Production of inclusive Y(1S) and Y(2S) in p-Pb collisions at √sNN=5.02 TeV

.ALICE Collaboration

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

Articlehistory:

Received9October2014

Receivedinrevisedform19November2014 Accepted19November2014

Availableonline22November2014 Editor:L.Rolandi

WereportontheproductionofinclusiveΥ(1S)andΥ(2S)inp–Pbcollisionsat√_s

NN=⁵.02 TeV atthe LHC.ThemeasurementisperformedwiththeALICEdetectoratbackward(−⁴.46<ycms<−².96)and forward(2.03<ycms<3.53)rapiditydowntozerotransversemomentum.Theproductioncrosssections oftheΥ(1S)and Υ(2S)arepresented,as wellas thenuclearmodificationfactorandthe ratioofthe forwardtobackwardyieldsofΥ(1S).AsuppressionoftheinclusiveΥ(1S)yieldinp–Pbcollisionswith respect tothe yield from ppcollisions scaled bythe number ofbinary nucleon–nucleon collisions is observedatforwardrapiditybutnotatbackwardrapidity.Theresultsarecomparedtotheoreticalmodel calculationsincludingnuclearshadowingorpartonicenergylosseffects.

©^{2014TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense} (http://creativecommons.org/licenses/by/3.0/).FundedbySCOAP³.

1. Introduction

Quarkoniaareboundstatesofaheavyquarkanditsanti-quark.

TheJ/ψ familyiscomprisedofcharmandanti-charmquarksand the Υ family ofbottom and anti-bottom quarks. The former are commonly called charmonia and the latter bottomonia. In ele- mentary pp collisions, the production of a quarkonium can be understood asthe creation of a heavy-quark pair (QQ¯) followed byitsbindingintoa quarkoniumstatewithgivenquantum num- bers[1].Thefirststepis welldescribed byperturbative quantum chromo-dynamics(QCD)whilethesecond stepisinherentlynon- perturbative.Threemainapproachesareusedtodescribequarko- nium production in hadronic collisions: the Colour Evaporation Model (CEM) [2,3], the Colour-Singlet Model (CSM) [4] and the Non-Relativistic QCD (NRQCD) framework [5]. However, none of those models is able to satisfactorily describe simultaneously all aspectsofquarkoniumproductioninppcollisions[6].

In ultra-relativistic Pb–Pb collisions, quarkonia are important probestostudythepropertiesofthedeconfinedstateofpartonic matter, the quark–gluon plasma (QGP). Such a state ispredicted by QCD at high temperature andpressure [7,8]. Since quarkonia areproducedattheearlystage ofthecollision,theyareexpected tointeractwiththeQGPthroughoutitsevolution.Inparticular,in thecolour-screeningscenario[9]quarkoniumstatesaresuppressed intheQGPwithdifferentdissociationprobabilitiesforthevarious massstates,depending ontheir binding energy.The CMSCollab- oration at the Large Hadron Collider (LHC) has reported on the

E-mailaddress:[email protected].

observation ofthe sequential suppression of bottomonium states in Pb–Pb collisions at √

s_NN=².76 TeV [10,11]. However, other hotnuclearmattereffectsbesidescolourscreening,aswellascold nuclear matter (CNM) effects, do complicate thissimple picture.

Ontheonehand,recentmeasurementsbytheALICECollaboration arecompatiblewitharegenerationmechanismplayingan impor- tant role inthe productionofJ/ψ inPb–Pb collisions attheLHC [12–14].Additional J/ψ areexpectedtobe producedfromdecon- finedcharmquarksbykineticrecombinationintheQGP[15,16]or bystatisticalhadronizationatthephaseboundary[17].Thisaddi- tional,hotnuclearmattereffect,competeswiththesuppressionby colourscreening.Duetothelower productioncrosssection ofb_b¯ pairscomparedto cc pairs,¯ theregeneration ofΥ(1S)isexpected tobesmallerthanthatofJ/ψ [18].Ontheotherhand,effectsre- latedto the presence ofCNM can also modifythe production of quarkoniainnucleus–nucleuscollisions.

Cold nuclear matter effects can be separated into initial and final-state effects. Initial-state effects occur prior to the forma- tionoftheheavy-quarkpair.Theseincludethemodificationofthe kinematicaldistributionofthe partonsinthenucleicomparedto thatinfreenucleons[19–22]aswellaspartonenergyloss[23–25].

First,thenuclearPartonDistribution Functions(nPDF) differfrom those in free nucleons (PDF). Since the gluon fusion mechanism dominates the production of heavy-quark pairs in high energy collisions, quarkonium production is particularly sensitive to the gluonnPDF,whichispresently notwell known.Bjorken-x(x_Bj) is defined asthe fraction ofthe hadron momentum carried by the parton. The gluon nPDF includes a shadowing region at low xBj (xBj⁰.01)correspondingtoasuppressionofgluons,anantishad- owingregion atintermediatexBj (0.01^xBj⁰.3)corresponding http://dx.doi.org/10.1016/j.physletb.2014.11.041

0370-2693/©^{2014TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense}(http://creativecommons.org/licenses/by/3.0/).Fundedby SCOAP³.

toanenhancementofgluons,andanadditionalsuppressionofglu- onsknownasEMCeffectathigherxBj(0.35^xBj⁰.7).Secondly, ifthequarkoniumproductionisdominatedbylowxBj gluons,then theColourGlassCondensate(CGC)modelcanbeusedtodescribe the nucleus as a coherent gluonic system that saturates at very large density [26]. Finally, partons can lose energy before creat- ing the heavy-quark pair, thereforemodifying the kinematicdis- tributionsofquarkonia.Final-stateeffectsarethosethataffectthe heavy-quarkpairduringthefinitetimeitneedstoformaquarko- nium state orafterthe state hasbeen formed[27].The QQ¯ pair caninteractwiththenuclearmatterandeventuallybreakup.The break-upcrosssection dependsonthenatureofthepre-resonant stateandisexpectedtobesmallforΥ(1S)athighenergy[27–29].

The final-state resonance can also interact with surrounding co- movers and lose energy or even break up [30–32]. Finally, in a recentapproachtopartonenergyloss[25],itishypothesizedthat the partonenergyloss is coherentand cannot be factorizedinto initialandfinalstateeffects.

Cold nuclear matter effects can be studied in proton–nucleus (p–A) collisions, where the QGP is not expected to be formed.

Charmonium states havebeen extensively measured in p–A col- lisions atvariouscollision energies up toLHC energies.Bottomo- niumproductionhasrecentlybeenstudiedthankstotheincreased energy and luminosity available in collider experiments at RHIC [33,34]andtheLHC [35].Duetothelargermassofthebottomo- niumstatescomparedtothecharmoniumones,themeasurement of Υ production in proton–nucleus collisions allows a study of coldnuclearmattereffectsinadifferentkinematicregime,there- forecomplementingtheJ/ψ studies[36,37].Inaddition,therecent measurement bythe ALICECollaboration inPb–Pb collisions ofa strongerΥ(1S)suppression atforwardrapidity[38] than atmid- rapidityhasstressedtheimportanceofunderstandingCNMeffects onΥ production(sinceinthecolourscreeningscenariosuchabe- haviourisnotexpectedastheenergydensityshouldbelarger or equalatmid-rapiditythanatforwardrapidity).

InthisLetter,wereportALICEresultsoninclusiveΥ production in p–Pb collisions at √

sNN=⁵.02 TeV, measured via the

μ

⁺

μ

⁻

decay channel. The ALICE measurement of the Υ(1S) and Υ(2S) productioncrosssectioninp–PbcollisionsatLHCenergiesispre- sentedatbackward(−⁴.46<ycms<−².96)andforward(2.03<

ycms<3.53)centre-of-massrapidities.Thepositiverapidityisde- finedby the directionoftheprotonbeam. The Υ(1S) production crosssectionsinp–Pbcollisionsarecomparedtothoseinppcol- lisions scaled bythe Pb-nucleusatomic massnumber APb=^208.

Thisnuclearmodificationfactorispresentedasafunctionofrapid- ity.Theratiooftheforwardtobackwardyieldsisalsodiscussed.

2. Experimentalapparatusanddatasample

TheALICEdetectordesignandperformanceareextensivelyde- scribedin[39] and[40].Theanalysispresented hereisbasedon the detection of muons in the ALICE forward muon spectrome- ter,whichcoversthelaboratorypseudorapidityrange−⁴<

η

lab<

−².5.Inaddition,theSiliconPixelDetector(SPD)isusedtorecon- structtheprimaryvertex,theVZEROdetectorprovidesaminimum biastriggerandtheVZEROandTZEROdetectorsare bothusedas luminometers.Ashortdescriptionofthesedetectorsisgiveninthe followingparagraphs.

Themuonspectrometerconsistsofasetofabsorbers,adipole magnet with a 3 T m field integral, five tracking stations and twotriggerstations.Thefront absorber,madeofcarbon,concrete andsteel and placed between0.9 and5 m fromthe Interaction Point (IP), filters out hadrons, thus decreasing the occupancy in thetrackingsystem. Muon trackingisperformedby fivestations, eachoneconsistingoftwoplanesofCathodePadChambers(CPC).

The first two stations are located upstream of the dipole mag- net, the third one is embedded inside the magnet gap and the fourthandfifthareplaceddownstreamofthedipole,justbeforea 1.2 mthick ironwall(7.2interactionlengths),whichabsorbssec- ondary hadronsescaping the front absorber andlow-momentum muons (having p<1.5 GeV/c at the exitof the front absorber).

The muon trigger systemislocated downstream oftheiron wall andconsistsof twostations, eachone equippedwithtwo planes ofResistivePlateChambers(RPC).Thetimeresolutionisoftheor- der of2 ns andtheefficiencyisbetter than95% [41].The muon trigger system delivers single muon anddimuon triggers with a programmable transverse momentum (p_T) threshold.Throughout itsentirelength,aconicalabsorberaroundthebeampipe(θ <2^◦) made oftungsten, lead and steelshields themuon spectrometer againstsecondary particlesproducedbytheinteractionoflarge-

η

primaryparticlesinthebeampipe.

Primary vertexreconstruction isperformedusingtheSPD, the two innermostlayers oftheInner TrackingSystem[42].It covers thepseudo-rapidityranges|

η

lab|<2 and|

η

lab|<1.4,fortheinner andouterlayers,respectively.

ThetwoVZEROhodoscopes[43],with32scintillatortileseach, are placed on each side of the IP, covering the pseudo-rapidity ranges 2.8<

η

lab<5.1 and −³.7<

η

lab<−¹.7. Each hodoscope issegmentedinto8sectorsofequalazimuthal coverageandfour equal pseudo-rapidity rings. The logical AND of the signalsfrom the two hodoscopes forms the Minimum Bias (MB) trigger, also usedasaluminositysignal.Asecond luminositysignalisdefined asthe logical AND ofthe twoTZERO arrays, located onopposite sides of the IP (4.6<

η

lab<4.9 and −³.3<

η

lab<−³.0). Each arrayconsistsof12quartzCherenkovcounters,readbyphotomul- tipliertubes.

Thedatasamplesusedforthisanalysiswerecollectedin2013.

The number of bunches colliding at the ALICE IP ranged from 72 to 288. The peak luminosity during data taking was about 10²⁹ s⁻¹cm⁻². The average number of visible interactions per bunch crossinginsuchconditionsisabout 0.06,corresponding to amultipleinteraction(pile-up)probabilityofabout3%.

Thetriggerconditionusedfordatatakingisadimuon-MBtrig- gerformedbythelogicalANDoftheMBtriggerandanunlike-sign dimuontriggerwithatriggerprobabilityforeachofthetwomuon candidates that increases with pT and is 50% at 0.5 GeV/c. In an additional offlineselection, thetiming information ofthetwo VZERO arrays is used to reject beam-halo andbeam-gas events.

The Zero DegreeCalorimeters (ZDC), positioned symmetricallyat 112.5 m fromthe IP,are usedoffline torejecteventswitha dis- placed vertex,originatingfromtheinteractionsofsatellite proton andleadbunches,asdescribedin[40].

ThetwoLHCbeamshavethesamemagneticrigiditybutdiffer- entprojectilechargetomassratio,whichresultsinthetwobeams havingdifferentenergies: E_p=^{4 TeV and E}Pb/A_Pb=¹.58 TeV. As aconsequence,thecentre-of-masssystemofnucleon–nucleoncol- lisions is shifted in rapidity by y=⁰.465 with respect to the laboratoryframe inthedirectionofthe protonbeam. Intermsof therapidityinthecentre-of-massframe ycms,themuonspectrom- eter acceptance is 2.03<ycms<3.53 when the proton beam is travellinginthedirectionofthespectrometer(p–Pbconfiguration), and−⁴.46<ycms<−².96 inthe opposite case(Pb–pconfigura- tion). Toaccessboth rapidity ranges,data were takeninthe two configurations.

About 9.3×^{106 (2}.1×¹⁰⁷⁾dimuon-MB-triggered eventswere analyzed for the p–Pb (Pb–p) configuration, corresponding to an integratedluminosityLint=⁵.01±⁰.19 nb⁻¹ (5.81±⁰.20 nb⁻¹).

The determination of the integrated luminosities and associated uncertaintiesisdescribedlater.

Fig. 1.Invariantmassdistributionofopposite-signdimuonsintherapidityregions−4.46<ycms<−2.96 (left)and2.03<ycms<3.53 (right)inp–Pbcollisions.Ineach case,thefullcurveshowsthetotalfitfunctionandthedashedcurvesthesignalcomponentforthethreeΥ states(seetextfordetails).

3. Dataanalysis

Muon track candidates are reconstructed in the muon spec- trometer using the standard tracking algorithm [44]. The tracks are required to exit the front absorber at a radial distance from thebeamaxis, Rabs,in therange17.6<Rabs<89.5 cm toreject trackscrossingtheregionoftheabsorberwiththehighestdensity material.Inthisregion,multiplescatteringandenergylosseffects arelargeandcanaffectthemassresolution.Thecontributionfrom fakeandbeam-gasinteractioninducedtracksisreducedbyselect- ingtrackspointingtotheinteractionvertex. Inaddition,tracksin thetrackingsystemarerequestedtomatchatracksegmentinthe triggersystem(triggertracklet).

TheΥ signalisobtainedfromtheinvariant massdistributions of opposite-sign dimuons with a laboratory pair-rapidity in the range 2.5<|^ylab|<4 down to zero transverse momentum. The raw number of Υ is obtainedby fitting the invariant mass dis- tributions.Asumoftwo exponentialfunctionsisusedto param- eterizethe background continuum, andeach Υ resonance shape is described by an extended Crystal Ball(CB) function [45]. The CB function ismade ofa Gaussian core anda power-law tailon each side and is found to reproduce the shape of the Υ peak obtainedinMonte Carlo(MC) simulations. Sincethe CB tails are poorlyconstrainedby thedata,theyare fixedfromtheresultsof the MC simulations. It is also necessary to fix the mass differ- ence between states by using the PDG values [46] and to force the width of the Υ(2S) and Υ(3S) to scale proportionally with theΥ(1S)width accordingto the ratioofthe resonance masses.

MCsimulations validatedthese assumptions.The Υ(1S)signal to backgroundratio (S/B)¹ is between0.8to 1.8, allowing theposi- tion and width of the Υ(1S) peak to be free parameters in the fit.Thesignificance(S/√

S+^B)forΥ(1S)isbetween6.3 and11.6 forthe rapidity bins considered in the analysis. The significance of the Υ(2S) in the rapidity ranges −⁴.46<ycms<−².96 and 2.03<ycms<3.53 islargerthan 3,whichallows areliable mea- surement. However, dueto the limited statistics,the significance oftheΥ(3S) stateis toolowto separate thesignalfromthe un- derlying background. Fig. 1 illustrates the fittingmethod for the rapidity intervals −4.46<ycms<−2.96 (left panel) and 2.03<

ycms<3.53 (rightpanel).ThemeasuredΥ(1S)peakpositionisin agreementwiththeresonancemassvaluefromPDG[46] andthe

1 Thesignaltobackgroundratioandsignificancenumbersarealwaysevaluated determiningthe number ofsignaland backgroundcounts inaninvariant mass rangecentredonthe Υ massand correspondingto±^{3 times thewidthofthe} peak.

measured width(155±^{25 MeV}/c² in−⁴.46<ycms<−².96 and 160±^{22 MeV}/c² in 2.03<ycms<3.53) agrees with the results fromMCsimulations.Asimilaragreementwasobservedforallra- piditybinsconsideredinthisLetter.

Toinvestigatethesystematicuncertaintiesonthesignalextrac- tion procedure, different fits were performed parameterizing the backgroundcontinuum withthe sumoftwo power-lawfunctions andusingalternativeinvariantmassfittingranges.Sincesomepa- rametersare fixedinthefittingprocedure,therelatedsystematic uncertainties werealso studied.TheCB tailparameterswere var- ied according to their spreadobtained by several fits of the MC distributionsindifferentmassranges.ThewidthoftheΥ(2S)and Υ(3S) were varied according to the size of the uncertainties of the Υ(1S) width obtained from the fit. The latter method was similarly used to estimate the systematic uncertainty related to the fixing of the Υ(2S) and Υ(3S) peak position.The raw num- ber of Υ(1S) and Υ(2S) in the rapidity range −⁴.46< ycms<

−².96 are 161±²¹(stat)±⁹(syst) and 42±¹⁴(stat)±⁵(syst), respectively. In the 2.03< ycms<3.53 rapidity range, they are 305±³⁴(stat)±¹³(syst) for Υ(1S) and 83±²³(stat)±¹⁰(syst) forΥ(2S).

The acceptance-times-efficiency ofthe muon spectrometer for themeasurementofΥ, A×

ε

,iscalculatedwithMCsimulations.

The pT and y distributionsofthegenerated Υ(1S)were extrapo- lated,withaprocedureequivalentto theone adoptedfortheJ/ψ [47],to√

sNN=⁵.02 TeV fromexistingppmeasurements [48–50].

Nuclear shadowingcalculations[51] wereusedtoincludetheex- pectedCNM effects.Thesystematicuncertaintywasestimatedby varying the pT and y input distributions by an amount suffi- ciently large (based on theoretical estimations) to include the a priori unknown impact of CNM effects. Since the available data favoura zeroorsmallpolarization ofΥ(1S) [52–54],an unpolar- izedproductionwas assumed. Particle transportisperformedus- ingGEANT3[55]andarealisticdetectorresponseisappliedtothe simulatedhitsinordertoreproducetheperformanceoftheappa- ratusduringdatataking.Thetimedependenceofthetrackingand trigger efficiencies is taken into account by incorporating in the MC simulationsthe deadchannelmaps obtainedfromthe online detectorinformationandthetriggerchamberefficienciesobtained fromarealdataanalysis.Inaddition,arealisticdescriptionofthe residualmisalignmentofthetrackingchambersisincludedinthe simulations. Thetrackingefficiencyis evaluatedwithdata byan- alyzing the clusterdistribution ofthe reconstructedtracks inthe detectionchamberswiththealgorithmdescribedin[44].Thesame algorithmcanbeusedtoestimatethetrackingefficiencyfromMC data. The systematic uncertainties on thisvalue are obtained by

comparingthetrackingefficiencyestimatedfromrealandMCdata.

The efficiency of the muon triggering system is calculated from dataandresultsfromtheanalysisoftriggertrackletdistributions reconstructed fromclustersin thefour planesof thetwo trigger stations.The corresponding systematic uncertainties are obtained by varying the trigger chamber efficiency in MC simulations by an amount equivalent to the statisticaluncertainties on the real dataestimation. The quality ofthe matching ofthe tracking and triggeringsysteminformationisensuredby a

χ

² cut.Inorderto quantify the systematic uncertainties on the matching efficiency, thecut was varied inthesameproportions whileanalyzing both realandMCdata.Theobserveddifferenceinthematchingproba- bilitiesprovidestheuncertainties.

The A×

ε

valuesandthecorresponding systematicuncertain- tiesforΥ (1S)measuredduringthep–PbandthePb–pdatataking periodsare(29.0±².0)% and(20.1±¹.6)%,respectively.Thevalue of A×

ε

is lower for the Pb–p period mainly due to a reduced tracking efficiency. The Υ(2S) A×

ε

and the corresponding sys- tematicuncertainties were evaluated withthe samemethod and thesameinputdistributionsasfortheΥ(1S).Theobserveddiffer- encesbetweentheΥ(2S)andΥ(1S) A×

ε

arelessthan0.5%.The shapevariations betweenthedifferentinputdistributionsusedin thestudyoftheA×

ε

systematicuncertaintieswerelargeenough to cover the differences between the Υ(1S) and Υ(2S) distribu- tionsobservedbyLHCbintherapidityrange2<ycms<4.5 inpp collisions[49,56,57].

TherawnumberofΥ(1S)obtainedwiththefitprocedurede- scribedpreviously,N[Υ (^1S)]^{,iscorrectedforthebranchingratioof} thedimuondecaychannel, BR_{Υ (1S)→}μ⁺μ⁻=⁰.0248±⁰.0005[46]

and for the acceptance-times-efficiency, (A×

ε

)_{Υ (}1S). The Υ(1S) crosssectionisobtainedas

σ

_pPb^{Υ (1S)}

=

^N

[Υ (

^1S

)]/(

^A

× ε )

Υ (1S)

BR_{Υ (1S)→μ}+μ⁻

×

L

,

(1) where the integrated luminosity L =^NMB/

σ

MB is the ratio be- tween the number of MB events and the MB trigger cross sec- tion.Sincetheanalyzed datasampleismadeofdimuontriggered events, it is necessary to use a scaling factor, F, to obtain the numberofMBeventsfromthenumberoftriggeredevents.Thein- verseofthe F factorcorrespondstotheprobability ofhavingthe dimuontriggerconditionverifiedinan MBevent.Itsaveragevalue is F=¹¹²⁹±²(stat)±¹¹(syst) and F =⁵⁸⁹±²(stat)±⁶(syst) forthep–PbandPb–pdatatakingperiods,respectively.Theseval- uesandthe corresponding statisticaluncertainties were obtained by averaging the resultsof two different methods,one basedon theratiooftriggerratesandtheotherbasedontheofflineselec- tionofdimuoneventsintheMBdatasample[36].Thesystematic uncertainties reflect the difference between the results obtained withthetwomethods.TheMBtriggercrosssection

σ

MBwasmea- suredwithavanderMeerscan[58]andfoundtobe2.09±⁰.07 b (2.12±⁰.07 b)forthep–Pb(Pb–p)configuration,wheretheuncer- taintiesforthetwoconfigurationsarepartiallycorrelated[59].The luminosity was also independently determined, ina similar way, by means of the TZERO-based luminosity signal. The two mea- surementsdifferbyatmost1%throughoutthewholedata-taking period.Suchasmallvariationwascombinedquadraticallywiththe NMB and

σ

MB uncertainties, toget a total luminosity uncertainty of3.8%forthep–Pbconfiguration(forwardrapidities)and3.5%for thePb–pconfiguration(backwardrapidities).

4. Results

The Υ(1S) production cross sections in p–Pb collisions at

√s_NN=⁵.02 TeV are:

Table 1

Summaryoftherelativesystematicuncertaintiesoneachquantityenteringinthe calculations oftheresults.TypeI (II)standsfor uncertaintiescorrelated(uncor- related) with rapidity. Type IIuncertainties aregiven as arange including the smallest and the largest values observed inthe bins considered inthis analy- sis.Resultsarepresentedforthebackward(−⁴.46<ycms<−².96)andforward (2.03<ycms<3.53)rapidityregions.

Source Backward rapidity Forward rapidity

Signal extraction:Υ(1S) 5%–6% (II) 4%–6% (II)

Signal extraction:Υ(2S) 12% (II) 12% (II)

Input MC parameterization:Υ(1S) 2%–5% (II) 4%–6% (II) Input MC parameterization:Υ(2S) 5% (II) 5% (II)

Tracking efficiency 6% (II) 4% (II)

Trigger efficiency 2% (II) 2% (II)

Matching efficiency 1% (II) 1% (II)

σpp^{Υ (1S)}(interpolation) 11%–13% (II) 7%–12% (II)

L (correlated) 1.6% (I) 1.6% (I)

L (uncorrelated) 3.1% (II) 3.4% (II)

σ

_pPb^{Υ (1S)}

( −

⁴

.

46

<

ycms

< −

²

.

96

)

=

⁵

.

57

±

⁰

.

72

(

stat

) ±

⁰

.

60

(

syst

)

μb

, σ

_pPb^{Υ (1S)}

(

2

.

03

<

ycms

<

3

.

53

)

=

⁸

.

45

±

⁰

.

94

(

stat

) ±

⁰

.

77

(

syst

)

μb

.

The Υ(2S) production cross sections in p–Pb collisions at

√sNN = ⁵.02 TeV, obtained in a similar way but with BR_{Υ (2S)→}μ⁺μ⁻=⁰.0193±⁰.0017[46],are:

σ

_pPb^{Υ (2S)}

(−

⁴

.

46

<

y_cms

< −

²

.

96

)

=

¹

.

85

±

⁰

.

61

(

stat

) ±

⁰

.

32

(

syst

)

μb

, σ

_pPb^{Υ (2S)}

(

2

.

03

<

y_cms

<

3

.

53

)

=

²

.

97

±

⁰

.

82

(

stat

) ±

⁰

.

50

(

syst

)

μb

.

Asummaryofthedifferentsourcesofsystematicuncertainties and their relative value is givenin Table 1. The uncertainties of typeIIarenotfullyuncorrelatedwithrapidityandnotrivialfactor- ization incorrelated anduncorrelatedpartscan be made. Hence, theyarelabelledasuncorrelated,buttheycannotbequadratically combinedtoobtaintherapidityintegratedresult.

The Υ(1S) candidates were further divided in four rapidity ranges, namely −⁴.46< y_cms<−³.53, −³.53< y_cms<−².96, 2.03<ycms<2.96 and2.96<ycms<3.53.Twoofthemaresym- metricwithrespectto ycms=^{0.Fig. 2showstheinclusive}Υ(1S) differentialcrosssectiond

σ

/dy asafunctionofrapidity.Thever- ticalerrorbarsrepresentthestatisticaluncertaintiesandtheopen boxestheuncorrelatedsystematicuncertainties.Alsoshownisthe inclusiveΥ (1S)y-differentialinterpolated crosssection inpp col- lisions atthe samecentre-of-mass energy(obtained asexplained laterinthetext)scaledby APb.

The CNM effectscan bequantified withthenuclear modifica- tionfactor,

R^{Υ (}_pPb^1S)

= σ

_pPb^{Υ (1S)}

A_Pb

× σ

pp^{Υ (1S)}

,

(2)

where

σ

_pp^{Υ (1S)} isthe Υ(1S)cross section inpp collisionsat √ s= 5.02 TeV.

Since

σ

pp^{Υ (1S)} at √

s=⁵.02 TeV has not yet been measured, it was computed using a data driven √

s interpolation method.

A detailed description of the adopted procedure is given in [60].The LHCb Collaboration hasmeasured the Υ (1S) cross sec- tion in pp collisions at √

s=².76,7 and 8 TeV, over the ranges

Fig. 2.InclusiveΥ(1S)productioncrosssectionasafunctionofrapidityinp–Pb collisionsat√

sNN=⁵.02 TeV.Theverticalerrorbarsrepresentthestatisticalun- certaintiesandtheopenboxestheuncorrelatedsystematicuncertainties.Thecor- relatedsystematicuncertaintyis 1.6% and isdirectly quotedinthe figure.It is obtainedbysumminginquadraturethecorrelateduncertaintyontheintegrated luminosityandthe uncertaintyonthebranching ratioofΥ(1S)todimuon.The bandscorrespondtotheinclusiveΥ(1S)ppcrosssectionobtainedwiththeproce- duredescribedinthetextandscaledbyAPb.

pT<15 GeV/c and 2<y<4.5, in 5rapidity bins of equal size [49,56,57]. The LHCb results were re-binned to obtain the cross section in(approximately)the rapidity ranges ofinterest for this analysis: 2<y<3, 2< y<3.5, 3< y<3.5, 3< y<4.5, and 3.5<y<4.5.Foreach bin,thecrosssection asa functionofen- ergywasfittedaccordingto21differentshapes:15arebasedon LeadingOrderCEM(LO-CEM)calculationsforΥ production,corre- spondingtovariouschoicesofPDFsandofthefactorizationscale;

3arebasedontheenergy-dependenceofbarebottom-quark pair production (FONLL) [61]; the remaining three are a power law, a linear and an exponential function. The obtained fit parame- ters were used to compute the cross section at √

s=⁵.02 TeV.

In order to take into account the rather poor agreement of the data with the fitting functions (

χ

²/ndf >2 for all fits, where ndf is the number of degrees of freedom), all the uncertainties on the fit results were rescaled by

χ

²/ndf. Fits with

χ

²/ndf values larger than three times the minimum value obtained for the rapidity range considered were discarded. The weighted av- erage ofthe surviving results was computed (using the rescaled fit uncertainty as a weight) and retained as central value. The average(rescaled)fit-resultuncertaintywasevaluatedforeachra- piditybin:itrangesfrom7%to12%.Asanadditionaluncertainty, the maximum difference between the average and the individ- ualfit results was computed: it ranges from 2% to 7%. Finally, a thirduncertaintywasconsidered,totakeintoaccounttheshiftof 0.035 rapidity units between the ranges adopted in the interpo- lationprocedure and those usedfor the measurement of R^{Υ (}_pPb^1S). Suchan uncertaintyisquantifiedbythe maximumdifference be- tween the cross sections in the two ranges, evaluated with the LO-CEMandFONLLmodels,andamountsto1%fortheforwardra- pidity region and3% forthe backward rapidity region. Since the interpolationis performedseparately foreach rapidity range,the associateduncertainties are assumedto be uncorrelatedwithra- pidity.Fortheforwardandbackwardrapidity rangesusedforthe integratedresults, the obtainedinterpolated cross-sections times branchingratioare1451±¹¹⁴(syst)pb and770±⁸⁷(syst)pb,re- spectively.

Usingthe interpolatedvalues of

σ

pp^{Υ (1S)},the nuclearmodifica- tionfactorsare

Fig. 3.NuclearmodificationfactorofinclusiveΥ(1S)inp–Pbcollisionsat√_s

NN= 5.02 TeV.TheresultsarecomparedtothoseforinclusiveJ/ψ[36].Theverticaler- rorbarsrepresentthestatisticaluncertaintiesandtheopenboxestheuncorrelated systematicuncertainties(fortheJ/ψ,theuncorrelatedandpartiallycorrelatedun- certaintieshavebeenaddedinquadrature).ThefullboxesaroundRpPb=1 show thesizeofthecorrelateduncertainties,whichinthecaseoftheΥincludeonlythe correlateduncertaintyontheluminosity(seeTable 1).

R^{Υ (}_pPb^1S)

(−

⁴

.

46

<

y_cms

< −

²

.

96

)

=

⁰

.

86

±

⁰

.

11

(

stat

) ±

⁰

.

13

(

uncorr

) ±

⁰

.

01

(

corr

),

R^{Υ (}_pPb^1S)

(

2

.

03

<

y_cms

<

3

.

53

)

=

⁰

.

70

±

⁰

.

08

(

stat

) ±

⁰

.

08

(

uncorr

) ±

⁰

.

01

(

corr

).

Undertheassumptionofa2→1 productionprocess(gg→Υ), the sampledx_Bj rangesare 5.5·¹⁰⁻⁵<x_Bj<2.5·^{10−4 and3}.6· 10⁻²<x_Bj<1.6·^{10−1 atforwardandbackwardrapidity,respec-} tively.Thus, themeasurement atforwardrapidity teststhe shad- owingregionandtheoneatbackwardrapiditytheanti-shadowing region.Inthecaseofa 2→2 productionprocess (gg→Υg) the coveredx_Bj rangesarenaturally expectedtobeenlarged.InFig. 3 the inclusiveΥ(1S)nuclear modificationfactor inp–Pbcollisions at√

s_NN=⁵.02 TeV isshowninfourclassesofrapidity.Theverti- cal errorbars representthestatisticaluncertainties andtheopen boxestheuncorrelatedsystematicuncertainties.Anadditionalcor- related uncertaintyis indicated by the full box around R_pPb=^1.

The R_pPb showsa suppression ofthe inclusive Υ(1S) production yields atforward rapidity in p–Pbcompared to pp collisions. At backwardrapidity,theΥ(1S) RpPbiscompatiblewithunitywithin uncertainties, and thereforedoesnot favoura strong gluon anti- shadowing.AlsoshowninFig. 3istheALICEmeasurementofthe inclusive J/ψ RpPb [36]. Although the uncertainties are large, it appears that at positive y_cms the Υ(1S) andJ/ψ R_pPb are rather similar. It is worth notingthat dueto its larger mass,the Υ(1S) RpPb atforward rapidity ishigher thanthe J/ψ one according to all available model calculations [25,26,28,62]. At negative rapidi- ties, theJ/ψ R_pPb aresystematically abovetheΥ(1S)onebutthe twoRpPbareconsistentwithinuncertainties.Althoughtherapidity rangesarenotidentical,the R_pPbmeasuredbyLHCb[63]arecon- sistent with the ALICEmeasurements within uncertainties, albeit systematicallylarger[60].

The ratio [Υ (2S)/Υ (1S)] ^{of the production cross section of} Υ (2S)→

μ

⁺

μ

⁻toΥ (1S)→

μ

⁺

μ

⁻canbeobtainedas

Υ (

2S

)/Υ (

1S

)

=

^N

[Υ (

2S

)]/(

A

× ε )

Υ (2S) N

[Υ (

^1S

)]/(

^A

× ε )

Υ (1S)

.

(3)

Thebranchingratioofthedimuondecaychannel doesnoten- ter the calculation. Additionally, since the same data sample is used,L ^{cancelsout intheratio.Thesystematic}uncertaintieson

Fig. 4.NuclearmodificationfactorofinclusiveΥ(1S)inp–Pbcollisionsat√

sNN=5.02 TeV asafunctionofrapidity.Theverticalerrorbarsrepresentthestatisticaluncer- taintiesandtheopenboxestheuncorrelatedsystematicuncertainties.ThefullboxesaroundRpPb=1 showthesizeofthecorrelateduncertainties.Alsoshownareseveral modelcalculations:(left)partonenergyloss[25]withandwithoutEPS09shadowingatNLOandCEMwithEPS09shadowingatNLO[62];(right)CGCbased[26]andCSM withEPS09shadowingatLO[28].ForthelattertheeffectofvariationintheshadowingandEMCcurvesishighlightedasdescribedinthetext.(Forinterpretationofthe coloursinthisfigure,thereaderisreferredtothewebversionofthisarticle.)

the ratioswere obtainedby quadratically combiningthe system- aticuncertaintiesenteringineachelementofEq.(3).Nevertheless, sincethedecaykinematicsofthetwo Υ statesareclose, thesys- tematicuncertaintieson tracking,trigger andmatchingefficiency, estimatedforthe samedetectorin thesame workingconditions, canceloutintheratio.Theresultsare:

Υ (

2S

)/Υ (

1S

)

pPb

(−

⁴

.

46

<

y_cms

< −

²

.

96

)

=

⁰

.

26

±

⁰

.

09

(

stat

) ±

⁰

.

04

(

syst

), Υ (

2S

)/Υ (

1S

)

pPb

(

2

.

03

<

ycms

<

3

.

53

)

=

⁰

.

27

±

⁰

.

08

(

stat

) ±

⁰

.

04

(

syst

).

The same ratio has been measured by ALICE in pp collisions at √

s=^{7 TeV in the rapidity range 2}.5< y_cms<4.0 [64] and is 0.26±⁰.08(tot), where the uncertainty is the quadratic sum of the statistical and systematic uncertainties. The LHCb Collab- oration has measured the same ratio in pp collisions at √

s= 2.76,7 and 8 TeV andasafunctionofrapidityintherange2.0<

ycms<4.5 [49,56,57]. The measured [Υ (2S)/Υ (1S)] ^{is found to} be, within uncertainties, independent of √

s and rapidity. For p_T<15 GeV/c (14 GeV/c for 8 TeV) themeasured values inthe range3.0<ycms<3.5 are 0.22±⁰.03(tot),0.24±⁰.02(tot) and 0.25±⁰.01(tot)for√

s=².76,7 and 8 TeV,respectively.Ourmea- sured ratio [Υ (2S)/Υ (1S)] in p–Pbcollisions is compatible with thesameratio inpp collisions.Withinouruncertainties, thereis thereforenoevidenceofadifferentmagnitudeofCNMeffectsfor the Υ(2S) with respect to the Υ (1S). At mid-rapidity, however, theCMSCollaboration has measured thedoubleratio,i.e.the ra- tio[Υ (2S)/Υ (1S)] ^{inp–Pbdividedby that inpp}collisions, tobe 0.83±⁰.05(stat)±⁰.05(syst),suggestingastrongersuppressionof theΥ(2S)thanoftheΥ(1S)inp–Pbcollisions[35].

TheinclusiveΥ(1S) R_pPb integratedoverthebackward orfor- wardrapidity ranges, arecompared toseveralmodel calculations inFig. 4.Intheleft panel,theresultsarecompared toanext-to- leading order (NLO) CEM calculationusing the EPS09parameter- ization ofthe nuclear modification ofthe gluon PDF (commonly referred to asgluon shadowing) at NLO [62] (blue shaded band) andto a partonenergy loss calculation [25] with (green shaded band) or without (red band) EPS09 gluon shadowing at NLO. In thecaseoftheCEM+^EPS09calculation,thebandreflectstheun- certaintiesofthecalculation,dominatedbytheonesoftheEPS09 parameterization[19].Inthecasesofthepartonenergylossmodel

calculations, the bandsrepresentthe uncertaintyfrom theEPS09 parameterizationorfromthepartontransportcoefficient andthe parameterizationusedforthepp referencecrosssection.Noneof the calculationsfullydescribethe backwardandforwardrapidity data and all tend to overestimate the observed Υ(1S) RpPb. The parton energy loss with EPS09calculation reproduces the Υ(1S) R_pPb at forwardrapidity buttendto overestimate itatbackward rapidity.Theopposite trendisfoundifonlypartonenergylossis considered.

Intherightpanel,theresultsarecomparedtoacalculationof a 2→2 production model(gg→Υg) atleading order (LO) us- ing the EPS09 shadowing parameterization also at LO [28]. Two bandsare showntohighlighttheuncertainties linkedto twodif- ferent effects. The extent of the blue band shows the EPS09 LO relateduncertainties in theshadowing region,i.e.atlow xBj. The redbandshowstheuncertaintyintheEMCregion,i.e.athighxBj. As the authors of [28] discuss, the gluon nPDF is poorly known inthisregionandtheΥ(1S)R_pPb atbackwardrapiditycouldadd usefulconstraintstothemodelcalculations.Itisworthnotingthat thetwo bluebandsintheleft andrightpanelsofFig. 4differby their central curve and the extent of the uncertainties. The two approachesare similar andalthoughthe productionmodels used are different,mostofthedifference comesfromtheusage ofthe NLOorLOEPS09gluonshadowingparameterizations.Itcanbear- guedthatusingan NLOparameterizationismoreappropriatethan an LO one,howeveritisworthremarking thatother gluonshad- owing parameterizations [20,21] (also at NLO) are available and that theuncertaintyband oftheEPS09LOparameterizationprac- tically includesthem.Therefore, theblueuncertaintyband inthe rightpanelofFig. 4canbeconsideredasincludingtheuncertainty duetodifferentgluonshadowingparameterizations.Thebackward rapidityΥ(1S)R_pPbdisfavoursthestronggluonanti-shadowingin- cludedintheEPS09parameterization.Intherightpanel ofFig. 4, a calculation based on the CGC framework coupled witha CEM productionmodel isalsoshown(green shadedband)forpositive y_cms.Itisworthnotingthatthiscalculation,althoughonlyslightly underestimating theΥ(1S) R_pPb,isnot abletoreproducetheJ/ψ RpPbinthesamerapidityrange[36].

Thequantity RFBisdefinedastheratioofthenuclearmodifica- tionfactorsatforwardandatbackwardrapiditiesinarangesym- metricwithrespectto ycms=^{0.Itcanbecomputeddirectlyfrom} the ratio of the cross sections (see Eq. (1)) of Υ(1S) at forward and backward rapidities. R_FB is therefore independent of

σ

pp^{Υ (1S)}.

Fig. 5.(Left)ForwardtobackwardratioRFBofinclusiveΥ(1S)yieldscomparedtotheJ/ψ RFB[36].Theverticalerrorbarsrepresentthestatisticaluncertaintiesandthe openboxestheuncorrelatedsystematicuncertainties.(Right)InclusiveΥ(1S)RFBcomparedtotheoreticalmodelcalculations.Thestatisticalandsystematicuncertaintiesfor theexperimentalvalueareaddedinquadrature.Forthecalculations,uncertaintiesarequotedwhenavailable.

The drawback of the RFB ratio is that it can only be measured in the restricted rapidity range 2.96<|^ycms|<3.53, hence los- ing about two thirds of the number of measured Υ. The mea- sured forward to backward ratio is RFB(2.96<|^ycms|<3.53)= 0.95±⁰.24(stat)±⁰.14(syst).Uncertaintiesareobtainedby sum- ming in quadrature the contribution of each individual element enteringtheratio.TheinclusiveΥ(1S) R_FBiscomparedinFig. 5to theinclusiveJ/ψ R_FB [36] inthe samerapidity range(leftpanel) and to several model calculations (right panel). In the rapidity range2.96<|^ycms|<3.53 theΥ(1S)R_FB iscompatiblewithunity andis larger than that of the J/ψ. All models describe the data withinthepresentuncertaintiesofthemeasurement.

5. Conclusion

In summary, we reported the ALICE measurement of Υ pro- duction in p–Pb collisions at √

s_NN=⁵.02 TeV at the LHC. The Υ(1S) production cross section and nuclear modification factor werepresentedinthe rapidityranges −⁴.46<y_cms<−².96 and 2.03<y_cms<3.53 down to zero transverse momentum. At for- ward rapidity, R_pPb shows a suppression of Υ(1S) production in p–Pbcompared to pp collisions. At backward rapidity,the Υ(1S) R_pPbisconsistentwithunity,suggestingthatgluonanti-shadowing is smaller than expected in the EPS09 parameterization. Models includingthe nuclear modificationof thegluon PDF [28,62] ora contributionfromcoherentpartonenergyloss[25]tendtooveres- timateourmeasured RpPbandcannotsimultaneouslydescribethe forwardandbackward rapiditysuppressions. ACGC basedmodel [26] is in agreement withour Υ results atforward rapidity but cannot describe the J/ψ RpPb [36]. The forward to backward ra- tio RFB of the inclusive Υ(1S) yields in 2.96<|^ycms|<3.53 is compatiblewithunity within largeuncertainties. Withinour un- certainties,the [Υ (^2S)/Υ (1S)] ^{ratioshows noevidence ofdiffer-} entCNMeffectsonthetwostates.Additionalmeasurementswith higherstatistics are needed to further constrain the models and extrapolatetheCNMeffectstoPb–Pbcollisions.

Acknowledgements

The ALICECollaboration wouldlike to thank all its engineers andtechniciansfortheirinvaluablecontributionstotheconstruc- tion of the experiment and the CERN accelerator teams for the outstandingperformanceoftheLHCcomplex.

The ALICECollaborationgratefullyacknowledges theresources andsupportprovidedby allGridcentresandtheWorldwide LHC ComputingGrid(WLCG)Collaboration.

The ALICE Collaboration acknowledges the following funding agencies fortheir supportin buildingandrunning theALICEde- tector: StateCommitteeofScience,World FederationofScientists (WFS) and Swiss Fonds Kidagan, Armenia, Conselho Nacional de DesenvolvimentoCientífico eTecnológico (CNPq),Financiadorade EstudoseProjetos(FINEP),FundaçãodeAmparoàPesquisadoEs- tadodeSãoPaulo(FAPESP);NationalNaturalScienceFoundationof China(NSFC),theChineseMinistryofEducation(CMOE)andMin- istryofScience andTechnology ofthePeople’sRepublicofChina (MSTC); Ministry of Education andYouth ofthe Czech Republic;

Danish Natural Science Research Council, the Carlsberg Founda- tion andthe DanishNationalResearch Foundation;TheEuropean ResearchCouncilundertheEuropeanCommunity’sSeventhFrame- workProgramme;HelsinkiInstituteofPhysicsandtheAcademyof Finland;FrenchCNRS-IN2P3,the‘RegionPaysdeLoire’,‘RegionAl- sace’, ‘Region Auvergne’ andCEA, France; German BMBFand the HelmholtzAssociation;GeneralSecretariatforResearchandTech- nology,MinistryofDevelopment,Greece;HungarianOTKAandNa- tionalOfficeforResearch andTechnology (NKTH); Departmentof Atomic Energy, Government of India and Department of Science and Technology,Ministry of Science and Technology of the Gov- ernment ofIndia; IstitutoNazionale diFisicaNucleare(INFN)and CentroFermi–MuseoStoricodellaFisicaeCentroStudieRicerche

“Enrico Fermi”, Italy; MEXT Grant-in-Aid for Specially Promoted Research, Japan; JointInstitute forNuclear Research, Dubna; Na- tionalResearchFoundationofKorea(NRF);CONACYT,DGAPA,Méx- ico,ALFA-ECandtheEPLANETProgram(EuropeanParticlePhysics LatinAmericanNetwork); StichtingvoorFundamenteelOnderzoek der Materie(FOM) andtheNederlandse Organisatievoor Weten- schappelijk Onderzoek (NWO), Netherlands; Research Council of Norway (NFR); Polish Ministry of Science and Higher Education;

NationalScienceCentre,Poland;MinistryofNationalEducation/In- stitute forAtomic PhysicsandCNCS-UEFISCDI, Romania; Ministry ofEducationandScience ofRussianFederation, RussianAcademy ofSciences,RussianFederalAgencyofAtomicEnergy,RussianFed- eralAgency forScience andInnovationsandthe Russian Founda- tionforBasicResearch;MinistryofEducationofSlovakia;Depart- mentofScienceandTechnology,RepublicofSouthAfrica;CIEMAT, EELA,MinisteriodeEconomíayCompetitividad(MINECO)ofSpain, XuntadeGalicia(ConselleríadeEducación),CEADEN,Cubaenergía, Cuba,andIAEA(InternationalAtomicEnergyAgency);SwedishRe- search Council (VR) and Knut and Alice Wallenberg Foundation