Suppression of ϒ(1S)ϒ(1S) at forward rapidity in Pb–Pb collisions at √sNN=2.76 TeV

Contents lists available atScienceDirect

Physics Letters B

www.elsevier.com/locate/physletb

Suppression of Υ ( 1S ) at forward rapidity in Pb–Pb collisions at √

s _NN = ² . 76 TeV

.ALICE Collaboration

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

Articlehistory:

Received3July2014

Receivedinrevisedform18September 2014

Accepted1October2014 Availableonline6October2014 Editor:L.Rolandi

WereportonthemeasurementoftheinclusiveΥ (1S)productioninPb–Pbcollisionsat√_s

NN=².76 TeV carriedoutatforwardrapidity(2.5<y<4)and downtozerotransversemomentumusingitsμ⁺μ⁻

decaychannelwiththeALICEdetectorattheLargeHadronCollider.A strongsuppressionoftheinclusive Υ (1S)yield is observedwithrespect toppcollisions scaledbythe number ofindependentnucleon–

nucleoncollisions.Thenuclearmodificationfactor,foreventsinthe0–90%centralityrange,amountsto 0.30±⁰.05(stat)±⁰.04(syst).TheobservedΥ (1S)suppression tendstoincreasewiththecentralityof thecollisionandseemsmorepronouncedthanincorrespondingmid-rapiditymeasurements.Ourresults arecomparedwithmodelcalculations,whicharefoundtounderestimatethemeasuredsuppressionand failtoreproduceitsrapiditydependence.

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

1. Introduction

At high temperature and high density, Quantum Chromody- namics predicts the existence of a deconfined state of strongly- interacting matter (Quark–Gluon Plasma, QGP) with properties governed by the quark and gluon degrees of freedom [1]. This state can be studied in ultra-relativistic heavy-ion collisions and isexpectedto be produced when thetemperature ofthe system exceedsthe criticaltemperature Tc150–195 MeV [2,3]. Among theparticleswhichcanbemeasuredtoinvestigatetheQGPprop- erties,heavyquarksareofspecialinterestsincetheyareproduced in the initial parton–parton interactions and they interact with the medium throughout its evolution. In particular, the studyof theheavy quark–antiquark boundstate (quarkonium) isexpected to provide essential information on QGP properties. The colour- screeningmodel [4]predicts that charmonia andbottomonia (cc andbb bound states,respectively) dissociate in the medium, re- sultingina suppressionof theobserved yields.More specifically, the quarkonium binding properties are expected to be modified in the deconfined medium and, out of the various charmonium andbottomoniumstates,the lesstightlybound mightmelt close to Tc andthemosttightlybound wellabove Tc [5].A sequential suppressionpatternwithincreasingtemperatureisthenexpected tobe realized.Based on resultsfromquenchedlattice QCD [6,7], the most tightly bound bottomonium state, Υ (1S), is predicted to melt ata temperature larger than 4Tc, while the Υ (2S) and theΥ (3S)shouldmeltat1.6 and1.2Tc,respectively.Themelting

E-mailaddress:[email protected].

temperature fortheJ/ψ charmoniumstateisexpectedtobeclose to that of the Υ (2S) and the Υ (3S) bottomonium states.In the case ofrecent spectral-function approaches withcomplex poten- tial[8,9],theobtaineddissociationtemperaturesarelower.

Inthecharmoniumsector, a significantsuppressionoftheJ/ψ yield hasbeenobservedatSPS [10–12](√

sNN=¹⁷.3 GeV),RHIC [13,14] (√

s_NN =³⁹,62.4,200 GeV) and LHC [15–17] (√ s_NN = 2.76 TeV)energies.A qualitative descriptionofthe resultscanbe obtained assuming that in addition to the dissociationby colour screening,a regenerationprocesstakesplaceforhigh-energycolli- sions.TheregenerationmechanismisparticularlyimportantatLHC energies,wherethemultiplicity ofcharmquarksislarge [18–22].

The ψ(2S) charmoniumstate haslower binding energy than the J/ψ one and cannot be produced by the decays of higher mass states.AtSPSenergies[23],thesuppressionofψ(2S)yieldisabout 2.5 timeslargerthanfortheJ/ψ state.Withthehighcollisionen- ergies and luminosities recently available at RHIC andLHC, it is alsopossibletostudybottomoniumproductioninheavy-ioncolli- sions[24–28].ComparedwiththeJ/ψ case,theprobabilityforthe Υ statestoberegeneratedinthemedium ismuchsmallerdueto thelower productioncrosssection ofbb pairs[29].However, the feed-down fromhigher mass bottomonia(between 40% and 50%

forΥ (1S) [30]) complicatesthe datainterpretation. Furthermore, the suppression dueto the QGPmust be disentangled fromthat dueto ColdNuclear Matter (CNM) effects(such asnuclear mod- ification of the parton distribution functions or break-up of the quarkonium state in CNM)which, asof now, are not accurately known neither at RHIC energies [24] nor in the forward rapid- ityregionsprobedatLHC.AtRHIC,theinclusiveΥ (1S+^2S+^3S) productionhasbeenmeasuredinAu–Aucollisionsatmid-rapidity http://dx.doi.org/10.1016/j.physletb.2014.10.001

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

by the STAR [24] andPHENIX [25] Collaborations. The observed suppression isconsistentwiththemeltingoftheΥ (2S)andΥ (3S) states. At LHC, the CMS Collaboration has measured the mid- rapidityproductionofbottomoniumstatesinPb–Pbcollisions.The Υ (1S) yield issuppressed byapproximately afactor oftwo with respecttotheexpectationfromppcollisionsobtainedbyscalingof thehardprocessyieldwiththenumberofbinarynucleon–nucleon collisions. Moreover, the Υ (2S) and the Υ (3S) are almost com- pletelysuppressed[26,27].

InthisLetter, we report on theinclusiveΥ (1S)production at forwardrapidity (2.5<y<4) anddown to zero transverse mo- mentum (pT>0) in Pb–Pb collisions at √

sNN=².76 TeV. The measurement was carried out in the

μ

⁺

μ

⁻ decaychannel with theALICEdetector.TheyieldofΥ (1S)inPb–Pbcollisionsrelative topp,normalizedtothenumberofnucleon–nucleoncollisions at thesame energy(nuclearmodificationfactor, RAA) isreported in twocentralityintervalsandtworapidity intervals.Theresultsare comparedwithCMSΥ (1S)mid-rapiditydata[27]andwithmodel calculations[31,32].

2. Experimentalapparatusanddatasample

The ALICEdetector isdescribed indetail in reference[33].In this section, we briefly summarize the main features of the de- tectorsusedforthis analysis.The muon spectrometer, coveringa pseudo-rapidityrange−⁴<

η

lab<−².5 inthelaboratoryframe,¹ consists primarily of a tracking apparatus composed of five sta- tionsoftwoplanesofCathodePadChambers(CPC)each,a dipole magnetdeliveringa3T·^{mintegratedmagneticfieldusedtobend} thechargedparticlesinthetrackingsystemareaandatriggering system including four planes of Resistive Plate Chambers (RPC).

The detector incorporates a 10 interaction length front absorber usedtofilterthemuonsupstreamofthetrackingapparatusanda 7.2interactionlength iron wall locatedbetweenthetracking and the triggering systems. The iron wall plays an important role in the muon identification, since it stops the light hadrons escap- ing fromthefront absorber andthe low momentum background muonsproducedmainlyin

π

andKdecays.

TheV0detector[34]consistsoftwoscintillatorarrayscovering the full azimuthand the pseudo-rapidity ranges 2.8<

η

lab<5.1 (V0-A)and−³.7<

η

lab<−¹.7 (V0-C).Bothscintillatorarrayshave an intrinsictime resolution better than 0.5 ns [34,35] and their timing information was used for offline rejection of events pro- ducedbytheinteractionsofthebeamwithresidualgas(orbeam- gasinteractions).

TheZeroDegreeCalorimeters (ZDC),which arelocatedat 114 meterson each sideof theALICEinteraction point,were usedto reduce the beam-halo backgroundby means of an offline timing cut [35]. Another cut on the energy deposited in the ZDC sup- presses the backgroundcontribution fromelectromagnetic Pb–Pb interactions.

Finally,the Silicon PixelDetector (SPD) is used to reconstruct theprimaryvertex.Thisdetectorconsistsoftwocylindricallayers coveringthefullazimuthandthepseudo-rapidityranges|

η

|<2.0 and|

η

_|<1.4 fortheinnerandouterlayer,respectively.

The Minimum-Bias (MB) trigger is definedasthe coincidence ofa signal in thetwo V0 arrays. The efficiencyofsuch a trigger forselecting inelastic Pb–Pb interactions islarger than 95% [36].

Inordertoenrichthedatasamplewithdimuons,thetriggerused in this analysisrequires the detectionof an opposite-sign muon

1 IntheALICEreferenceframe,thepositivez-directionisalongthecounterclock- wisebeamdirection.Thus,themuonspectrometercoversanegativepseudorapidity (ηlab)rangeandanegativeyrange.InthisLettertheresultsarepresentedwitha positiveynotationkeepingtheηlabvaluessigned.

pair in the triggering system incoincidence withthe MB condi- tion. The muon trigger system selects tracks having a transverse momentum, pμ

T, largerthan 1 GeV/c. Thisthresholdisnot sharp andthequoted value correspondsto a50% triggerprobability on amuoncandidate.Eventswereclassifiedaccordingtotheirdegree ofcentrality,whichiscalculatedthroughthestudyoftheV0sig- nal amplitude distribution [37].This analysiswas carried out for the eventscorrespondingto themostcentral 90% oftheinelastic Pb–Pb crosssection. In thiscentralityrange,theefficiencyof the MB trigger forselecting inelastic Pb–Pb interactions is 100% and the contamination from electromagnetic processes is negligible.

The analyzed data sample corresponds to an integratedluminos- ityL_int=⁶⁸.8±⁰.9(stat)⁺₋⁶₅^._.⁰₁(syst) μb⁻¹ [38].

3. Dataanalysis

SeveralstepsarenecessarytoestimatetheΥ (1S)nuclearmod- ification factor.Theyaredescribed inthefollowingsection.Addi- tionaldetailsontheanalysiscanbefoundin[28].

Muon track candidates were reconstructed starting from the hits inthe tracking chambers[39]. Eachreconstructed trackwas then required to match a track segment in the trigger cham- bers(trigger tracklet) andto haveatransverse momentum pμ

T >

2 GeV/c. The latter requirement helps in reducing the contribu- tionofsoftmuonsfrom

π

/K decayswithoutaffectingmuonsfrom Υ (1S) decays. A further selection was applied by requiring the muon tracksto exitthe front absorber at a radial distance from the beamaxis, Rabs,inthe range17.6<Rabs<89.5 cm.Thisse- lectionrejectstracks crossingtheregionoftheabsorberwiththe materialofhighestdensity,wheremultiple-scatteringandenergy- loss effectsare large andaffectthemass resolution.Finally,each trackwasrequiredtopointtotheinteractionvertexinordertore- jectthecontributionsfromfaketracksandbeam-gasinteractions.

Trackswerethencombinedtoformopposite-signmuonpairsand a 2.5<y<4 cuton thepairrapidity was introduced to remove dimuonsattheedgeoftheacceptance.

The raw numberof Υ (1S) was obtained by means ofa fit to the dimuoninvariant mass distributionswiththe combinationof severalfunctions(seeFig. 1).Thebackgroundwasparametrizedas thesumoftwoexponentialfunctionswithallparametersletfree.

Suchfunctionsreproducewellthedataonthelargeinvariantmass rangeofourfits,5–18 GeV/c².MonteCarlosimulationsshowthat each Υ resonanceshapeiswelldescribed byan extendedCrystal Ball(CB) function[40] madeofaGaussian coreandapower-law tail on both sides. The low invariant mass tail is due to non- Gaussian multiplescatteringinthefront absorber,whilethehigh invariant mass one isdueto alignmentandcalibration biases. In thefit,thepositionandthewidthoftheΥ (1S)peakwereleftfree, astheycanbeconstrainedbythedatathemselves.Thepositionof the Υ (2S)and Υ (3S) peakswere fixed tothat ofthe Υ (1S) ac- cordingtothePDG[41] massdifference,whiletheirwidthswere forcedtoscaleproportionallytothatoftheΥ (1S)accordingtothe ratiooftheresonancemasses.Thisscalingwasverified tobeful- filledinMCsimulations.TheCBtailsarepoorlyconstrainedbythe dataandwerefixedusingMCsimulations.Fitswereperformedon the y-integrated,0–90% centralitydistribution,aswell asfortwo centrality intervals, 0–20% (central collisions) and20–90% (semi- peripheral collisions), or two rapidity ranges, 2.5<y<3.2 and 3.2< y<4. The tailparameters depend on rapidity but remain constant withrespecttocentrality.Foreachofthementionedin- tervals,thesignificance(S/√

S+^{B),evaluatedonarangecentered} on the Υ (1S) peak position and ranging between ±^{3 times its} width,islargerthanfiveandthesignal-to-backgroundratiolarger thanone.InthecaseoftheΥ (2S)andΥ (3S),thesignificanceand the signal-to-background ratioare too low toseparate thesignal

Fig. 1.Invariantmassdistributionofopposite-signdimuonswithpT>0 forthedifferentcentralityandrapidityintervalsconsideredintheanalysis(seetextfordetails).The solidbluelinerepresentsthetotalfitfunction(sumoftwoexponentialandthreeextendedCrystalBallfunctions)andthedashedredlineistheΥ (1S)signalcomponent only.Thegreendottedlineandthemagentadashed–dottedlinerepresenttheΥ (2S)andtheΥ (3S)peaks,respectively.

from the underlying background. The Υ (1S) mass, as extracted fromthefit,isconsistentwiththeresonancemassvaluefromthe PDG[41].Depending on the considered rapidity range,its width rangesfrom(107±²⁵)MeV/c²to(159±⁴⁰)MeV/c² andiscon- sistentwiththeresultsfromMCsimulations.

In order to estimate the systematic uncertainties on the sig- nalextraction,thefitswereperformedoverseveralinvariantmass rangesanda sumoftwopower-lawfunctionswasusedasanal- ternativeparametrizationofthebackground.Concerningthereso- nancepeaks,alternativechoicesweremadeforthevaluesofthefit parametersthatwerekeptfixedinthedefaultprocedureoutlined above.First, the widthandthe positionof theΥ (2S) andΥ (3S) werevariedbyanamountcorrespondingtothesizeoftheuncer- taintiesonthe corresponding fitparameters fortheΥ (1S).Then, theCBtailparameters were variedaccordingtothe uncertainties intheirdeterminationfromfitsoftheMCsignaldistributions.For

each source of systematic uncertainty (background parametriza- tion, fixed widths and positions as well astail parameters), the Root MeanSquare (RMS)of thedistribution of signal countsob- tainedwiththedifferentfitswasestimatedandthecorresponding relativeuncertaintiesweresummedinquadrature.

WiththeseprescriptionsthenumberofΥ (1S)countsis134± 20(stat)±^{7(syst) in the rapidity range 2}.5<y<4 and 0–90%

centrality.Dependingoncentralityandrapidity,thesystematicun- certainties rangebetween 5% and 10%. Theyare almost constant withcentralityandreach amaximuminthe3.2<y<4 rapidity interval.

The measured number of Υ (1S) was corrected forthe detec- tor acceptance and efficiency (A×

ε

) estimatedby means of an Embedding Monte Carlo (EMC) method. The MC hits of muons fromΥ (1S)decayswereembeddedintoMBeventsattheraw-data level.Thestandardreconstructionalgorithm[39]wasthenapplied

totheseevents.Thismethodreproducesthedetectorresponseto the signal in a highly realistic background environment and ac- countsforpossiblevariationsofthereconstructionefficiencywith thecollisioncentrality.Thep_Tandydistributionsofthegenerated Υ (1S)were obtainedfromexistingppmeasurements [42–44]us- ing the extrapolationprocedure described in [45]. EKS98nuclear shadowing calculations[46] were used to includean estimate of CNMeffects.Sinceavailabledatafavor asmallornullpolarization for Υ (1S) [47–49], an unpolarized production was assumed (in bothppandPb–Pbcollisions).Thevariationsoftheperformanceof thetrackingandtriggeringsystemsthroughoutthedata-takingpe- riodaswellastheresidualmisalignmentofthetrackingchambers weretakenintoaccountintheEMC.

Fourcontributions enterthe systematicuncertaintyon A×

ε

: (i) theinput Υ (1S) pT and y distributionsforEMC,(ii) thetrack- ing efficiency, (iii) the triggerefficiency and(iv) the matching of trigger tracklets withtracks inthe tracking system. Type (i) un- certaintiescorrespondtothemaximumdifferencebetween A×

ε

evaluated by using the default input parametrizations and those obtainedby usingparametrizations corresponding to pp andPb–

Pbcollisionsatdifferentenergiesandcentralities.Thetrackingand triggerefficienciesdeterminedfromdata[39]andfromMCsimu- lationswerecomparedtoevaluatetype (ii)and (iii)contributions.

For the type (iv) systematicuncertainties, the estimate was per- formedbyvaryingbyasimilaramount,inbothMCandrealdata, thevalue ofthe

χ

² cut ofthe matchingprobability betweenre- constructedtracksinthetrackingsystemandtriggertracklets.The comparisonoftheresultsofthetwoapproachesprovidestheun- certainty.

For Υ (1S) produced in 2.5<y<4 with pT>0, the value of A ×

ε

is 0.226±⁰.025(syst) in semi-peripheral collisions and decreases to 0.216±⁰.024(syst) in central collisions. For the centrality-integrated sample the value of A×

ε

is 0.219± 0.024(syst). Depending on centralityand rapidity, the systematic uncertaintiesrangebetween11% and12%.

TherawnumberofΥ (1S),N[Υ (^1S)]^{,wascorrectedfortheac-} ceptanceandefficiency,(A×

ε

),andforthebranchingratioofthe dimuondecaychannel,BR_{Υ (1S)→}μ⁺μ⁻=⁰.0248±⁰.0005[41].The yield, Y_{Υ (1S)}, was thenobtainedby normalizingthe resulttothe equivalentnumberofMBevents,NMB,via

Y_{Υ (1S)}

=

^N

[Υ (

1S

)]

(

A

× ε ) ×

^BRΥ (1S)→μ⁺μ⁻

×

^NMB

.

(1)

Sincetheanalysisisbasedonadimuontriggersample,theequiv- alent number of MB events was obtained by multiplying the number oftriggered events by an enhancement factor, F,which correspondstotheinverseoftheprobabilityofhavingthedimuon trigger condition verified in an MB event. The F factoraveraged over the data-taking period is F =²⁷.5±¹.0(syst), where the systematic uncertainty reflects the spread of its values observed in the different periods of data taking. Within the rapidity in- terval 2.5<y<4, the Υ (1S) yield is Y_{Υ (1S)}=(5.2±⁰.8(stat)± 0.7(syst))×^{10−5.The valuesofthe yields inthe other centrality} andrapidityrangesconsideredintheanalysisaregiveninTable 1.

Themedium effectson theyields canbe quantifiedby means ofthenuclearmodificationfactor

R_AA

=

^{YΥ (1S)}

^TAA

× σ

_{Υ (}^pp_1S)

^,

⁽²⁾

where^TAA^{istheaveragenuclearoverlapfunction,whichcanbe} interpretedastheaveragenumberofnucleon–nucleonbinarycol- lisions normalizedtothe inelasticnucleon–nucleon crosssection, and

σ

_{Υ (}^pp_1S) is theΥ (1S)productioncross section inpp collisions at√

s=².76 TeV.

Table 1

Yieldsforthedifferentcentralityandrapidityintervalsconsideredintheanalysis.

Statisticaluncertaintiesarereferredtoasstat,uncorrelatedsystematicuncertainties asuncorr andcorrelatedsystematicuncertaintiesascorr.Whenresults areinte- gratedonrapidity(centrality),thedegreeofcorrelationismentionedwithrespect tocentrality(rapidity).

Centrality Rapidity (Yield±stat±uncorr±corr)×10⁵ 0–20% 2.5<y<4 11.3±2.5±0.7±1.3

20–90% 2.5<y<4 3.2±⁰.6±⁰.2±⁰.4 0–90% 2.5<y<3.2 3.2±⁰.6±⁰.4±⁰.1 0–90% 3.2<y<4 1.9±⁰.4±⁰.3±⁰.1

Table 2

Correspondencebetweenthecentralityclass,theaveragenumberofparticipantnu- cleons^Npart^{,theaveragenumberof}participantnucleonsweightedbythenumber ofbinarynucleon–nucleoncollisions^Nwpart^{,andtheaveragenuclearoverlapfunc-} tionTAA.Thevaluesareobtainedasdescribedin[36].

Centrality Npart N^w_part TAA(mb⁻¹)

0–90% 124±^{2 262}±^{4 6}.3±⁰.2

0–20% 308±^{5 323}±^{5 18}.9±⁰.6

20–90% 72±^{3 140}±^{6 2}.7±⁰.1

Thenumberofparticipantnucleons,^Npart^,andtheTAA^cor- respondingto eachcentralityclass usedinthisanalysiswereob- tained froma Glauber model calculation [36]. Table 2showsthe correspondence between the centrality class, ^Npart ^{and T}AA^. Theaveragenumberofparticipantnucleonsweightedbythenum- ber of binary nucleon–nucleon collisions, ^Npart^{w , is also shown.}

The weighted averagewas calculatedforeach centralityclass ac- cording to the values reported in [36] for narrow intervals. The ^Nwpart ^{quantity represents a more precise evaluation of the av-} erage centrality for a giveninterval, since the Υ (1S) production is a hard process andits initial yield scales with thenumber of binary nucleon–nucleon collisions, in the absence of initial-state effects.

Due to the limited number of events collected in pp colli- sions at √

s=².76 TeV, we cannot measure

σ

_{Υ (}^pp_1S). Instead, the LHCb data [50] are used for the R_AA estimate.² LHCb quotes

σ

_{Υ (}^pp_1S)×^BRΥ (1S)→μ⁺μ⁻=⁰.670±⁰.025(stat)±⁰.026(syst) nb in the 2.5< y<4 rapidity range. For the rapidity intervals stud- ied in this analysis(2.5< y<3.2 and 3.2<y<4) there is no exactmatchingwiththerapidityrangesprovidedbyLHCb.There- fore,a rapidityinterpolationwas performedtoprovidethevalues corresponding toourintervals.TheLHCbdata,withthestatistical anduncorrelatedsystematicuncertainties summedinquadrature, were fitted with Gaussian or even-degree polynomial functions.

The functionswerethen integratedovertherequiredrapidity re- gion and, for each range, the Υ (1S) pp cross section result is the average ofthe values obtainedwith thevarious fitting func- tions. The associated uncorrelated systematic uncertainty is ob- tained summinginquadraturethe largestfit uncertaintyandthe halfspreadoftheresultsobtainedwiththedifferentfittingfunc- tions.ThecorrelatedsystematicuncertaintyassociatedtotheLHCb values istakenasa further correlated contributionto theuncer- taintyofourinterpolationresult.Moredetailsontheppreference aregivenin[28].

Therelativesystematicuncertaintiesoneachquantityentering the RAAcalculationarelistedinTable 3.

2 WhenALICE preliminaryresults werereleased,the LHCbdata werenotyet availableandσ_{Υ (}^pp_1S)wasestimatedusingadata-drivenmethodasexplainedin[28].

Dependingontherapidityinterval,theppreferenceobtainedwiththisapproach andtheLHCbdata[50]differby30–35%.Takingintoaccountuncertainties,itim- pliesachangeonthemodificationfactorby1.3 to2.2σ,dependingonrapidity.

Table 3

SummaryoftherelativesystematicuncertaintiesoneachquantityenteringtheRAA calculationforcentralityandrapidityranges.ThetypeI(II)standsforcorrelated (uncorrelated)uncertainties.WhentwovaluesaregivenfortypeIIuncertainties, thefirstvalue isgivenforthe 0–20%(2.5<y<3.2) centrality(rapidity)inter- val,thesecondoneforthe20–90%(3.2<y<4)interval.Thevaluesofsystematic uncertaintiesforthe RAAintegratedover0–90%incentralityand2.5<y<4 in rapidityarequotedinthelastcolumn.

Source Centrality Rapidity Integrated

Signal extraction 5–6% (II) 5–10% (II) 5%

Input EMC distributions 4% (I) 5–7% (II) 4%

Tracking efficiency 10% (I) 9–11% (II) 10%

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

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

^TAA ^{3–4% (II) 3% (I) 3%}

NMB 4% (I) 4% (I) 4%

BR_{Υ (}1S)→μ⁺μ⁻×σ_{Υ (}^pp_1S) 4% (I) 4–7% (II) 4% (I) 4%

Table 4

Valuesofthe RAA measuredinthe centralityand rapidityrangesconsidered in thisanalysis.Statisticaluncertaintiesarereferredtoasstat,uncorrelatedsystem- aticuncertaintiesarereferredtoasuncorrandcorrelatedsystematicuncertainties arereferredtoascorr.

Centrality Rapidity RAA±^stat±^uncorr±^corr

0–20% 2.5<y<4 0.22±⁰.05±⁰.02±⁰.03 20–90% 2.5<y<4 0.44±⁰.09±⁰.03±⁰.05 0–90% 2.5<y<3.2 0.30±⁰.05±⁰.04±⁰.02 0–90% 3.2<y<4 0.29±0.07±0.05±0.02

Fig. 2.InclusiveΥ (1S)RAAasafunctionoftheaveragenumberofparticipantnu- cleons.ALICEdatarefertotherapidityrange2.5<y<4 andareshowntogether withCMS[27]datawhicharereportedin|^y|<2.4.Theverticalbarsrepresentthe statisticaluncertaintiesandtheboxesthe point-to-pointuncorrelated systematic uncertainties.Therelativecorrelateduncertainties(12%forALICEand14%forCMS) areshownasaboxatunity.Thepoint-to-pointhorizontalerrorbarscorrespondto theRMSoftheNpartdistribution.

4. Results

The pT-integratednuclearmodificationfactormeasured inthe rapidityinterval2.5<y<4 is0.30±⁰.05(stat)±⁰.04(syst) forthe 0–90%centrality range andindicates a strong suppressionof the inclusiveΥ (1S) production. The numerical values of the nuclear modificationfactor for the other centrality andrapidity intervals consideredintheanalysisaregiveninTable 4.

InFig. 2,the RAA isshownasa function of^Npart^{. Sinceour} centralityintervalsare large, a horizontalerror barwas assigned point-to-point. It corresponds to the RMS of the N_part distribu- tion[36].Theobservedsuppressiontendstobemorepronounced incentral(0–20%)thaninsemi-peripheral(20–90%)collisions.The CMS[27] datain |^y|<2.4 areshowninthesame figure.Incen- tralcollisions,thesuppressionseems strongeratforwardrapidity

Fig. 3.InclusiveΥ (1S)RAAasafunctionof^Npart^{,comparedwith}calculationsfrom atransport[31](top)andadynamical[32](bottom)model(seetextfordetails).

ThesameconventionsasinFig. 2areusedtoshowtheuncertainties.

thanatmid-rapidity.Insemi-peripheralcollisions,a similar effect mightbepresentwithasmallersignificance.

In Fig. 3, the ALICE results are compared with the calcula- tionsfromatransport[29,31](top)andadynamical[32](bottom) model.The transport model [31] employs a kinetic rate-equation approach in an evolving QGP andincludes both suppression and regeneration effects. In the model [31], CNM effects were calcu- lated by varying an effective absorption cross section between 0 and2 mb,resultinginan uncertaintyband usedtorepresentthe RAA.Thetransportmodelclearlyunderestimatestheobservedsup- pression, even ifthe shape of thecentrality dependence isfairly reproduced. The dynamical model [32] doesnot include CNM or regenerationeffects.Thecalculationofthebottomoniumsuppres- sionisbasedonacomplex-potentialapproachinanevolvingQGP described with a hydrodynamical model. It is assumed that the initial temperatureprofileinrapidity isaboost-invariant plateau, asinferred from the Bjorken picture [51] of heavy-ioncollisions.

Theresultsobtainedwitha Gaussianprofilecorresponding tothe Landaupicture[52] arealsoshown.Three valuesofplasma shear viscositytoentropydensityratio(4

π η

/s)areusedinthecalcula- tions,includingthelimitingcasewhere4

π η

/s=^{1.Themodelcal-} culationsunderestimatethemeasured suppression,independently of the temperature profiles and the model parameter assump- tionsadopted.Theresultcalculatedwith4

π η

/s=^{1 intheBjorken} scenario showsthe largest suppression andfairly reproduces the shapeofthedata.Ithastobenotedthatthecomparisonbetween theRAAvaluesandtheoreticalpredictionsdependsonwhetherthe resultsareshownasafunctionofNpartorN^w_part.Inparticular, if^Nwpart ^{isadopted,the} semi-peripheral R_AA data point isfairly describedbyboththetransportandthedynamicalmodels.

TherapiditydependenceoftheinclusiveΥ (1S) R_AA,integrated overcentrality(0–90%)forp_T>0,ispresentedinFig. 4.TheALICE

Fig. 4.InclusiveΥ (1S)RAAasafunctionofrapiditymeasuredinPb–Pbcollisions at√

sNN=².76 TeV byALICEin2.5<y<4 andCMS[27]in|^y|<2.4,compared withthecalculationsfromatransport[29,31](top)andadynamical[32](bottom) model(seetextfordetails).Openpointsarereflectedwithrespecttothemeasured onesandthesameconventionsasinFig. 2areusedtoshowtheuncertainties.The relativecorrelateduncertaintyontheALICEmeasurementis7% (andisshownasa boxatunity).

resultsare comparedwiththoseofCMS[27] (|^y|<2.4).The ob- servedsuppressionseemsstrongeratforwardthanatmid-rapidity.

Thepredictionsofthe transportmodel[29,31]are alsoshown inFig. 4(top).ThemodelpredictsanearlyconstantRAAasafunc- tionoftherapiditywhichisindisagreementwithCMSandALICE data.InFig. 4 (bottom), thedataare compared withthecalcula- tionsofthedynamicalmodel[32].Allparametersets usedinthe modelcalculationspredictarapiditydependencewhichistheop- positeofthemeasuredone.

Inboth thetransportandthe dynamicalmodels,the inclusive Υ (1S)suppressionislargelyduetothein-mediumdissociationof highermassbottomonia.The evenlargersuppressionobservedin the ALICE data might then point to a significant dissociation of direct Υ (1S). However, to reach a morequantitative assessment, theroleplayedbyCNMeffectsatforwardrapidityshouldbemore accuratelyverifiedandconstrainedbydata.

5. Conclusions

In summary,we havepresented the measurement of the nu- clearmodificationfactorofinclusiveΥ (1S)productionatforward rapidity (2.5<y<4) and down to zero transverse momentum (pT>0) in Pb–Pb collisions at √

sNN =².76 TeV. The observed suppressionof inclusiveΥ (1S) seemsstronger incentral (0–20%) than insemi-peripheral (20–90%)collisions and tends to show a pronouncedrapiditydependenceoverthelargedomaincoveredby ALICE(2.5<y<4)andCMS(|^y|<2.4).TheALICEinclusiveΥ (1S) suppression isunderestimated by the transport model [29,31] as well as by the dynamical model [32] considered in this Letter.

The suppression predictedby the transport model calculationsis approximately constant with rapidity while the measured one is more pronounced atforwardthan atmid-rapidity. Inthe caseof thedynamicalmodel,thecalculatedrapiditytrendistheopposite oftheobserved one.A precisemeasurementof Υ (1S)feed-down fromhigher massbottomonia,aswell asan accurate estimate of CNM effects in the kinematic rangeprobed by ALICEis required in orderto make a morestringentcomparison withmodels. The Υ (1S) production in p–A collisions has recently been measured with theALICEmuon spectrometer [53] andshould help togain furtherinsightonthesizeoftheCNMeffects.

Acknowledgements

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

The ALICE Collaboration acknowledges the following funding agencies fortheir support inbuildingandrunning the ALICEde- tector: StateCommittee ofScience, WorldFederationofScientists (WFS) and Swiss Fonds Kidagan, Armenia, Conselho Nacional de DesenvolvimentoCientífico e Tecnológico(CNPq), Financiadorade EstudoseProjetos(FINEP),FundaçãodeAmparoàPesquisadoEs- tado de São Paulo (FAPESP); NationalNaturalScience Foundation of China (NSFC), the Chinese Ministry of Education (CMOE) and theMinistryofScienceandTechnologyofChina(MSTC);Ministry ofEducationandYouthoftheCzechRepublic;DanishNaturalSci- ence Research Council, the Carlsberg Foundation and the Danish NationalResearchFoundation;TheEuropeanResearchCouncilun- der the European Community’s Seventh Framework Programme;

Helsinki Institute ofPhysics andthe Academy of Finland; French CNRS-IN2P3,the‘RegionPaysdeLoire’,‘RegionAlsace’,‘RegionAu- vergne’andCEA,France;GermanBMBFandtheHelmholtzAssoci- ation;GeneralSecretariatforResearchandTechnology,Ministryof Development,Greece;HungarianOTKAandNationalOfficeforRe- searchandTechnology(NKTH);DepartmentofAtomicEnergyand Department ofScience andTechnology ofthe Government ofIn- dia;IstitutoNazionalediFisicaNucleare(INFN)andCentroFermi – MuseoStoricodellaFisicaeCentroStudieRicerche“EnricoFermi”, Italy; MEXT Grant-in-Aid forSpecially Promoted Research,Japan;

Joint Institute for Nuclear Research, Dubna; National Research Foundation of Korea (NRF); CONACYT, DGAPA, México, ALFA-EC andtheEPLANETProgram(EuropeanParticle PhysicsLatinAmeri- canNetwork); StichtingvoorFundamenteelOnderzoekderMaterie (FOM)andtheNederlandseOrganisatievoorWetenschappelijkOn- derzoek (NWO), Netherlands; Research Council of Norway (NFR);

PolishMinistryofScienceandHigherEducation;NationalAuthor- ity for Scientific Research – NASR (Autoritatea Na ¸tional˘a pentru Cercetare ¸Stiin ¸tific˘a – ANCS); Ministry of Education and Science of the Russian Federation, Russian Academy of Sciences, Russian Federal AgencyofAtomic Energy,Russian FederalAgencyforSci- ence and Innovations and the Russian Foundation for Basic Re- search; Ministryof Educationof Slovakia; Departmentof Science andTechnology,RepublicofSouthAfrica;CIEMAT,EELA,Ministerio deEconomíayCompetitividad(MINECO)ofSpain,XuntadeGalicia (Consellería de Educación), CEADEN,Cubaenergía,Cuba, andIAEA (International Atomic Energy Agency); Swedish Research Council (VR) andKnut andAlice Wallenberg Foundation (KAW); Ukraine Ministry of Education and Science; United Kingdom Science and Technology Facilities Council (STFC); The U.S. Department of En- ergy, the United StatesNational Science Foundation, the State of Texas,andtheStateofOhio.