithrespect toproton–proton(pp)collisions[1,2],whichwashistorically inter-pretedto bedueto the formationofa high-densityquark–gluon medium[3].Theabundanceofstrangeparticlesatdifferent center-of-massenergiesisinlinewithcalculationsfromthermal statisti-cal models [4–6]. In gold–gold (AuAu) collisions at RHIC, strong azimuthal correlationsof final-statehadronswere observed, sug-gestingthattheproducedmediumbehaveslikeanear-perfectfluid undergoingapressure-drivenanisotropicexpansion[2].Studiesof strangenessandlightflavorproductionanddynamicsinheavyion collisions have provided further insight into the medium’s fluid-likenatureandevidenceforitspartoniccollectivity[2,7].

In recent years, the observation of a long-range “ridge” at smallazimuthalseparations intwo-particlecorrelationsin pp [8]

andproton-lead(pPb) [9–11] collisions withhighevent-by-event charged-particle multiplicity (referred to hereafter as “multiplici-ty”)hasprovidedanindicationforcollectiveeffectsinsystemsthat areanorderofmagnitudesmallerinsizethanheavyioncollisions.

 _{E-mailaddress:cms-publication-committee-chair@cern.ch.}

Thenatureoftheobservedlong-rangeparticlecorrelationsinhigh multiplicitypp andpPb collisionsisstillunderintensedebate[12]. Whilethe collectiveflowofa fluid-likemedium providesa natu-ral interpretation[13–16],other modelsattributethisbehaviorto theinitialcorrelationofgluons[17–21],ortheanisotropic escape ofparticles[22].

Studies of identified particle production and correlations in highmultiplicity pp and pPb collisionsprovide detailed informa-tion abouttheunderlying particleproductionmechanism. Identi-fiedparticle(includingstrange-hadron)transversemomentum(pT)

spectra andazimuthal anisotropies in lead–lead (PbPb) collisions attheCERNLHC havebeenstudied[23,24]anddescribed by hy-drodynamicmodels[25,26].Similarmeasurementshavebeen per-formed in pPb collisions as a function of multiplicity, where an indication of acommon velocity boostto the produced particles, knownas“radial flow”[27,28],andforamassdependenceofthe anisotropicflow[29,30]havebeenobserved.WhencomparingpPb andPbPb systemsatsimilarmultiplicities,a strongerradial veloc-ityboostisseen inthesmallerpPb collision system[27,30].This couldberelatedtoamuchhigherinitialenergydensityinahigh multiplicitybutsmallersystem,resultinginalargerpressure gra-dientoutwardalongtheradialdirection, aspredictedinRef.[31]. Toperform a quantitative comparison, a common average radial-flowvelocityfromdifferentcollisionsystemscanbeextractedfrom asimultaneousfittothespectraofvariousparticlespecies, based on the blast-wave model [32]. Inspired by hydrodynamics, the blast-wavemodelassumesa commonkinetic freeze-out

tempera-http://dx.doi.org/10.1016/j.physletb.2017.01.075

0370-2693/©2017TheAuthor(s).PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/4.0/).Fundedby SCOAP3_.

pp events represent an even smaller system than pPb events, a strongerradial-flowboostmightbepresentcomparedtopPb and PbPb events ata comparable multiplicity [31]. Furthermore, ina pPb collision,thesystemisnotsymmetricinpseudorapidity(

η

).If afluid-likemediumisformed,itsenergydensitycouldbedifferent onthep- and Pb-goingsides, whichcouldlead toan asymmetry inthecollectiveradial-floweffectasafunctionof

η

. Hydrodynam-icalmodelspredictthattheaveragepT(or, equivalently,the

aver-agetransversekineticenergy



KET



,where



KET



≡ 

mT



−

m,with mT

=



m2

₊

_p2

T and m the particle mass) of produced particles

is larger in the Pb-goingdirection than in the p-going direction, whilethistrendcould bereversedinmodelsbasedongluon sat-uration [37]. Measurement of identified particle pT spectra as a

functionof

η

couldthushelptoconstraintheoreticalmodels. ThisLetterpresentsmeasurementsofstrange-particle pT

spec-trainpp,pPb,andPbPb collisionsasafunctionofthemultiplicity inthe events.Specifically, we examine thespectra ofK0

S,



, and



− particles, where the inclusion of the charge-conjugate states isimplied for



and



− particles. The data were collected with theCMSdetectorattheLHC.Withtheimplementationofa ded-icated high-multiplicitytrigger, thepp and pPb datasamples ex-hibitmultiplicitiescomparabletothatobservedinperipheralPbPb collisions,where “peripheral”refers to

∼

50–100% centrality, with centralitydefinedasthefractionofthetotalinelasticcrosssection. Themostcentralcollisionshave0%centrality.Thisoverlapinmean multiplicityallowsthethreesystems,withdrasticallydifferent col-lisiongeometries, tobe compared.The largesolid-angle coverage oftheCMSdetectorpermitsthestrange-particle pT spectratobe

studiedindifferentrapidityranges,andthusthestudyofpossible asymmetrieswithrespecttothep- andPb-goingdirectionsinpPb collisions.

2. Detectoranddatasamples

Thecentralfeature oftheCMSapparatusisasuperconducting solenoid of 6 m internal diameter, which provides an axial field of3.8T.Withinthesolenoidvolume area siliconpixelandstrip tracker(with13and14layers inthecentralandendcapregions, respectively), alead tungstatecrystalelectromagnetic calorimeter (ECAL), and a brass and scintillator hadron calorimeter (HCAL), each composed ofa barrel andtwo endcapsections. The tracker covers the pseudorapidity range

|

η

|

<

2

.

5. Reconstructed tracks with 1

<

pT

<

10GeV typically have resolutions of 1.5–3% in pT

and 25–90 (45–150) μm in the transverse (longitudinal) impact parameter [38]. The ECAL and HCAL each cover

|

η

|

<

3

.

0 while forwardhadron calorimeters(HF) cover 3

<

|

η

|

<

5.Muons with

|

η

|

<

2

.

4 aremeasuredwithgas-ionizationdetectorsembeddedin the steel flux-return yoke outside the solenoid. A more detailed descriptionoftheCMSdetector,together withadefinitionofthe coordinate system and the relevant kinematic variables, can be foundinRef. [39].The MonteCarlo(MC) simulationofthe

parti-ingthedatafromthetworunperiods.Becauseoftheasymmetric beam conditions,the nucleon–nucleoncenter-of-mass inthe pPb collisions moves with speed

β

=

0

.

434 in the laboratory frame, corresponding to a rapidity of 0.465. As a consequence, the ra-pidity ofa particle in thenucleon–nucleon center-of-mass frame ( ycm)isdetectedinthelaboratoryframe( ylab)withashift, ylab

=

ycm

+

0

.

465.ThepPb particleyieldsreportedinthisLetterare

pre-sentedintermsof ycm,ratherthan ylab,forbettercorrespondence

withtheresultsfromthepp andPbPb collisions.

3. Selectionofeventsandtracks

The triggers, eventreconstruction,andeventselection are the sameasthosediscussedforpp,pPb,andPbPb collisionsinRefs.[8, 41]. Theyare briefly outlined inthe following paragraphs for pp andpPb collisions,whicharethemainfocusofthisLetter.A sub-setofperipheralPbPb datacollectedin2011withaminimum-bias triggerisreprocessedusingthesameeventselectionandtrack re-construction algorithmasforthepresentpPb andpp analyses,in order to more directly compare the three systems at the same multiplicity. Details of the 2011 PbPb analysis can be found in Refs.[41,42].

Minimum-bias pPb events are triggered by requiring at least one trackwith pT

>

0

.

4GeV tobe found inthepixeltracker.

Be-cause of hardware limitations in the data acquisition rate, only a small fraction

(

∼

10−3

)

of triggered minimum-bias events are recorded.Inordertocollectalargesampleofhigh-multiplicitypPb collisions,adedicatedhigh-multiplicitytriggerisimplemented us-ingtheCMSLevel-1(L1)andhigh-leveltrigger(HLT)systems[43]. AtL1,thetotaltransverseenergysummedovertheECALandHCAL isrequiredtoexceedeither20or40GeV,dependingonthe mul-tiplicityrequirementasspecified below.Charged particlesare re-constructedattheHLTlevelusingthepixeldetectors.Itisrequired that these tracks originate within a cylindrical region (30 cm in length along thedirectionofthe beamaxisand0

.

2 cm in radius inthedirectionperpendicularto thataxis)centered onthe nom-inalinteraction point. Foreach event,the numberofpixel tracks (Nonline

trk ) with

|

η

|

<

2

.

4 and pT

>

0

.

4GeV isdetermined for each

reconstructed vertex. Only tracks with a distance of closest ap-proach0

.

4 cm orlesstooneoftheverticesareincluded.TheHLT selectionrequiresNonline_trk forthevertexwiththelargestnumberof tracks to exceeda specific value. Data are collected in pPb colli-sionswiththresholds Nonline

trk

>

100 and130foreventswithanL1

transverseenergythresholdof20GeV,and Nonline_trk

>

160 and190 forevents withan L1thresholdof 40GeV. Whileall eventswith Nonline

trk

>

190 areaccepted,onlyafractionoftheeventsfromthe

otherthresholdsareretained.Thisfractionisdependentonthe in-stantaneous luminosity.Datafromboththeminimum-biastrigger and the high-multiplicitytrigger are retained foroffline analysis. Similar high-multiplicity triggers, with differentthresholds, were developedforpp collisions,withdetailsgiveninRef.[8].

The CMS Collaboration / Physics Letters B 768 (2017) 103–129 105

In the subsequent analysis of all collision systems, hadronic eventsare selectedby requiring thepresence ofatleast one en-ergydepositlargerthan3GeV ineachofthetwoHFcalorimeters. Eventsarealsorequiredtocontainaprimaryvertexwithin 15 cm ofthenominalinteractionpointalong thebeamaxisand0

.

15 cm in the transverse direction, where the primary vertex is the re-constructedvertexwiththelargesttrackmultiplicity.Atleasttwo reconstructed tracks are required to be associated with this pri-maryvertex,acondition thatisimportantonlyforminimum-bias events.Beam-relatedbackgroundissuppressedbyrejectingevents inwhichlessthan25%ofallreconstructedtrackssatisfythe high-purityselectiondefinedinRef.[38].InthepPb datasample,there is a 3% probability to haveat least one additional interaction in the same bunch crossing (pileup). The procedure used to reject pileupeventsinpPb collisionsisdescribedinRef.[41].Itisbased onthenumberoftracksassociatedwitheachreconstructedvertex andthe distancebetweendifferentvertices.Apurityof99.8%for single pPb collision events is achieved for the highest multiplic-itypPb range studiedin thisLetter. Forthepp data, theaverage numberofcollisionsper bunchcrossingis1.2.However,pp inter-actions thatare well separated fromeach other do not interfere. Thus,amongeventsidentifiedascontainingpileup,theeventis re-tainediftheseparationbetweentheprimaryvertexandanyother vertexexceeds 1 cm.Insuch events,onlytracksfromthehighest multiplicityvertexareused.

Withtheabovecriteria,97%(98%)ofthesimulatedpPb events generatedwith the epos lhc [44] (hijing 2.1 [45]) programs are selected.Similarly,94% (96%)ofthepp eventssimulatedwiththe pythia6Tune Z2[46] (pythia 8Tune4C [47]) programs are se-lected.

Theevent-by-event charged-particle multiplicity N_trkoffline is de-finedusingprimary tracks,i.e., tracksthat satisfy thehigh-purity criteria of Ref. [38] and, in addition, the following criteria de-signed to improve track quality and ensure the tracks emanate fromtheprimaryvertex.Theimpactparametersignificanceofthe trackwithrespecttotheprimaryvertexinthedirectionalongthe beamaxis,dz

/

σ

(

dz

)

,is requiredto be lessthan 3,asis the

cor-responding impact parameter in the transverse plane, dT

/

σ

(

dT

)

.

Therelative pT uncertainty,

σ

(

pT

)/

pT,must belessthan 10%.To

ensure high tracking efficiency and to reduce the rate of mis-reconstructed tracks, the tracks are required to satisfy

|

η

|

<

2

.

4 and pT

>

0

.

4GeV. Based on simulated samples generated with

the hijing program, the efficiency for primary track reconstruc-tion is found to be greater than 80% for charged particles with pT

>

0

.

6GeV and

|

η

|

<

2

.

4.For the multiplicity rangestudied in

thisLetter, nodependenceofthe trackingefficiencyon multiplic-ityisfoundandtherateofmisreconstructedtracksis1–2%.

Thepp, pPb, andPbPb dataare divided intoclasses basedon Noffline_trk . The quantity Ncorrected_trk is the corresponding multiplicity corrected for detector and algorithm inefficiencies in the same kinematicregion (

|

η

|

<

2

.

4 andpT

>

0

.

4GeV).The fractionofthe

totalmultiplicity found ineach interval and the averagenumber oftracksboth beforeandafteraccountingforthecorrectionsare listedin

Table 1

forthe pp dataandinRef. [41]forthepPb and PbPb data.Theuncertaintyintheaveragevalue



Ncorrected_trk



is eval-uatedfromtheuncertaintyinthetrackingefficiency,whichis3.9% fora single track[48]. Forthe pp data,six multiplicity intervals, indicatedin

Table 1

,aredefined,whichareinclusiveforthelower boundsandexclusivefortheupperbounds,asindicatedin

Table 1

. TheaverageN_trkofflinevalueofminimum-biaseventsissimilartothat forthemultiplicity rangeNoffline_trk

<

35.ForthepPb andPbPb data, eightintervalsaredefined.Theseeightintervalsareindicated,e.g., inthe legend ofFig. 2.Note that, unlike pp and PbPb collisions, Noffline_trk forpPb collisions isnotdetermined inthe center-of-mass frame. However, the difference in the Noffline

trk definition between

Table 1

Fractionof the fullevent sample ineach multiplicity interval and the average multiplicity perintervalfor pp data.Themultiplicities Noffline

trk and Ncorrectedtrk are determinedfor|η|

<

2.4 and pT>0.4GeV beforeandafterefficiencycorrections, respectively.Thethirdandfourthcolumnslistthe averagevaluesofNoffline

trk and Ncorrectedtrk .

Multiplicity interval (Noffline

trk ) Fraction  Noffline trk   Ncorrected trk  [0,35) 0.93 12 14±1 [35,60) 0.06 43 50±2 [60,90) 6×10−3 _{68 79±3} [90,110) 2×10−4 _{97 112±4} [110,130) 1×10−5 _{116 135±5} [130,∞) 7×10−7 _{137 158±6}

thelaboratoryandthecenter-of-massframesisfoundtobe min-imal andso thisdifferenceisignored. Thedetectorcondition has beencheckedtobestableforeventswithdifferentmultiplicities.

4. TheK0

S,



,and



−reconstructionandyields

ThereconstructionandselectionproceduresforK0_S,



,and



− candidatesarepresentedinRefs.[30,49].Toincreasetheefficiency for trackswith low momenta andlarge impact parameters, both characteristicofthestrange-particledecayproducts,theloose se-lection of tracks, asdefined in Ref. [38], is used. The K0

S and



candidates(generically referred toas“V0_{s”) arereconstructed,by}

combiningoppositelychargedparticlestodefine asecondary ver-tex.Eachofthetwotracksmusthavehitsinatleastfourlayersof thesilicontracker,andtransverseandlongitudinalimpact parame-tersignificanceswithrespecttotheprimaryvertexgreaterthan1. Thedistanceofclosestapproachofthepairoftrackstoeachother is required to be less than 0

.

5 cm. The fitted three-dimensional vertex of the pair of tracks is required to have a

χ

2 _value

di-vided by thenumber of degreesoffreedom lessthan 7.Each of thetwo tracksis assumedtobe apioninthe caseoftheK0

S

re-construction. As the proton carries nearly all of the momentum in the



decay, the higher-momentum trackis assumedto be a protonandtheothertrackapioninthecaseofthe



reconstruc-tion.Toreconstruct



−particles,a



candidateiscombinedwith anadditionalchargedparticlecarryingthecorrectsign,todefinea commonsecondaryvertex.Thisadditionaltrackisrequiredtohave hitsinatleastfourlayersofthesilicontracker,andboththe trans-verseandlongitudinalimpactparametersignificanceswithrespect totheprimaryvertexarerequiredtoexceed 3.

Due to the long lifetime of the K0

S and



particles, the

sig-nificance of the V0 _{decay length, whichis the three-dimensional}

distancebetweentheprimaryandV0verticesdividedbyits uncer-tainty,isrequiredtoexceed5.ToremoveK0

S candidates

misiden-tified as



particles and vice versa, the



(K0_S) candidate mass assuming both tracksto be pions(the lower-momentumtrackto beapionandthehigher-momentumtrackaproton)mustdifferby morethan 20

(

10

)

MeV fromthe nominal[50]K0_S(



)massvalue. To remove photon conversions to an electron–positron pair, the massofaK0_Sor



candidateassumingbothtrackstohavethe elec-tron massmust exceed15MeV. The angle

θ

point betweenthe V0 momentumvector andthevectorconnectingtheprimary andV0 vertices is requiredto satisfy cos

θ

point

>

0

.

999.This reduces the contributions ofparticlesfromnuclearinteractions, random com-binationsoftracks,andsecondary



particlesoriginatingfromthe weakdecaysof



and

particles.

To optimize the reconstruction of



− particles, requirements onthethree-dimensionalimpactparametersignificanceofits de-cay productswithrespectto theprimary vertexare applied.This significancemustbe largerthan3 (4)fortheproton(pion)tracks

Fig. 1. InvariantmassdistributionofK0

S(left),



(middle),and



−(right)candidatesinthepTrange1–3GeV for220≤Nofflinetrk <260 inpPb collisions.Theinclusionof thecharge-conjugatestatesisimpliedfor



and



−particles.Thesolidlinesshowtheresultsoffitsdescribedinthetext.Thedashedlinesindicatethefittedbackground component.

from the



decay, and larger than 5 for the direct pion candi-datefrom the



− decay. Tofurther reduce thebackground from randomcombinationsoftracks,thecorresponding impact param-etersignificance of



− candidates cannot exceed 2.5. The three-dimensionaldecaylengthsignificance,withrespecttotheprimary vertex,ofthe



− candidateandtheassociated



candidatemust exceed3and12,respectively.

The K0

S,



, and



− reconstruction efficiencies are about 15,

5, and 0.7% for pT

≈

1GeV, and 20, 10, and2% for pT

>

3GeV,

averaged over

|

η

|

<

2

.

4. These efficiencies account for the ef-fectsof acceptance,and forthe branching fractions of the decay modes in which the strange particles are reconstructed. The in-variantmassdistributions ofreconstructedK0_S,



, and



− candi-dateswith1

<

pT

<

3GeV areshownin

Fig. 1

forpPb eventswith

220

≤

Noffline_trk

<

260.Prominentmasspeaksare visible,withlittle background.Thesolidlines showtheresultofamaximum likeli-hoodfit.Inthisfit,thestrange-particlepeaksaremodeled asthe sumoftwoGaussian functionswithacommonmean.The “aver-age

σ

”valuesin

Fig. 1

arethesquarerootoftheweightedaverage ofthevariancesofthetwoGaussian functions.Thebackgroundis modeledwithaquadraticfunctionfortheK0_Sresults,withthe ana-lyticformAq1/2

+

Bq3/2withq

=

m

− (

mπ

+

mp

)

forthe



results,

andwiththeformCqD_withq

₌

_m

₋₍

_m



+

mπ

)

forthe



−results,

whereA,B,C ,andD arefittedparameters.Thesefitfunctionsare foundtoprovideagooddescriptionofthesignalandbackground with relatively few free parameters. The fits are performedover therangesof strange-particleinvariant massesindicatedin

Fig. 1

toobtaintherawstrange-particleyields Nraw K0

S//−

.

Therawstrange-particleyieldsarecorrectedtoaccountforthe branching fraction ofthe reconstructed decay mode, and forthe acceptanceandreconstructionefficiencyofthestrangeparticle, us-ing simulatedeventsamplesbased on the pythia 6(pp) or epos (pPb and PbPb)eventgeneratorand Geant4 modeling ofthe de-tector: Ncorr K0 S//−

=

Nraw K0 S//− Rcorr

,

(1)

where Rcorr is a correction factor from simulation given by the

ratiooftherawreconstructedyieldtothetotalgeneratedyieldfor therespectivestrangeparticle,withNcorr

K0 S//−

thecorrectedyield. Theraw



particleyieldincludescontributionsfromthedecays of



− and

particles. This “nonprompt” contribution is largely determined by the relative



− to



yield (because the contri-butionfrom

particlesis negligible).Thestringentrequirements placed on cos

θ

point remove a large fraction ofthe nonprompt



componentbut, fromsimulation, up to 10% of the



candidates

at high pT are nonprompt.If the relative



− to



yieldin

sim-ulation is modeled precisely, the contamination from nonprompt



particleswillberemovedbythecorrectionprocedureofEq.(1). Otherwise,anadditionalcorrectiontoaccountfortheresidual con-tamination is necessary. As the



− particle yields are explicitly measured in this analysis, this residual correction factor can be determineddirectlyfromthedataas:

f_,residual_np

=

1

+

f_,raw_np,MC



Ncorr_−

/

Ncorr NMC_−

/

NMC_

−

1



,

(2)

where f_,raw_np,MC denotes the fractionof nonprompt



particles in the raw reconstructed



sample asdetermined fromsimulation, while Ncorr_−

/

Ncorr and N

MC −

/

N

MC

 are the



−-to-



yield ratios

from thedata afterapplying thecorrectionsof Eq.(1),andfrom generator-level simulation, respectively. The final prompt



par-ticle yield is given by Ncorr_

/

f_,residual_np . Based on epos MC studies, which has a similar



−/



ratio to the data, the residual non-prompt contributions to the



yields are found to be negligible in pPb and PbPb collisions, while in pp collisions the correction is 1–3% depending on the pT value of the



particle. Note that Ncorr

 inEq.(2)isderivedusingEq.(1),whichinprinciplecontains

theresidualnonprompt



contributions.Nonetheless,byapplying Eq.(2) inan iterativefashion, we expect N_corr to approacha re-sultcorresponding toprompt



particles only.Asecond iteration ofcorrectionisfoundtohaveaneffectoflessthan0.1%onthe



particle yield.As a cross-checkwe treat the sample ofsimulated eventsgeneratedwiththeHIJINGprogramlikedataandverifythat we obtain thecorrectyields atthe generatorlevel afterapplying thecorrectionproceduredescribedabove.

5. Systematicuncertainties

Table 2 summarizes thedifferentsources ofsystematic uncer-tainty intheyieldsofeach strangeparticlespecies. Thevaluesin parenthesescorrespondtothesystematicuncertaintiesinthe for-ward rapidity regions (

−

2

.

4

<

ycm

<

−

1

.

5 and 0

.

8

<

ycm

<

1

.

5)

for pPb data, ifthey differfromthose atmid-rapidity. The dom-inant sources of systematic uncertainty are associated with the strange-particlereconstruction,especiallytheefficiency determina-tion.

The systematicuncertainty in determining the efficiency of a single track is 3.9% [48]. The tracking efficiency is strongly cor-related with the lifetime of a particle because when and where a particle decaysdetermine how efficientlythe detectorcaptures its decayproducts. Weobserve agreementofthe K0_S lifetime dis-tribution (c

τ

) betweendataandsimulation,andsimilarly forthe

The CMS Collaboration / Physics Letters B 768 (2017) 103–129 107

Table 2

SummaryofsystematicuncertaintiesforthepTspectraofKS0,



,and



−particlesinthecenter-of-massrapidityrange |ycm|<1.0 (forpPb events,atforwardrapidities,ifdifferent)forthethreecollisionsystems.

Source K0 S(%) (%) −(%) pT(GeV) <1.5 >1.5 <1.5 >1.5 Single-track efficiency 7.8 7.8 7.8 7.8 11.7 Yield extraction 2 (3) 2 (3) 2 (4) 2 (4) 3 Selection criteria 3.6 (3.6) 2.2 (3.6) 3.6 (6.4) 2.2 (6.4) 7 Momentum resolution 2 2 2 2 2 Nonpromptcorrection 2 2 Pileup (pp only) 3 1 3 1 3

Proton direction (pPb only) 3 (3) 3 (3) 3 (5) 3 (5) 4

Rapidity binning 1 (2) 1 (2) 1 (3) 1 (3) 2

Efficiency correction 5

Total (pp) 9.6 8.7 9.8 8.9 15.4

Total (pPb) 9.6 (10.0) 9.2 (10.0) 9.8 (12.6) 9.4 (12.6) 15.6

Total (PbPb) 9.1 8.6 9.3 8.9 15.1



and



−,whichprovidesacross-check ofthesystematic uncer-tainty.Thistranslatesinto a systematicuncertaintyin the recon-structionefficiencyof7.8% forthe K0_S and



particles,and11.7% forthe



−particles.Differentbackgroundfitfunctionsand meth-ods to extract the yields for the K_S0,



, and



− are compared. The background fit function is varied to a fourth-order polyno-mialfortheK0_S and



studies,andtoalinearfunctionforthe



− study.Theyields areobtainedbyintegratingover aregionthatis

±

5timestheaverageresolutionandcenteredatthemean,rather thanoverthe entirefittedmass range.Possible contamination by residualmisidentified V0 _{candidates(i.e.,a K}0

S particle

misidenti-fiedasa



particle, orvice versa) isinvestigatedby varying the invariant mass range used to reject misidentified V0 candidates. Onthe basis of thesestudies we assign systematicuncertainties of2–4% to the yields. Systematic effects related to the selection ofthestrange-particlecandidatesareevaluatedbyvaryingthe se-lectioncriteria,resultinginanuncertaintyof1–7%.Theimpactof finitemomentumresolutiononthespectraisestimatedusingthe eposeventgenerator.Specifically,thegenerator-levelp_Tspectraof the strange particles are smeared by the momentum resolution, which is determined through comparison of the generator-level andmatched reconstructed-level particle information. The differ-ence betweenthe smeared and original spectra is less than 2%. Thesystematic uncertaintyassociated withnonprompt



correc-tions to the



spectra is evaluated through propagation of the systematic uncertainty in the Ncorr_−

/

Ncorr ratio in Eq. (2) to the

fresidual

,np factor,andisfoundto belessthan2%.Systematic

uncer-taintiesintroduced bypossibleresidual pileupeffectsforpp data are estimated to be 1–3%. This uncertainty is evaluated through bothtightening(only onereconstructedvertexallowedperevent) andloosening (no eventrejectionon the basis ofthe numberof vertices)thepileuprejectioncriteria[41].Theuncertainty associ-atedwithpileupisnegligibleforthepPb andPbPb datasincethere arevery feweventsinthose sampleswithmore thanone recon-structed vertex. In pPb collisions, the direction of the p and Pb beamswerereversedduringcourseofthedatacollection,as men-tioned in Section 2. Comparison of the particle pT spectra with

andwithoutthe beamreversalyields an uncertaintyof 2–5%for all particletypes. The effectof the choice ofthe rapidity bins is assessedbydividingeachbinintotwo,therebydoublingthe num-berof bins,resultingin a systematicuncertaintyof1–3% forthe pT spectra.Forthe



−,the reconstructionefficiencycorrectionis

smoothedby averaging adjacent bins inorder tocompensate for thelimitedstatisticalprecisionoftheMCsample.Variationsinthe smoothingprocedureleadtoasystematicuncertaintyof5%forthe pTspectraofthe



−.

All sources of systematic uncertainty are uncorrelated and summedinquadraturetodefinethetotalsystematicuncertainties inthe pTspectraofeachstrangeparticle.Thetotalsystematic

un-certaintiesbetweenthepp,pPb,andPbPb systemsaresimilarand largelycorrelated.When calculatingratiosofparticleyields,most ofthesystematicuncertaintiespartiallyorentirelycancel.For ex-ample,thesystematicuncertaintiesduetotrackingefficiencyand pileupforthe



/2K0_S ratioarenegligible.

6. Results

6.1. Multiplicitydependenceatmid-rapidity

The pT spectra of K0_S,



, and



− particles with

|

ycm

|

<

1 in

pp collisions at

√

s

=

7TeV (top),pPb collisionsat

√

s

=

5

.

02TeV (middle),andPbPb collisionsat

√

sN N

=

2

.

76TeV (bottom)are

pre-sentedin

Fig. 2

,fordifferentmultiplicity intervals. Duetodetails in the implementation of the dedicated high-multiplicity trigger thresholdsused toselectthe pp events,themultiplicity intervals forpp eventsdiffer slightlyfromthosefor pPb andPbPb events. The pTdifferential yieldisdefinedasdN2

/(

2

π

pT

)

dpTd y.Forthe

purposeofbetter visibility,the dataare scaledby factors of2−n_,

as indicated in the figure legend. A clear evolution of the spec-tral shapewithmultiplicity can be seenforeach particle species ineach collision system. Forhighermultiplicity events,the spec-tratendtobecomeflatter(i.e.,“harder”),indicatingalarger



KET



value.Withineachcollisionsystem,heavierparticles(e.g.,



−) ex-hibit a harder spectrum than lighter particles (K0_S), especially for high-multiplicityevents.

To examine the differences in the multiplicity dependence of the spectra in greater detail, the ratios



/2K0

S and



−/



of the

yields areshownin

Fig. 3

asa functionof pT fordifferent

multi-plicityrangesinthepp,pPb,andPbPb systems.Theresultsforthe



/2K0

S ratioare showninFig. 3(top).For pT



2GeV,the



/2K0S

ratioisseentobesmallerinhigh-multiplicityeventsthanin low-multiplicity eventsforagiven pT value.In pp andpPb collisions,

thistrendissimilartowhathasbeenobservedbetweenperipheral andcentralPbPb collisions[23];thistrendisnotasevidentforthe PbPb datain Fig. 3 (topright)because inthe presentstudyonly PbPb events of 50–100% centrality are considered. At higher pT,

this multiplicity ordering of the



/2K0_S ratio is reversed. In hy-drodynamic modelssuch asthose presentedinRefs. [51,52],this behaviorcanbeinterpretedastheeffectofradialflow.Astronger radialflowisdevelopedinhigher-multiplicityevents,whichboosts heavierparticles(e.g.,



) tohigher pT,resultingina suppression

ofthe



/2K0_S ratioatlow pT.Comparingthevariouscollision

low-Fig. 2. ThepTspectraofK0S,



,and



−particlesinthecenter-of-massrapidityrange|ycm|<1 inpp collisionsat√s=7TeV (top),pPb collisionsat√s=5.02TeV (middle), andPbPb collisionsat√sN N=2.76TeV (bottom)fordifferentmultiplicityintervals.Theinclusionofthecharge-conjugatestatesisimpliedfor



and



−particles.The

datainthedifferentmultiplicityintervalsarescaledbyfactorsof2−n_{forbettervisibility.Thestatisticaluncertaintiesaresmallerthanthemarkersandthesystematic}

uncertaintiesarenotshown.

andhigh-multiplicity eventsisseen tobelargestforthepp data. Inthehydrodynamic modelofRef. [31],smallercollisionsystems likepp producealargerradial-floweffectthanlargersystemslike pPb or PbPb, for similar multiplicities, which could explain this observation.ForpT

>

2GeV,thebaryonenhancementcouldbe

ex-plainedbyrecombinationmodels,inwhichfreequarksrecombine to form hadrons [53]. In previous studies (e.g., Ref. [54]), it has beenshownthat theaverage pT value ofvariousparticle species

hasonlya slightcenter-of-massenergydependence(10% athigh multiplicity).Thisdependenceisnot sufficientto explainthe dif-ferencesobservedin

Fig. 3

betweenthevarioussystems.

Foreach multiplicity interval,the



/2K0_S ratio reachesa max-imum that has a similar value for all three collision processes, and then decreases at higher pT. The location of the maximum

increaseswithmultiplicityfromaround pT

=

2 to 3GeV.

The results forthe



−/



ratio are shown in Fig. 3 (bottom). Inthiscase,thedifferencebetweenthelow- andhigh-multiplicity eventsismuchsmallerthanforthe



/2K0_S ratio,forallthree col-lisionssystems.Forallsystems,the



−/



ratioincreaseswithpT

andreachesaplateauataround pT

=

3GeV.Duetothelarge

sys-tematic uncertainty, it isnot possibleto draw a conclusion with respecttotheradial-flowinterpretation.

Motivatedby thehydrodynamicmodel,weperforma simulta-neous fit of a blast-wave function [32] to the K0_S and



spectra in Fig. 2. The fits are restricted to low pT because that is the

region in which the blast-wave model is valid. The blast-wave model is strictly appropriate only fordirectly produced particles, while about1/3 oftheK0_S mesonsmaybe fromhighermass res-onances [55]. The



− particle isnot used in thefit asthere are not many



−atlow pT.Thefitsareperformedforeach collision

systemseparately.Thefitrangesare0

.

1

<

pT

<

1

.

5GeV fortheK0S

and0

.

6

<

pT

<

3

.

0GeV forthe



.Thefittedfunctionis: 1 pT dN dpT

∼

R



0 r dr mTI0



pTsinh

ρ

Tkin

K1



mTcosh

ρ

Tkin

,

(3)

where

ρ

=

tanh−1

β

T

=

tanh−1

β

s

(

r

/

R

)

n



is the velocity profile, R is the radius of the medium (set to unity in the fit), r is the radial distancefrom the centerof the medium in the transverse plane,n istheexponentofthevelocityprofile,

β

_Tisthetransverse

The CMS Collaboration / Physics Letters B 768 (2017) 103–129 109

Fig. 3. RatiosofpTspectrafor



/2KS0(top)and



−/(bottom)inthecenter-of-massrapidityrange|ycm|

<

1.0 forpp collisionsat√s=7TeV (left),pPb collisionsat √

s=5.02TeV (middle),andPbPb collisionsat√sN N=2.76TeV (right).Two(forpp)orthree(forpPb andPbPb)representativemultiplicityintervalsarepresented.The

inclusionofthecharge-conjugatestatesisimpliedfor



and



− particles.Theerrorbarsrepresentthestatisticaluncertainties,whiletheboxesindicatethesystematic uncertainties.

expansionvelocity (also known asthe radial-flow velocity),

β

_s is thetransverse expansion velocity on the surface of the medium, Tkin isthekineticfreeze-outtemperature,andI0andK1 are

mod-ifiedBesselfunctions.Thefittedparametersthatgoverntheshape aren,

β

_s,andTkin.

Inthe blast-wave model,common values of Tkin andaverage

radial-flow velocity

β

T



are assumed for all particle species, as

isexpected ifthe systemis locally thermalizedand undergoes a radial-flowexpansion.Itisusefultodirectlycomparetheextracted values of Tkin and

β

T



fromthe different systems to study the

system-sizedependenceatsimilarmultiplicities.

TheextractedvaluesofTkinand

β

T



areshownin

Fig. 4

forthe

sixpp andfortheeightpPb andPbPb multiplicityintervals.Inthis figure,themultiplicityincreasesfromlefttoright.Theellipses cor-respondtoone standard deviationstatisticaluncertainties, which forpp collisionsaresmalleratlowandhighmultiplicityduetothe useofevents collected withminimumbias andhigh-multiplicity triggers. Systematicuncertainties, which are evaluated by propa-gatingthe systematicuncertaintiesfromthe spectratothe blast-wave fits and altering the fit ranges, are on the order of a few percentandarenotshown.Examplesofthefitsareshownin

Fig. 5

foralow- andhigh-multiplicityrangeinpPb collisions.Ingeneral, thefit quality is goodfor high-multiplicity eventsexcept forthe lowest pT range,while for low-multiplicityevents there are

dis-crepanciesontheorderof5%.However,thediscrepanciesbetween thefitanddataliewithinthesystematicuncertainty.

Theprecisemeaningofthe Tkin and

β

T



parametersismodel

dependent,andthey shouldnot beinterpreted literallyasthe ki-netic freeze-out temperature andradial-flow velocity of the sys-tem. The mainpurpose ofFig. 4 is toprovide a qualitative com-parisonofthespectralshapesinthethreesystems.Inthecontext

Fig. 4. Theextractedkineticfreeze-outtemperature,Tkin,versustheaverage radial-flowvelocity,βT,fromasimultaneousblast-wavefittotheK0Sand



pTspectra at|ycm|<1 fordifferentmultiplicityintervalsinpp,pPb,andPbPb collisions.The sixpp andeightpPb andPbPb multiplicityintervalsareindicatedinthelegendof Fig. 2.Fortheresultsinthisplot,themultiplicityincreasesfromlefttoright.The correlationellipsesrepresentthestatisticaluncertainties.Systematicuncertainties, whichareevaluatedtobeontheorderofafewpercent,arenotshown.

oftheblast-wavemodel,whencomparingatsimilarmultiplicities, theTkinparameterhasthesamevaluewithin15%amongthethree

systems, while the

β

T



parameter is larger when the system is

smaller, i.e.,

β

T



pp

>

β

T



pPb

>

β

T



PbPb. Thisis qualitatively

con-sistent withthe predictionofRef. [31].The results ofblast-wave fitsare knowntodependontheparticlespecies. Duetothe lim-ited setofparticlesinthisanalysis,futurestudieswillbe needed tofurthersubstantiatetheconclusions.

Theevolution ofthe pT spectra withmultiplicity canbe

com-paredmoredirectlybetweenthethreesystemsthrough examina-tion ofthe



KET



value. The



KET



values at

|

ycm

|

<

1 for K0S,



,

Fig. 5. Examplesofsimultaneousblast-wavefitsofthepTspectraofKS0and



particlesinlow- andhigh-multiplicitypPb events.Theinclusionofthecharge-conjugatestates isimpliedfor



particles.TheratiosofthefitstothedataasafunctionofpTareshowninthebottompanels.Theuncertaintiesarestatisticalonlyandaretoosmalltobe visibleformostofthepoints.

Fig. 6. Theaveragetransversekineticenergy,KET,at|ycm|<1 forK0S,



,and



− particlesasafunctionofmultiplicityinpp,pPb,andPbPb collisions.Theinclusionof thecharge-conjugatestatesisimpliedfor



and



−particles.Forthe



−,onlyresultsfrompPb collisionsareshown.Theerrorbarsrepresentthestatisticaluncertainties, whiletheboxesindicatethesystematicuncertainties.

and



− particlesasafunctionofmultiplicityareshownin

Fig. 6

. Extrapolation ofthe pT spectra down to pT

=

0GeV is a crucial

stepinextractingthe



KET



values,whiletheimpactofthe

extrap-olationupto pT

≈ ∞

isnegligible,bothonthevalueof



KET



and

its uncertainty.Forthe



− particle,onlyresultsin pPb collisions are shown due to the limitation of the low-pT reach in pp and

PbPb collisions,ascan beseen fromFig. 2.Blast-wavefits tothe individual spectra, which only consider the spectrum shape but donot impose anyphysics constraint, areused to obtainthe ex-trapolation.The fractionoftheextrapolatedyieldwithrespectto thetotal yield is about1.2–2.5%forthe K0

S, 5.8–15.1%forthe



,

and5.4–20.4% forthe



− particles, depending on the multiplic-ity.Alternativemethods toperform theextrapolationare usedto evaluateasystematicuncertainty,includinguseofthepredictions fromthesimultaneousblast-wavefittotheK0

S and



pT spectra,

anda linear extrapolation fromthe yields ina low range of pT.

Thesystematicuncertaintiesfrom

Table 2

arealsoincludedinthe evaluationofthe



KET



uncertainties.

Forthelowestmultiplicityrange,the



KET



valuesforeach

par-ticlespecies areseen to besimilar. Forall particle species,



KET



increases with increasing multiplicity. However, the slope of the increase differs for different particles, with the heavier particles exhibiting a faster growth in



KET



for all systems. For a given

multiplicity range,the



KET



value isroughly proportional to the

particle’s mass. In PbPb collisions, this can be understood to be dueto theonsetofradial flow [2,7].The observeddifference be-tween particlespeciesathighmultiplicity isseentobelargerfor pp andpPb eventsthanforPbPb events.Note,however,the differ-enceinthecenter-of-massenergiesbetweenthethreesystems. 6.2. RapiditydependenceinpPb events

The rapidity dependence of the pT spectra of the K0_S and



particlesisstudiedinthepPb data.Noresultsfor



−particlesare presentedduetostatisticallimitations.AsapPb collisionis asym-metric in rapidity, it is interesting to compare the spectra along thePb-going( ycm

<

0) andp-going( ycm

>

0)directions[37].The

The CMS Collaboration / Physics Letters B 768 (2017) 103–129 111

Fig. 7. ThepTspectraofK0S and



particlesindifferent ycm rangesfor pPb collisionsat √s=5.02TeV.Theinclusionofthecharge-conjugatestatesisimpliedfor



andparticles.Resultsareshownforthreemultiplicityranges:0≤Noffline

trk <35 (top),120≤N offline

trk <150 (middle),and220≤N offline

trk <260 (bottom).Withineachpanel, thecurvesontoprepresentPb-goingeventsandthecurvesonbottomp-goingevents.Thedatainthedifferentrapidityintervalsarescaledbyfactorsof2−2n _forbetter visibility.Thestatisticaluncertaintiesaresmallerthanthemarkersandthesystematicuncertaintiesarenotshown.

pTspectraofK0_S and



particlesindifferentycmrangesareshown

in

Fig. 7

forsmall(top),intermediate(middle),andlarge(bottom) averagemultiplicities.

The



/2K0_S ratios fromthe

−

1

.

5

<

ycm

<

−

0

.

8 (Pb-going) and

0

.

8

<

ycm

<

1

.

5 (p-going)rapidityregions are comparedin

Fig. 8

formultiplicity ranges 0

≤

Noffline_trk

<

35 and 220

≤

Noffline_trk

<

260. Forboththelow-multiplicityandthehigh-multiplicityevents,the



/2K0_SratiofromthePb-goingdirectionliesabovetheresultsfrom thep-goingdirection,withthelargestdifferenceobservedathigh pT inthehigh-multiplicitysample.

Asafurtherstudy,wecalculate



KET



,followingtheprocedure

outlined inSection 6.1, andexamine its dependence on ycm for

K0_S and



particlesinthepPb collisions.Theresultsareshownin

Fig. 8. RatiosofpTspectra,



/2K0S,fromthe−1.5<ycm<−0.8 (Pb-going)and0.8

<

ycm<1.5 (p-going)rapidityregionsinpPb collisionsat√s=5.02TeV.Theinclusion ofthecharge-conjugatestatesisimpliedfor



particles.Resultsarepresentedfortwomultiplicityranges0≤Noffline

trk <35 (left)and220≤Ntrkoffline<260 (right).Theerror barsrepresentthestatisticaluncertainties,whiletheboxesindicatethesystematicuncertainties.

Fig. 9. Theaveragetransversekineticenergy,KET,asafunctionofycmfortheK0S and



particlesinthreerangesofmultiplicityinpPb collisionsat √

s=5.02TeV.The inclusionofthecharge-conjugatestatesisimpliedfor



particles.Theerrorbarsrepresentthestatisticaluncertainties,whiletheboxesindicatethesystematicuncertainties.

arelarge,the



KET



valuesareseentobecomeslightlyasymmetric

asmultiplicity increases. At low multiplicities (0

≤

Noffline_trk

<

35), theratiosof



KET



betweenthePb-goingside(

−

1

.

5

<

ycm

<

−

0

.

8)

andthep-going side(0

.

8

<

ycm

<

1

.

5) are 1

.

01

±

0

.

01 (syst.) for

K0

S particlesand1

.

04

±

0

.

05 (syst.)for



particles,both ofwhich

areconsistentwithunitywithin thesystematicuncertainties (the statistical uncertainties are negligible). However, in the highest multiplicity range,220

≤

Noffline_trk

<

260,theratios become 1

.

06

±

0

.

01 (syst.)forK0_S particlesand1

.

12

±

0

.

06 (syst.)for



particles, suggestingthatan asymmetry in



KET



isdeveloped betweenthe

Pb-going andp-going sides. This trend is qualitatively consistent withthehydrodynamicpredictionforpPb collisions[37].

7. Summary

Measurements of strange hadron (K0_S,



+

, and



−

+

+) transversemomentumspectrainpp, pPb,andPbPb collisionsare presentedoverawiderangeofeventcharged-particlemultiplicity andparticle rapidity.The studyis basedon samples of pp colli-sionsat

√

s

=

7TeV,pPb collisionsat

√

s

=

5

.

02TeV,andPbPb col-lisionsat

√

sN N

=

2

.

76TeV,collectedwiththeCMSdetectoratthe

LHC.Inthecontextofhydrodynamic models,themeasured parti-clespectraarefittedwitha blastwave function,which describes anexpandingfluid-likesystem.Whencomparingatasimilar mul-tiplicity,theextractedradial-flowvelocityparametersarefoundto belargerinpp andpPb collisionsthanthatinPbPb collisions.The averagetransverse kinetic energy



KET



of strangehadronsis

ob-served to increase with multiplicity,with a strongerincrease for

heavier particles.At similar multiplicities, thedifference in



KET



betweenthestrange-particlespeciesislargerinthesmallerpp and pPb systemsthaninthePbPb system.ForpPb collisions,



KET



in

thePb-goingdirectionforK0_S(



+

)is6%(12%)largerthaninthe p-goingdirectionforeventswiththehighestparticlemultiplicities.

Acknowledgements

WecongratulateourcolleaguesintheCERNaccelerator depart-ments for the excellent performance of the LHC and thank the technical andadministrativestaffs atCERNand atother CMS in-stitutes for their contributions to the success of the CMS effort. Inaddition,wegratefullyacknowledgethecomputingcentersand personneloftheWorldwideLHCComputingGridfordeliveringso effectively thecomputinginfrastructure essentialto our analyses. Finally, we acknowledge the enduring support for the construc-tion andoperationofthe LHCandtheCMSdetectorprovided by thefollowingfundingagencies:BMWFWandFWF(Austria);FNRS and FWO (Belgium); CNPq, CAPES, FAPERJ, and FAPESP (Brazil); MES (Bulgaria); CERN; CAS, MoST, and NSFC (China); COLCIEN-CIAS (Colombia); MSES and CSF (Croatia); RPF (Cyprus); MoER, ERC IUT andERDF (Estonia); Academy of Finland, MEC, andHIP (Finland); CEA and CNRS/IN2P3 (France); BMBF, DFG, and HGF (Germany);GSRT(Greece);OTKAandNIH(Hungary);DAEandDST (India);IPM(Iran); SFI(Ireland);INFN (Italy);MSIP andNRF (Re-public ofKorea);LAS(Lithuania);MOE andUM(Malaysia);BUAP, CINVESTAV, CONACYT, LNS, SEP, and UASLP-FAI (Mexico); MBIE (New Zealand); PAEC (Pakistan); MSHE and NSC (Poland); FCT (Portugal); JINR(Dubna); MON,RosAtom, RASandRFBR (Russia);

The CMS Collaboration / Physics Letters B 768 (2017) 103–129 113

MESTD(Serbia); SEIDIandCPAN(Spain); SwissFundingAgencies (Switzerland); MST (Taipei); ThEPCenter, IPST, STAR and NSTDA (Thailand);TUBITAKandTAEK(Turkey);NASUandSFFR(Ukraine); STFC(UnitedKingdom);DOEandNSF(USA).

Individuals have received support from the Marie-Curie pro-gramandtheEuropeanResearchCouncilandEPLANET(European Union); the Leventis Foundation; the A.P. Sloan Foundation; the AlexandervonHumboldt Foundation;theBelgianFederal Science Policy Office; the Fonds pour la Formation à la Recherche dans l’Industrie et dans l’Agriculture (FRIA-Belgium); the Agentschap voor Innovatie door Wetenschap en Technologie (IWT-Belgium); theMinistryofEducation, YouthandSports (MEYS)ofthe Czech Republic;theCouncilofScienceandIndustrialResearch,India;the HOMING PLUSprogram ofthe Foundation forPolish Science, co-financed from European Union, Regional Development Fund; the MobilityPlusprogramoftheMinistryofScienceandHigher Edu-cation(Poland);theOPUSprogramoftheNationalScienceCenter (Poland);the Thalis andAristeia programs cofinancedby EU-ESF andtheGreek NSRF;the NationalPrioritiesResearch Programby Qatar National Research Fund; the Programa Clarín-COFUND del Principado de Asturias; the Rachadapisek Sompot Fund for Post-doctoralFellowship,ChulalongkornUniversity(Thailand);the Chu-lalongkorn Academic into Its 2nd Century Project Advancement Project(Thailand);andtheWelchFoundation,contractC-1845.

References

[1] E.Andersen,etal.,Enhancementofcentral



,



and yieldsinPb–Pb colli-sionsat158AGeV/c,Phys.Lett.B433(1998)209,http://dx.doi.org/10.1016/ S0370-2693(98)00689-3.

[2] J.Adams,etal.,STAR,Experimentalandtheoreticalchallengesinthesearchfor thequarkgluonplasma:TheSTARCollaboration’scriticalassessmentofthe evidencefromRHICcollisions,Nucl.Phys.A757(2005)102,http://dx.doi.org/ 10.1016/j.nuclphysa.2005.03.085,arXiv:nucl-ex/0501009.

[3] J.Rafelski,B.Muller,Strangenessproductioninthequark–gluonplasma,Phys. Rev.Lett.48(1982)1066,http://dx.doi.org/10.1103/PhysRevLett.48.1066. [4] B.I.Abelev,etal.,STAR,EnhancedstrangebaryonproductioninAu+Au

colli-sionscomparedtop+p at√sN N=200 GeV,Phys.Rev.C77(2008)044908,

http://dx.doi.org/10.1103/PhysRevC.77.044908,arXiv:0705.2511.

[5] E. Andersen, et al., WA97, Strangeness enhancement at mid-rapidity inPb Pbcollisionsat 158-A-GeV/c,Phys.Lett.B449(1999)401,http://dx.doi.org/ 10.1016/S0370-2693(99)00140-9.

[6] J. Adams, et al., STAR, Multistrange baryon production inAu–Au collisions at √sN N=130 GeV, Phys. Rev. Lett. 92 (2004) 182301, http://dx.doi.org/

10.1103/PhysRevLett.92.182301,arXiv:nucl-ex/0307024.

[7] K. Adcox, et al., PHENIX, Formation of dense partonic matter in rela-tivistic nucleus–nucleus collisions at RHIC: experimentalevaluation by the PHENIXCollaboration,Nucl.Phys.A757(2005)184,http://dx.doi.org/10.1016/ j.nuclphysa.2005.03.086,arXiv:nucl-ex/0410003.

[8] CMSCollaboration, Observationoflong-rangenear-sideangular correlations inproton–protoncollisionsat theLHC,J.HighEnergy Phys.09(2010)091, http://dx.doi.org/10.1007/JHEP09(2010)091,arXiv:1009.4122.

[9] CMSCollaboration,Observationoflong-range,near-sideangularcorrelations inpPbcollisionsattheLHC,Phys.Lett.B718(2013)795,http://dx.doi.org/ 10.1016/j.physletb.2012.11.025,arXiv:1210.5482.

[10] ATLASCollaboration,Observationofassociatednear-sideandaway-side long-rangecorrelations in√sNN=5.02 TeV proton–leadcollisions with the AT-LAS detector, Phys. Rev.Lett. 110(2013) 182302, http://dx.doi.org/10.1103/ PhysRevLett.110.182302,arXiv:1212.5198.

[11] ALICECollaboration, Long-rangeangular correlations onthe nearand away side in pPb collisions at √sNN=5.02 TeV, Phys. Lett. B 719 (2013) 29, http://dx.doi.org/10.1016/j.physletb.2013.01.012,arXiv:1212.2001.

[12] K.Dusling, W. Li, B. Schenke, Novel collective phenomena in high-energy proton–protonand proton–nucleuscollisions,Int.J.Mod. Phys.E 25(2016) 1630002,http://dx.doi.org/10.1142/S0218301316300022,arXiv:1509.07939. [13] P.Bo ˙zek,Collectiveflowin p–Pbandd–Pbcollisionsat TeVenergies,Phys.

Rev.C85(2012)014911,http://dx.doi.org/10.1103/PhysRevC.85.014911,arXiv: 1112.0915.

[14] P.Bo ˙zek, W.Broniowski, Correlationsfrom hydrodynamic flow inpPb col-lisions, Phys. Lett. B 718 (2013) 1557, http://dx.doi.org/10.1016/j.physletb. 2012.12.051,arXiv:1211.0845.

[15] A.Bzdak,B.Schenke,P.Tribedy,R.Venugopalan,Initialstategeometryandthe roleofhydrodynamicsinproton–proton,proton–nucleusanddeuteron–nucleus

collisions,Phys.Rev.C87(2013)064906,http://dx.doi.org/10.1103/PhysRevC. 87.064906,arXiv:1304.3403.

[16] K.Werner,I.Karpenko,T.Pierog,“Ridge”inproton–protonscatteringat7 TeV, Phys. Rev. Lett. 106 (2011) 122004, http://dx.doi.org/10.1103/PhysRevLett. 106.122004,arXiv:1011.0375.

[17] K.Dusling,R.Venugopalan,ExplanationofsystematicsofCMSp+Pbhigh mul-tiplicitydi-hadrondataat √sNN=5.02 TeV,Phys.Rev.D87(2013)054014, http://dx.doi.org/10.1103/PhysRevD.87.054014,arXiv:1211.3701.

[18] K.Dusling,R.Venugopalan,EvidenceforBFKLandsaturationdynamicsfrom dihadronspectraattheLHC,Phys.Rev.D87(2013)051502,http://dx.doi.org/ 10.1103/PhysRevD.87.051502,arXiv:1210.3890.

[19] A.Dumitru,L.McLerran,V.Skokov,Azimuthalasymmetriesandtheemergence of“collectivity”frommulti-particlecorrelationsinhigh-energypAcollisions, Phys.Lett.B743(2015)134,http://dx.doi.org/10.1016/j.physletb.2015.02.046, arXiv:1410.4844.

[20] M.Gyulassy,P.Levai,I.Vitev,T.S.Biro,Non-Abelianbremsstrahlungand az-imuthalasymmetriesinhighenergy p+A reactions,Phys.Rev.D90(2014) 054025,http://dx.doi.org/10.1103/PhysRevD.90.054025,arXiv:1405.7825. [21] W.Li,Observationofa‘ridge’correlationstructureinhighmultiplicityproton–

proton collisions: a brief review, Mod. Phys. Lett. A 27 (2012) 1230018, http://dx.doi.org/10.1142/S0217732312300182,arXiv:1206.0148.

[22] L. He,T.Edmonds,Z.-W.Lin,F.Liu,D.Molnar,F.Wang,Anisotropicparton escape isthedominantsourceofazimuthalanisotropyintransportmodels, Phys. Lett.B753(2016)506,http://dx.doi.org/10.1016/j.physletb.2015.12.051, arXiv:1502.05572.

[23] ALICE Collaboration,Centralitydependence ofπ, K ,p production inPb–Pb collisionsat√sNN=2.76 TeV,Phys.Rev.C88(2013)044910,http://dx.doi.org/ 10.1103/PhysRevC.88.044910,arXiv:1303.0737.

[24] ALICE Collaboration, Elliptic flow of identified hadrons inPb–Pb collisions at √sNN=2.76 TeV,J. High Energy Phys. 06 (2015)190, http://dx.doi.org/ 10.1007/JHEP06(2015)190,arXiv:1405.4632.

[25] H.Song,S.Bass,U.W.Heinz,Spectraandellipticflowforidentifiedhadronsin 2.76ATeVPb+Pbcollisions,Phys.Rev.C89(2014)034919,http://dx.doi.org/ 10.1103/PhysRevC.89.034919,arXiv:1311.0157.

[26] X.Zhu,F.Meng,H.Song,Y.-X.Liu,Hybridmodelapproachforstrangeand multistrangehadronsin2.76ATeVPb+Pbcollisions,Phys. Rev.C91(2015) 034904,http://dx.doi.org/10.1103/PhysRevC.91.034904,arXiv:1501.03286. [27] ALICECollaboration,Multiplicitydependenceofpion,kaon,protonandlambda

productioninp–pbcollisionsat√sNN=5.02 TeV,Phys.Lett.B728(2014)25, http://dx.doi.org/10.1016/j.physletb.2013.11.020,arXiv:1307.6796.

[28] CMSCollaboration,Studyoftheproductionofchargedpions,kaons,and pro-tons inpPbcollisionsat √sNN=5.02 TeV,Eur.Phys. J.C74 (2014)2847, http://dx.doi.org/10.1140/epjc/s10052-014-2847-x,arXiv:1307.3442.

[29] ALICE Collaboration,Long-rangeangularcorrelations ofπ,Kandpinp–Pb collisionsat√sNN=5.02 TeV,Phys.Lett.B726(2013)164,http://dx.doi.org/ 10.1016/j.physletb.2013.08.024,arXiv:1307.3237.

[30] CMS Collaboration, Long-rangetwo-particle correlations ofstrange hadrons with charged particles in pPb and PbPb collisions at LHC energies, Phys. Lett.B742(2015)200,http://dx.doi.org/10.1016/j.physletb.2015.01.034, arXiv:1409.3392.

[31] E.Shuryak,I.Zahed,High-multiplicity pp andp A collisions:hydrodynamics atitsedge,Phys.Rev.C88(2013)044915,http://dx.doi.org/10.1103/PhysRevC. 88.044915,arXiv:1301.4470.

[32] E.Schnedermann,J.Sollfrank,U.Heinz,Thermalphenomenologyofhadrons from200A GeV S+Scollisions,Phys.Rev.C48(1993)2462,http://dx.doi.org/ 10.1103/PhysRevC.48.2462,arXiv:nucl-th/9307020.

[33] C. Albajar, et al., UA1, A study of the general characteristics of proton– antiprotoncollisionsat√s=0.2 TeV to0.9TeV,Nucl.Phys.B335(1990)261, http://dx.doi.org/10.1016/0550-3213(90)90493-W.

[34] T. Alexopoulos, et al., Multiplicity dependence of the transverse momen-tumspectrumforcentrallyproducedhadronsinantiproton–protoncollisions at √s=1.8 TeV,Phys. Rev.Lett. 60(1988) 1622, http://dx.doi.org/10.1103/ PhysRevLett.60.1622.

[35] B.I.Abelev,etal.,STAR,Systematicmeasurementsofidentifiedparticlespectra inpp,d+Au andAu+Au collisionsfromSTAR,Phys.Rev.C79(2009)034909, http://dx.doi.org/10.1103/PhysRevC.79.034909,arXiv:0808.2041.

[36] A.OrtizVelasquez,P.Christiansen,E.CuautleFlores,I.MaldonadoCervantes, G. Pai ´c, Color reconnection and flowlike patterns in pp collisions, Phys. Rev.Lett.111(2013)042001,http://dx.doi.org/10.1103/PhysRevLett.111.042001, arXiv:1303.6326.

[37] P.Bo ˙zek,A.Bzdak,V.Skokov,Therapiditydependenceoftheaverage trans-versemomentuminp+Pbcollisionsat theLHC:thecolorglasscondensate versushydrodynamics,Phys.Lett.B728(2014)662,http://dx.doi.org/10.1016/ j.physletb.2013.12.034,arXiv:1309.7358.

[38] CMS Collaboration, Description and performance of track and primary-vertex reconstruction with the CMS tracker, J. Instrum. 9 (2014) P10009, http://dx.doi.org/10.1088/1748-0221/9/10/P10009,arXiv:1405.6569.

[39] CMSCollaboration,TheCMSexperimentattheCERNLHC,J.Instrum.3(2008) S08004,http://dx.doi.org/10.1088/1748-0221/3/08/S08004.

ton and particle production in high-energy hadronic and nuclear colli-sions, Comput. Phys. Commun. 83 (1994) 307, http://dx.doi.org/10.1016/ 0010-4655(94)90057-4,arXiv:nucl-th/9502021.

[46] T. Sjöstrand,S. Mrenna, P.Skands,PYTHIA 6.4physics and manual,J. High EnergyPhys.05(2006)026,http://dx.doi.org/10.1088/1126-6708/2006/05/026, arXiv:hep-ph/0603175.

[47] T.Sjöstrand,S.Mrenna,P.Skands,AbriefintroductiontoPYTHIA8.1,Comput. Phys. Commun. 178 (2008) 852, http://dx.doi.org/10.1016/j.cpc.2008.01.036, arXiv:0710.3820.

quarkgluonplasma,Annu.Rev.Nucl.Part.Sci.58(2008)177,http://dx.doi.org/ 10.1146/annurev.nucl.58.110707.171134,arXiv:0807.4939.

[54] CMSCollaboration,Studyoftheinclusiveproductionofchargedpions,kaons, and protons in pp collisions at √s=0.9, 2.76, and 7 TeV, Eur. Phys. J. C 72 (2012) 2164, http://dx.doi.org/10.1140/epjc/s10052-012-2164-1, arXiv: 1207.4724.

[55] ALICECollaboration,ProductionofK∗(892)0_and

_φ

_{(1020)inp–Pbcollisionsat} √

sNN=5.02 TeV,Eur.Phys.J.C76(2016)245,http://dx.doi.org/10.1140/epjc/ s10052-016-4088-7,arXiv:1601.07868.

TheCMSCollaboration

V. Khachatryan,

A.M. Sirunyan,

A. Tumasyan

YerevanPhysicsInstitute,Yerevan,Armenia

W. Adam,

E. Asilar,

T. Bergauer,

J. Brandstetter,

E. Brondolin,

M. Dragicevic,

J. Erö,

M. Flechl,

M. Friedl,

R. Frühwirth

1

,

V.M. Ghete,

C. Hartl,

N. Hörmann,

J. Hrubec,

M. Jeitler

1

,

A. König,

M. Krammer

1

,

I. Krätschmer,

D. Liko,

T. Matsushita,

I. Mikulec,

D. Rabady,

N. Rad,

B. Rahbaran,

H. Rohringer,

J. Schieck

1

,

J. Strauss,

W. Treberer-Treberspurg,

W. Waltenberger,

C.-E. Wulz

1

InstitutfürHochenergiephysikderOeAW,Wien,Austria

V. Mossolov,

N. Shumeiko,

J. Suarez Gonzalez

NationalCentreforParticleandHighEnergyPhysics,Minsk,Belarus

S. Alderweireldt,

T. Cornelis,

E.A. De Wolf,

X. Janssen,

A. Knutsson,

J. Lauwers,

S. Luyckx,

M. Van De Klundert,

H. Van Haevermaet,

P. Van Mechelen,

N. Van Remortel,

A. Van Spilbeeck

UniversiteitAntwerpen,Antwerpen,Belgium

S. Abu Zeid,

F. Blekman,

J. D’Hondt,

N. Daci,

I. De Bruyn,

K. Deroover,

N. Heracleous,

J. Keaveney,

S. Lowette,

S. Moortgat,

L. Moreels,

A. Olbrechts,

Q. Python,

D. Strom,

S. Tavernier,

W. Van Doninck,

P. Van Mulders,

I. Van Parijs

VrijeUniversiteitBrussel,Brussel,Belgium

H. Brun,

C. Caillol,

B. Clerbaux,

G. De Lentdecker,

G. Fasanella,

L. Favart,

R. Goldouzian,

A. Grebenyuk,

G. Karapostoli,

T. Lenzi,

A. Léonard,

T. Maerschalk,

A. Marinov,

A. Randle-conde,

T. Seva,

C. Vander Velde,

P. Vanlaer,

R. Yonamine,

F. Zenoni,

F. Zhang

2

UniversitéLibredeBruxelles,Bruxelles,Belgium

L. Benucci,

A. Cimmino,

S. Crucy,

D. Dobur,

A. Fagot,

G. Garcia,

M. Gul,

J. Mccartin,

A.A. Ocampo Rios,

D. Poyraz,

D. Ryckbosch,

S. Salva,

R. Schöfbeck,

M. Sigamani,

M. Tytgat,

W. Van Driessche,

E. Yazgan,

N. Zaganidis

The CMS Collaboration / Physics Letters B 768 (2017) 103–129 115

C. Beluffi

3

,

O. Bondu,

S. Brochet,

G. Bruno,

A. Caudron,

L. Ceard,

S. De Visscher,

C. Delaere,

M. Delcourt,

L. Forthomme,

B. Francois,

A. Giammanco,

A. Jafari,

P. Jez,

M. Komm,

V. Lemaitre,

A. Magitteri,

A. Mertens,

M. Musich,

C. Nuttens,

K. Piotrzkowski,

L. Quertenmont,

M. Selvaggi,

M. Vidal Marono,

S. Wertz

UniversitéCatholiquedeLouvain,Louvain-la-Neuve,Belgium

N. Beliy,

G.H. Hammad

UniversitédeMons,Mons,Belgium

W.L. Aldá Júnior,

F.L. Alves,

G.A. Alves,

L. Brito,

M. Correa Martins Junior,

M. Hamer,

C. Hensel,

A. Moraes,

M.E. Pol,

P. Rebello Teles

CentroBrasileirodePesquisasFisicas,RiodeJaneiro,Brazil

E. Belchior Batista Das Chagas,

W. Carvalho,

J. Chinellato

4

,

A. Custódio,

E.M. Da Costa,

D. De Jesus Damiao,

C. De Oliveira Martins,

S. Fonseca De Souza,

L.M. Huertas Guativa,

H. Malbouisson,

D. Matos Figueiredo,

C. Mora Herrera,

L. Mundim,

H. Nogima,

W.L. Prado Da Silva,

A. Santoro,

A. Sznajder,

E.J. Tonelli Manganote

4

,

A. Vilela Pereira

UniversidadedoEstadodoRiodeJaneiro,RiodeJaneiro,Brazil

S. Ahuja

a

,

C.A. Bernardes

b

,

A. De Souza Santos

b

,

S. Dogra

a

,

T.R. Fernandez Perez Tomei

a

,

E.M. Gregores

b

,

P.G. Mercadante

b

,

C.S. Moon

a

,

5

,

S.F. Novaes

a

,

Sandra S. Padula

a

,

D. Romero Abad

b

,

J.C. Ruiz Vargas

a_{UniversidadeEstadualPaulista,SãoPaulo,Brazil} b_{UniversidadeFederaldoABC,SãoPaulo,Brazil}

A. Aleksandrov,

R. Hadjiiska,

P. Iaydjiev,

M. Rodozov,

S. Stoykova,

G. Sultanov,

M. Vutova

InstituteforNuclearResearchandNuclearEnergy,Sofia,Bulgaria

A. Dimitrov,

I. Glushkov,

L. Litov,

B. Pavlov,

P. Petkov

UniversityofSofia,Sofia,Bulgaria

W. Fang

6

BeihangUniversity,Beijing,China

M. Ahmad,

J.G. Bian,

G.M. Chen,

H.S. Chen,

M. Chen,

T. Cheng,

R. Du,

C.H. Jiang,

D. Leggat,

R. Plestina

7

,

F. Romeo,

S.M. Shaheen,

A. Spiezia,

J. Tao,

C. Wang,

Z. Wang,

H. Zhang

InstituteofHighEnergyPhysics,Beijing,China

C. Asawatangtrakuldee,

Y. Ban,

Q. Li,

S. Liu,

Y. Mao,

S.J. Qian,

D. Wang,

Z. Xu

StateKeyLaboratoryofNuclearPhysicsandTechnology,PekingUniversity,Beijing,China

C. Avila,

A. Cabrera,

L.F. Chaparro Sierra,

C. Florez,

J.P. Gomez,

B. Gomez Moreno,

J.C. Sanabria

UniversidaddeLosAndes,Bogota,Colombia

N. Godinovic,

D. Lelas,

I. Puljak,

P.M. Ribeiro Cipriano

UniversityofSplit,FacultyofElectricalEngineering,MechanicalEngineeringandNavalArchitecture,Split,Croatia

Z. Antunovic,

M. Kovac

UniversityofSplit,FacultyofScience,Split,Croatia

V. Brigljevic,

D. Ferencek,

K. Kadija,

J. Luetic,

S. Micanovic,

L. Sudic

InstituteRudjerBoskovic,Zagreb,Croatia

B. Calpas,

M. Kadastik,

M. Murumaa,

L. Perrini,

M. Raidal,

A. Tiko,

C. Veelken

NationalInstituteofChemicalPhysicsandBiophysics,Tallinn,Estonia

P. Eerola,

J. Pekkanen,

M. Voutilainen

DepartmentofPhysics,UniversityofHelsinki,Helsinki,Finland

J. Härkönen,

V. Karimäki,

R. Kinnunen,

T. Lampén,

K. Lassila-Perini,

S. Lehti,

T. Lindén,

P. Luukka,

T. Peltola,

J. Tuominiemi,

E. Tuovinen,

L. Wendland

HelsinkiInstituteofPhysics,Helsinki,Finland

J. Talvitie,

T. Tuuva

LappeenrantaUniversityofTechnology,Lappeenranta,Finland

M. Besancon,

F. Couderc,

M. Dejardin,

D. Denegri,

B. Fabbro,

J.L. Faure,

C. Favaro,

F. Ferri,

S. Ganjour,

A. Givernaud,

P. Gras,

G. Hamel de Monchenault,

P. Jarry,

E. Locci,

M. Machet,

J. Malcles,

J. Rander,

A. Rosowsky,

M. Titov,

A. Zghiche

DSM/IRFU,CEA/Saclay,Gif-sur-Yvette,France

A. Abdulsalam,

I. Antropov,

S. Baffioni,

F. Beaudette,

P. Busson,

L. Cadamuro,

E. Chapon,

C. Charlot,

O. Davignon,

L. Dobrzynski,

R. Granier de Cassagnac,

M. Jo,

S. Lisniak,

P. Miné,

I.N. Naranjo,

M. Nguyen,

C. Ochando,

G. Ortona,

P. Paganini,

P. Pigard,

S. Regnard,

R. Salerno,

Y. Sirois,

T. Strebler,

Y. Yilmaz,

A. Zabi

LaboratoireLeprince-Ringuet,EcolePolytechnique,IN2P3-CNRS,Palaiseau,France

J.-L. Agram

14

,

J. Andrea,

A. Aubin,

D. Bloch,

J.-M. Brom,

M. Buttignol,

E.C. Chabert,

N. Chanon,

C. Collard,

E. Conte

14

,

X. Coubez,

J.-C. Fontaine

14

,

D. Gelé,

U. Goerlach,

C. Goetzmann,

A.-C. Le Bihan,

J.A. Merlin

15

,

K. Skovpen,

P. Van Hove

InstitutPluridisciplinaireHubertCurien,UniversitédeStrasbourg,UniversitédeHauteAlsaceMulhouse,CNRS/IN2P3,Strasbourg,France

S. Gadrat

CentredeCalculdel’InstitutNationaldePhysiqueNucleaireetdePhysiquedesParticules,CNRS/IN2P3,Villeurbanne,France

S. Beauceron,

C. Bernet,

G. Boudoul,

E. Bouvier,

C.A. Carrillo Montoya,

R. Chierici,

D. Contardo,

B. Courbon,

P. Depasse,

H. El Mamouni,

J. Fan,

J. Fay,

S. Gascon,

M. Gouzevitch,

B. Ille,

F. Lagarde,

I.B. Laktineh,

M. Lethuillier,

L. Mirabito,

A.L. Pequegnot,

S. Perries,

A. Popov

16

,

J.D. Ruiz Alvarez,

D. Sabes,

V. Sordini,

M. Vander Donckt,

P. Verdier,

S. Viret

UniversitédeLyon,UniversitéClaudeBernardLyon1,CNRS-IN2P3,InstitutdePhysiqueNucléairedeLyon,Villeurbanne,France

T. Toriashvili

17