Suppression of Upsilon production in d plus Au and Au plus Au collisions at root S-NN=200 GeV

UNIVERSIDADE ESTADUAL DE CAMPINAS

SISTEMA DE BIBLIOTECAS DA UNICAMP

REPOSITÓRIO DA PRODUÇÃO CIENTIFICA E INTELECTUAL DA UNICAMP

Versão do arquivo anexado / Version of attached file:

Versão do Editor / Published Version

Mais informações no site da editora / Further information on publisher's website:

https://www.sciencedirect.com/science/article/pii/S0370269314004316

DOI: 10.1016/j.physletb.2014.06.028

Direitos autorais / Publisher's copyright statement:

©2014

by Elsevier. All rights reserved.

DIRETORIA DE TRATAMENTO DA INFORMAÇÃO

Cidade Universitária Zeferino Vaz Barão Geraldo

CEP 13083-970 – Campinas SP

Fone: (19) 3521-6493

http://www.repositorio.unicamp.br

Contents lists available atScienceDirect

Physics

Letters

B

www.elsevier.com/locate/physletb

Suppression

of

Υ

production

in

d

+

Au and

Au

+

Au collisions

at

√

s

_{N N}

=

200 GeV

STAR

Collaboration

L. Adamczyk

a

,

J.K. Adkins

w

,

G. Agakishiev

u

,

M.M. Aggarwal

ai

,

Z. Ahammed

bb

,

I. Alekseev

s

,

J. Alford

v

,

C.D. Anson

af

,

A. Aparin

u

,

D. Arkhipkin

d

,

E.C. Aschenauer

d

,

G.S. Averichev

u

,

J. Balewski

aa

,

A. Banerjee

bb

,

Z. Barnovska

n

,

D.R. Beavis

d

,

R. Bellwied

ax

,

A. Bhasin

t

,

A.K. Bhati

ai

,

P. Bhattarai

aw

,

H. Bichsel

bd

,

J. Bielcik

m

,

J. Bielcikova

n

,

L.C. Bland

d

,

I.G. Bordyuzhin

s

,

W. Borowski

at

,

J. Bouchet

v

,

A.V. Brandin

ad

,

S.G. Brovko

f

,

S. Bültmann

ag

,

I. Bunzarov

u

,

T.P. Burton

d

,

J. Butterworth

ao

,

H. Caines

be

,

M. Calderón de la Barca Sánchez

f

,

∗

,

D. Cebra

f

,

R. Cendejas

aj

,

M.C. Cervantes

av

,

P. Chaloupka

m

,

Z. Chang

av

,

S. Chattopadhyay

bb

,

H.F. Chen

aq

,

J.H. Chen

as

,

L. Chen

i

,

J. Cheng

ay

,

M. Cherney

l

,

A. Chikanian

be

,

W. Christie

d

,

J. Chwastowski

k

,

M.J.M. Codrington

aw

,

R. Corliss

aa

,

J.G. Cramer

bd

,

H.J. Crawford

e

,

X. Cui

aq

,

S. Das

p

,

A. Davila Leyva

aw

,

L.C. De Silva

ax

,

R.R. Debbe

d

,

T.G. Dedovich

u

,

J. Deng

ar

,

A.A. Derevschikov

ak

,

R. Derradi de Souza

h

,

S. Dhamija

r

,

B. di Ruzza

d

,

L. Didenko

d

,

C. Dilks

aj

,

F. Ding

f

,

P. Djawotho

av

,

X. Dong

z

,

J.L. Drachenberg

ba

,

J.E. Draper

f

,

C.M. Du

y

,

L.E. Dunkelberger

g

,

J.C. Dunlop

d

,

L.G. Efimov

u

,

J. Engelage

e

,

K.S. Engle

az

,

G. Eppley

ao

,

L. Eun

z

,

O. Evdokimov

j

,

R. Fatemi

w

,

S. Fazio

d

,

J. Fedorisin

u

,

P. Filip

u

,

E. Finch

be

,

Y. Fisyak

d

,

C.E. Flores

f

,

C.A. Gagliardi

av

,

D.R. Gangadharan

af

,

D. Garand

al

,

F. Geurts

ao

,

A. Gibson

ba

,

M. Girard

bc

,

S. Gliske

b

,

D. Grosnick

ba

,

Y. Guo

aq

,

A. Gupta

t

,

S. Gupta

t

,

W. Guryn

d

,

B. Haag

f

,

O. Hajkova

m

,

A. Hamed

av

,

L.-X. Han

as

,

R. Haque

ae

,

J.W. Harris

be

,

J.P. Hays-Wehle

aa

,

S. Heppelmann

aj

,

K. Hill

f

,

A. Hirsch

al

,

G.W. Hoffmann

aw

,

D.J. Hofman

j

,

S. Horvat

be

,

B. Huang

d

,

H.Z. Huang

g

,

P. Huck

i

,

T.J. Humanic

af

,

G. Igo

g

,

W.W. Jacobs

r

,

H. Jang

x

,

E.G. Judd

e

,

S. Kabana

at

,

D. Kalinkin

s

,

K. Kang

ay

,

K. Kauder

j

,

H.W. Ke

d

,

D. Keane

v

,

A. Kechechyan

u

,

A. Kesich

f

,

Z.H. Khan

j

,

D.P. Kikola

al

,

I. Kisel

o

,

A. Kisiel

bc

,

D.D. Koetke

ba

,

T. Kollegger

o

,

J. Konzer

al

,

I. Koralt

ag

,

W. Korsch

w

,

L. Kotchenda

ad

,

P. Kravtsov

ad

,

K. Krueger

b

,

I. Kulakov

o

,

L. Kumar

ae

,

R.A. Kycia

k

,

M.A.C. Lamont

d

,

J.M. Landgraf

d

,

K.D. Landry

g

,

J. Lauret

d

,

A. Lebedev

d

,

R. Lednicky

u

,

J.H. Lee

d

,

W. Leight

aa

,

M.J. LeVine

d

,

C. Li

aq

,

W. Li

as

,

X. Li

al

,

X. Li

au

,

Y. Li

ay

,

Z.M. Li

i

,

L.M. Lima

ap

,

M.A. Lisa

af

,

F. Liu

i

,

T. Ljubicic

d

,

W.J. Llope

ao

,

R.S. Longacre

d

,

X. Luo

i

,

G.L. Ma

as

,

Y.G. Ma

as

,

D.M.M.D. Madagodagettige Don

l

,

D.P. Mahapatra

p

,

R. Majka

be

,

S. Margetis

v

,

C. Markert

aw

,

H. Masui

z

,

H.S. Matis

z

,

D. McDonald

ao

,

T.S. McShane

l

,

N.G. Minaev

ak

,

S. Mioduszewski

av

,

B. Mohanty

ae

,

M.M. Mondal

av

,

D.A. Morozov

ak

,

M.G. Munhoz

ap

,

M.K. Mustafa

z

,

B.K. Nandi

q

,

Md. Nasim

ae

,

T.K. Nayak

bb

,

J.M. Nelson

c

,

L.V. Nogach

ak

,

S.Y. Noh

x

,

J. Novak

ac

,

S.B. Nurushev

ak

,

G. Odyniec

z

,

A. Ogawa

d

,

K. Oh

am

,

A. Ohlson

be

,

V. Okorokov

ad

,

E.W. Oldag

aw

,

R.A.N. Oliveira

ap

,

M. Pachr

m

,

B.S. Page

r

,

S.K. Pal

bb

,

Y.X. Pan

g

,

Y. Pandit

j

,

Y. Panebratsev

u

,

T. Pawlak

bc

,

B. Pawlik

ah

,

H. Pei

i

,

C. Perkins

e

,

W. Peryt

bc

,

A. Peterson

f

,

P. Pile

d

,

M. Planinic

bf

,

J. Pluta

bc

,

D. Plyku

ag

,

N. Poljak

bf

,

J. Porter

z

,

A.M. Poskanzer

z

,

N.K. Pruthi

ai

,

M. Przybycien

a

,

P.R. Pujahari

q

,

H. Qiu

z

,

http://dx.doi.org/10.1016/j.physletb.2014.06.028

0370-2693/©2014TheAuthors.PublishedbyElsevierB.V.ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/licenses/by/3.0/).Fundedby SCOAP3_.

A. Quintero

v

,

S. Ramachandran

w

,

R. Raniwala

an

,

S. Raniwala

an

,

R.L. Ray

aw

, C.K. Riley

be

,

H.G. Ritter

z

,

J.B. Roberts

ao

,

O.V. Rogachevskiy

u

,

J.L. Romero

f

,

J.F. Ross

l

,

A. Roy

bb

,

L. Ruan

d

,

J. Rusnak

n

,

N.R. Sahoo

bb

,

P.K. Sahu

p

,

I. Sakrejda

z

,

S. Salur

z

,

A. Sandacz

bc

,

J. Sandweiss

be

,

E. Sangaline

f

,

A. Sarkar

q

,

J. Schambach

aw

,

R.P. Scharenberg

al

,

A.M. Schmah

z

,

W.B. Schmidke

d

,

N. Schmitz

ab

,

J. Seger

l

,

P. Seyboth

ab

,

N. Shah

g

,

E. Shahaliev

u

,

P.V. Shanmuganathan

v

,

M. Shao

aq

,

B. Sharma

ai

,

W.Q. Shen

as

,

S.S. Shi

z

,

Q.Y. Shou

as

,

E.P. Sichtermann

z

,

R.N. Singaraju

bb

,

M.J. Skoby

r

,

D. Smirnov

d

,

N. Smirnov

be

,

D. Solanki

an

,

P. Sorensen

d

,

U.G. deSouza

ap

,

H.M. Spinka

b

,

B. Srivastava

al

,

T.D.S. Stanislaus

ba

,

J.R. Stevens

aa

,

R. Stock

o

,

M. Strikhanov

ad

,

B. Stringfellow

al

,

A.A.P. Suaide

ap

,

M. Sumbera

n

,

X. Sun

z

,

X.M. Sun

z

,

Y. Sun

aq

,

Z. Sun

y

,

B. Surrow

au

,

D.N. Svirida

s

,

T.J.M. Symons

z

,

A. Szanto de Toledo

ap

,

J. Takahashi

h

,

A.H. Tang

d

,

Z. Tang

aq

,

T. Tarnowsky

ac

,

J.H. Thomas

z

,

A.R. Timmins

ax

,

D. Tlusty

n

,

M. Tokarev

u

,

S. Trentalange

g

,

R.E. Tribble

av

,

P. Tribedy

bb

,

B.A. Trzeciak

bc

,

O.D. Tsai

g

,

J. Turnau

ah

,

T. Ullrich

d

,

D.G. Underwood

b

,

G. Van Buren

d

,

G. van Nieuwenhuizen

aa

,

J.A. Vanfossen Jr.

v

,

R. Varma

q

,

G.M.S. Vasconcelos

h

,

A.N. Vasiliev

ak

,

R. Vertesi

n

,

F. Videbæk

d

,

Y.P. Viyogi

bb

,

S. Vokal

u

,

A. Vossen

r

,

M. Wada

aw

,

M. Walker

aa

,

F. Wang

al

,

G. Wang

g

,

H. Wang

d

,

J.S. Wang

y

,

X.L. Wang

aq

,

Y. Wang

ay

,

Y. Wang

j

,

G. Webb

w

,

J.C. Webb

d

,

G.D. Westfall

ac

,

H. Wieman

z

,

G. Wimsatt

f

,

S.W. Wissink

r

,

R. Witt

az

,

Y.F. Wu

i

,

Z. Xiao

ay

,

W. Xie

al

,

K. Xin

ao

,

H. Xu

y

,

N. Xu

z

,

Q.H. Xu

ar

,

Y. Xu

aq

,

Z. Xu

d

,

W. Yan

ay

,

C. Yang

aq

,

Y. Yang

y

,

Y. Yang

i

,

Z. Ye

j

,

P. Yepes

ao

,

L. Yi

al

,

K. Yip

d

,

I.-K. Yoo

am

,

Y. Zawisza

aq

,

H. Zbroszczyk

bc

,

W. Zha

aq

,

J.B. Zhang

i

,

J.L. Zhang

ar

,

S. Zhang

as

,

X.P. Zhang

ay

,

Y. Zhang

aq

,

Z.P. Zhang

aq

,

F. Zhao

g

,

J. Zhao

as

,

C. Zhong

as

,

X. Zhu

ay

,

Y.H. Zhu

as

,

Y. Zoulkarneeva

u

,

M. Zyzak

o

a_{AGHUniversityofScienceandTechnology,Cracow,Poland} b_{ArgonneNationalLaboratory,Argonne,IL 60439,USA} c_{UniversityofBirmingham,Birmingham,UnitedKingdom} d_{BrookhavenNationalLaboratory,Upton,NY 11973,USA} e_{UniversityofCalifornia,Berkeley,CA 94720,USA} f_{UniversityofCalifornia,Davis,CA 95616,USA} g_{UniversityofCalifornia,LosAngeles,CA 90095,USA} h_{UniversidadeEstadualdeCampinas,SaoPaulo,Brazil} i_{CentralChinaNormalUniversity(HZNU),Wuhan430079,China} j_{UniversityofIllinoisatChicago,Chicago,IL 60607,USA} k_{CracowUniversityofTechnology,Cracow,Poland} l_{CreightonUniversity,Omaha,NE 68178,USA}

m_{CzechTechnicalUniversityinPrague,FNSPE,Prague,11519,CzechRepublic} n_{NuclearPhysicsInstituteASCR,25068ˇRež/Prague,CzechRepublic} o_{FrankfurtInstituteforAdvancedStudiesFIAS,Germany} p_{InstituteofPhysics,Bhubaneswar751005,India} q_{IndianInstituteofTechnology,Mumbai,India} r_{IndianaUniversity,Bloomington,IN 47408,USA}

s_{AlikhanovInstituteforTheoreticalandExperimentalPhysics,Moscow,Russia} t_{UniversityofJammu,Jammu180001,India}

u_{JointInstituteforNuclearResearch,Dubna,141980,Russia} v_{KentStateUniversity,Kent,OH 44242,USA}

w_{UniversityofKentucky,Lexington,KY,40506-0055,USA}

x_{KoreaInstituteofScienceandTechnologyInformation,Daejeon,RepublicofKorea} y

InstituteofModernPhysics,Lanzhou,China

z_{LawrenceBerkeleyNationalLaboratory,Berkeley,CA 94720,USA} aa_{MassachusettsInstituteofTechnology,Cambridge,MA02139-4307,USA} ab_{Max-Planck-InstitutfürPhysik,Munich,Germany}

ac_{MichiganStateUniversity,EastLansing,MI 48824,USA} ad_{MoscowEngineeringPhysicsInstitute,Moscow, Russia}

ae_{NationalInstituteofScienceEducationandResearch,Bhubaneswar751005,India} af_{OhioStateUniversity,Columbus,OH 43210,USA}

ag_{OldDominionUniversity,Norfolk,VA,23529,USA} ah_{InstituteofNuclearPhysicsPAN,Cracow,Poland} ai_{PanjabUniversity,Chandigarh160014,India}

aj_{PennsylvaniaStateUniversity,UniversityPark,PA 16802,USA} ak_{InstituteofHighEnergyPhysics,Protvino,Russia}

al_{PurdueUniversity,WestLafayette,IN 47907,USA} am_{PusanNationalUniversity,Pusan,RepublicofKorea} an_{UniversityofRajasthan,Jaipur302004,India} ao_{RiceUniversity,Houston,TX 77251,USA} ap_{UniversidadedeSaoPaulo,SaoPaulo,Brazil}

aq_{UniversityofScience&TechnologyofChina,Hefei230026,China} ar_{ShandongUniversity,Jinan,Shandong250100,China}

at_{SUBATECH,Nantes,France}

au_{TempleUniversity,Philadelphia,PA,19122,USA} av_{TexasA&MUniversity,CollegeStation,TX 77843,USA} aw_{UniversityofTexas,Austin,TX 78712,USA} ax_{UniversityofHouston,Houston,TX,77204,USA} ay_{TsinghuaUniversity,Beijing100084,China}

az_{UnitedStatesNavalAcademy,Annapolis,MD21402,USA} ba_{ValparaisoUniversity,Valparaiso,IN 46383,USA} bb_{VariableEnergyCyclotronCentre,Kolkata700064,India} bc_{WarsawUniversityofTechnology,Warsaw,Poland} bd_{UniversityofWashington,Seattle,WA 98195,USA} be_{YaleUniversity,NewHaven,CT 06520,USA} bf_{UniversityofZagreb,Zagreb,HR-10002,Croatia}

a

r

t

i

c

l

e

i

n

f

o

a

b

s

t

r

a

c

t

Articlehistory:

Received13December2013 Receivedinrevisedform23April2014 Accepted11June2014

Availableonline16June2014 Editor:V.Metag

Keywords:

Upsilonsuppression Quarkoniumin-medium Relativisticheavy-ioncollisions STAR

WereportmeasurementsofΥ mesonproductioninp+p,d+Au,andAu+Au collisionsusingtheSTAR detectoratRHIC.WecomparetheΥ yieldtothemeasuredcrosssectioninp₊p collisionsinorderto quantifyanymodificationsoftheyieldincoldnuclearmatterusingd+Au dataandinhotnuclearmatter usingAu+Au dataseparatedintothreecentralityclasses.Ourp+p measurementisbasedonthreetimes thestatisticsofourpreviousresult.WeobtainanuclearmodificationfactorforΥ (1S₊2S₊3S)inthe rapidityrange|y|<1 ind+Au collisionsofRdAu=0.79±0.24(stat.)±0.03(syst.)±0.10(p+p syst.). A comparison with models including shadowing and initial state parton energy loss indicates the presenceofadditionalcold-nuclearmattersuppression.Similarly,inthetop 10%most-centralAu₊Au collisions,wemeasureanuclearmodificationfactorofRA A=0.49±0.1(stat.)±0.02(syst.)±0.06(p+p syst.), which is a larger suppression factor than that seen in cold nuclear matter. Our results are consistentwith completesuppression ofexcited-stateΥ mesonsinAu+Au collisions.Theadditional suppression inAu+Au is consistentwith the level expectedin model calculations that include the presence of ahot, deconfined Quark–Gluon Plasma.However, understanding the suppression seenin

d+Au isstillneededbeforeanydefinitivestatementsabout thenatureofthesuppression inAu+Au canbemade.

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

1. Introduction

InthestudyofthepropertiesoftheQuark–GluonPlasma(QGP) an extensive effort has been devoted to measuring quarkonium yields since these have been predicted to be sensitive to color deconfinement [1]. Studies havemainly focused on charmonium, but with the high collision energies available at the Relativistic HeavyIonCollider(RHIC)andtheLargeHadronCollider(LHC)we cannowstudybottomoniuminhotnuclearmatterwithsufficient statistics.Fora recentreview of quarkonium in-medium,see e.g. Ref.[2,Section 5].Onepredictionisthatexcitedquarkoniumstates are expectedto dissociate ator above temperatures nearthat of the crossover to the deconfined QGP phase, Tc

≈

150–190 MeV

[3–5].Themoretightlyboundgroundstatesareexpectedto disso-ciateatevenhighertemperatures.The detailsofthe temperature dependence of the dissociation of the excited states and of the feed-downpatternoftheexcitedstatesintothegroundstatelead toasequentialsuppressionpatternoftheinclusiveupsilonstates withincreasingtemperature[6].Thebinding energyofthe

Υ (

2S

)

state(

∼

540 MeV)isabouthalfthatofthe

Υ (

1S

)

state(

∼

1

.

1 GeV); the

Υ (

3S

)

isstillmoreweaklyboundat

∼

200 MeV.Recentstudies takeintoaccountnotonlytheDebyescreeningeffectontheheavy quarkpotential butalsoan imaginarypartofthe potentialwhich modifiesthewidthsofthevariousquarkoniastates(e.g.[7–10]).In Ref.[8]itisestimatedthatthe

Υ (

2S

)

statewillmeltata temper-atureofT

≈

250 MeV,whereas thegroundstate

Υ (

1S

)

willmelt attemperaturesnearT

≈

450 MeV.

We focus here on the measurement of bottomonium mesons in collisions at

√

sN N

=

200 GeV. An observation of suppression

in the bottomonium sector in hot nuclear matter is important

*

Correspondingauthor.

for two reasons. First, it would be evidence for color deconfine-mentintheproducedmattersincetheaforementionedeffectsare all ultimatelybased on studiesof thehightemperature phase of QuantumChromodynamics(QCD)doneonthelattice,wherecolor isan activedegreeoffreedom.Second,bottomoniumsuppression providesawaytoestimatemodel-dependentboundsonthe tem-peraturewiththeboundsdependingontheparticularsuppression patternseen.

The crosssection forbottomoniumproduction issmallerthan thatofcharmonium[11–13]makingtheexperimentalstudyof

Υ

productionchallenging. However, thetheoretical interpretation of bottomoniumsuppressionislesscomplicatedthanthatof charmo-nium forseveral reasons.While charmonium productionatRHIC and higher energies can be affected by the statistical recombi-nation of charm quarks that are produced in different nucleon– nucleoncollisions within thesamenuclear interactionevent, this effect is negligible for bottomonium due to the much smaller bb production

¯

cross section (

σ

_bb¯ is measured to be in the range 1.34–1.84 μb [14] and calculated to be 1

.

87₋+₀0._.99₆₇ μb [15], com-paredto

σ

c¯c

≈

550–1400 μb[16,17]).Anothercomplicationinthe

charmoniumcaseisthateveninapurelyhadronicscenario, char-moniummesons can besuppressed duetotheir interaction with hadronic co-movers [18,19]. The cross section for inelastic colli-sionsof

Υ (

1S

)

withhadronsissmall[20].Hence,absorptioninthe medium by the abundantly produced co-moving hadronsis pre-dictedtobeminimal.Thecold-nuclear-matter(CNM)effectson

Υ

production,which are those seenin p

+

A collisions andcan be dueto shadowingofthe partondistributionfunctionsin the nu-cleus orenergy-loss in thenucleus, can still be important.There isevidenceofsome

Υ

suppressioninfixedtarget experimentsat 800 GeV

/

c labmomentumfromE772[21].However,theCNM sup-pressionobservedfor

Υ

issmallerthanthatfor J

/ψ

reportedby

NA50 [22]. Forall thesereasons, the

Υ

familyis expectedto be a cleaner and more direct probe of hot QCD, and of the corre-spondingcolordeconfinementeffects.

In this letter we present measurements of

Υ

production in p

+

p,d

+

Au,andAu

+

Au collisions at

√

sN N

=

200 GeV viathe

e+e−decaychannelobtainedbytheSTARexperiment.Weextract invariantcrosssectionsforallthreecollisionsystemsstudied.

Us-ing this p

+

p measurementasa baseline we obtain thenuclear

modificationfactor(RdAu and RA A) ofthethreestatescombined:

Υ (

1S

+

2S

+

3S

)

.The ratio RA A isused to quantifydeviations of

theyieldsind

+

Au andAu

+

Au comparedtothoseexpectedfrom a superpositionofelementaryp

+

p collisions.Thedataweretaken during 2008(d

+

Au),2009(p

+

p) and2010 (Au

+

Au)atRHIC, andcorrespondtointegratedluminositiesof28

.

2 nb−1,20

.

0 pb−1, and1

.

08 nb−1,respectively.Allthreedatasetsweretakenwiththe samedetectorconfiguration.Forthisreasonthedatafromour pre-vious p

+

p result (2006) was not included in thisanalysis; the amountofmaterialinthedetectoratthat pointwassubstantially largerthan itwas in thethreedatasets discussedhere. We com-pareourdatatomodelcalculationsofthecrosssectionbasedon perturbativeQCD(pQCD)[23],andtorecentmodelsof

Υ

produc-tionind

+

Au andAu

+

Au collisions[24–29].

2. Experimentalmethods

The maindetectors usedare the STARTimeProjection Cham-ber (TPC) [30] for tracking andthe STARBarrel Electro-Magnetic Calorimeter(BEMC)[31]fortriggering.BoththeTPCandBEMCare used forparticle identification. The starting point is theSTAR

Υ

trigger whose main components are a fast hardware Level-0 (L0) trigger, which fires when a tower in the BEMC has energy EL0−BEMC

≥

4

.

2 GeV, and a software Level-2 (L2) trigger, which

requires the presence of two high-energy clusters in the BEMC (

>

4

.

5 GeV and

>

3

.

0 GeV). The cluster pair must also have an opening angle greater than 90◦ andan invariant mass above 5 GeV

/

c2 ₍₆

_.

_{5 GeV}

_/

_c2_{)in p}

₊

_{p (d}

₊

_{Au). Notethat energy}

mea-suredatthetriggeringlevelispartiallycalibratedleadingtosmall butrandombiases.Hence,triggering thresholdsare notprecisein energy. The

Υ

trigger is required to be in coincidence withthe STARminimumbiastrigger.Forp

+

p collisionstheminimumbias triggerisbasedontheSTARBeam-BeamCounters,whileford

+

Au and Au

+

Au it is based on the STAR Zero-Degree Calorimeters (ZDC)andtheVertex-PositionDetectors (VPD).TheL0–L2 combi-nation was usedfor thed

+

Au data in 2008andfor p

+

p data in 2009. In the Au

+

Au 2010 run, an upgrade to the STAR data acquisition system allowed the processing of all the L0 triggers abovethe EL0−BEMC

=

4

.

2 GeV threshold,thus removingtheneed

fora Level-2trigger.

Someofthekeycomponentscommontoalltheseanalysesare the tracking,matching betweenTPC tracks andBEMC L0 andL2 clusters, and electron identification techniques. The main differ-encesbetweenthethree datasetsare summarizedasfollows.For Au

+

Au collisions we usethe charged particlemultiplicity mea-suredintheTPCinordertodeterminethecentralityofthe colli-sion.UsingaGlaubermodelsimulation,themultiplicityclassesin thecollision are usedto estimate the averagenumberof partici-pants (Npart) andnumber ofbinary collisions (Ncoll). Thetrigger,

tracking,andelectronidentificationefficienciesintheAu

+

Au case werestudiedasafunctionofcentrality(seeTable 1).Thepresence oftheunderlyingAu

+

Au eventbackgroundincreasestheenergy measuredinthecalorimetertowersandresultsinaslightincrease inthetrigger efficiencywithincreasing Npart (more central

colli-sions).Similarly,theincreaseinthetrackdensityintheTPCresults inadecreaseinthetrackingefficiencywhichisespecially notice-able at high Npart. We used the specific ionization of the tracks

Table 1

UpsilonreconstructionefficiencyinAu+Au.Thetotalefficiencyincludes trigger-ingefficiency,trackingefficiency,electronidentificationefficiency,andgeometrical acceptance.

Centrality Npart Rapidity Efficiency

0–60% 162±9 |y| <0.5 0.122 0.5<|y| <1.0 0.055 |y| <1.0 0.088 0–10% 326±4 |y| <0.5 0.089 0.5<|y| <1.0 0.039 |y| <1.0 0.064 10–30% 203±10 |y| <0.5 0.125 0.5<|y| <1.0 0.055 |y| <1.0 0.089 30–60% 80±10 |y| <0.5 0.126 0.5<|y| <1.0 0.056 |y| <1.0 0.091

intheTPCgas(dE

/

dx)forelectronidentification. Inaddition,the projectionofthetrackontothelocationoftheBEMCshower max-imumpositionwasrequiredtomatchthemeasured BEMCcluster position. Once a trackwas matched to a calorimeter cluster, the ratio of the energy of the cluster to the TPC momentum (E

/

p) wasalsousedforelectronidentification.Thecombinedacceptance times efficiencyfordetecting an

Υ

at mid-rapidity(

|

y

|

<

0

.

5) in Au

+

Au taking intoaccount allaforementionedeffectswas found tovaryfrom

∼

12% inperipheralcollisionsto

∼

9% incentral colli-sions.

The cuts used in these analyses were chosen such that the tracking andelectron identification efficiencies would be similar across the three datasets, allowing the systematic uncertainties to approximately cancelin the measurement of RA A. Forfurther

detail, the techniques used in these

Υ

measurements were de-scribedextensivelyforourpreviousp

+

p measurement[13]based on a 7

.

9 pb−1 dataset.Allevaluated sourcesofsystematic uncer-tainty aresummarizedinTable 2.Animportanteffectinaddition tothosediscussedin[13]isthechangeintrackingandmass res-olution with increasing detector occupancy. Simulated

Υ

events were embedded in real data andtheir reconstructed line shapes were studiedasafunction ofcollisionsystemanddetector occu-pancy. In p

+

p collisions, we find a mass resolutionof 1.3% for reconstructed

Υ (

1S

)

. Due to additional TPC alignment errors for thed

+

Au andAu

+

Au datathemassresolutionsofthe

Υ (

1S

)

in-creasedto2.7%ind

+

Au andperipheralAu

+

Au collisionsand2.9% incentralcollisions.Thisdecreasedmassresolutionwasaccounted forinthebinary scalingestimatesof

Υ (

1S

+

2S

+

3S

)

yields (see gray bandsinFigs. 1 and4).Systematicuncertaintiesinthose scal-ingestimates(lineshapes)areincludedintheerrorsinTable 2.For all resultswe quote,the

Υ

dataareintegratedover alltransverse momenta.

3. Resultsanddiscussion

Fig. 1 showsthe invariantmass distributions ofelectron pairs forp

+

p (top)andd

+

Au (bottom)inthekinematicregion

|

yΥ

|

<

0

.

5.Unlike-signpairsare shownasredfilledcircles andlike-sign pairsashollowbluecircles.

The data are fit with a parameterization consisting of the sum of various contributions to the electron-pair invariant-mass spectrum. The fit is performedsimultaneously withthe like-sign andunlike-signspectrausingamaximum-likelihoodmethod.The lines inFig. 1show theyield fromthecombinatorialbackground (dashed blue line), the result of adding the physics background from Drell–Yan andbb pairs

¯

(dot-dashed green line), andfinally the inclusionofthe

Υ

contribution(solidred line).The shapeof the Drell–Yan continuum is obtainedvia a next-to-leading order

Table 2

Systematicuncertaintiesaffectingthecrosssectionsandratios.Uncertainties stem-mingfromTPCmomentumresolutionareincludedinthelineshapeuncertainties.

Source Relative uncertainty

Luminosity, Vernier scan (p+p) ±7%

BBC efficiency ±9%

ZDC-Au trig. eff. (d+Au) ±3% Vertex finding eff. (p+p) ±1% Vertex finding eff. (d+Au) ±0.1% Vertex finding eff. (Au+Au) ±0.1%

d+Au min. biasσ ±4% Glauber model params. +0.9%,−0.5% Acceptance +1.7%,−3.0% L0 ADC threshold +8.7%,−2.3% L2 Ecluster +1.2%,−0.6% L2 cosθcut ∼0% L2 mass cut ∼0% Tracking efficiency ±2×5.88% Track-to-tower matching eff. +0.2%,−1.1%

E/p cut efficiency ±3%

dE/dx cut efficiency ±2×2.2%

d+Au excited state ratio +0%,−2% Au+Au excited state ratio +1%,−2%

Υ p+p line shape +6.0%,−4.1% Υd+Au line shape +1.8%,−1.2% ΥAu+Au line shape +0.8%,−0.9% Υ p⊥shape ±1.7% Υ (1S)purity, d+Au and Au+Au +0%,−7.5% Total syst.,σpp +21.1%,−19.0%

Total syst.,σdAu +17.5%,−15.6%

Total syst.,σAuAu +16.0%,−14.1%

Common normalization syst. +12.9%,−12.2%

RdAu, syst. +3.5%,−3.8%

RdAu(1S), syst. +3.5%,−8.4%

RA A, syst. +3.2%,−3.6%

RA A(1S), syst. +3.2%,−8.3%

(NLO)pQCDcalculationfromVogt[32].PYTHIA8wasusedto cal-culate the shape of the bb contribution

¯

[33]. We model each of the

Υ

stateswithaCrystalBallfunction [34],whichincorporates detectorresolutionandlossesfrombremsstrahlunginthedetector material.

Thefitisdonetotheunlikeandlike-signdatasimultaneously. Thefittothecombinatorialbackgroundcomponentextractedfrom thelike-signdataissharedbythefunctionalformusedto param-eterizetheunlike-signdata.Intheusuallike-signsubtraction pro-ceduresomeinformationwouldbelost.Incontrast,byperforming asimultaneousfittoboththelike-signandunlike-signsignalswe optimizethe statistical power of our data.The L2 trigger condi-tionhastheeffectofcutting offthelower invariantmasses. This cut-offshapeisparameterizedinthefitsusinganerrorfunction.

Weintegratetheunlike-signinvariantmassdistributioninthe region 8

.

8–11 GeV

/

c2 _{and subtract from the data the fit to the}

combinatorial,Drell–Yan,andbb background

¯

componentsinorder toobtain the yield of

Υ (

1S

+

2S

+

3S

)

. Afteraccountingfor effi-cienciesandsampledluminosity,we calculatea productioncross sectionin p

+

p collisionsof:Bee

×

d

σ

/

dy

|

|y|<0.5

=

64

±

10

(

stat.

+

fit

)

+₋14₁₂pb.Ourpreviousresultof114

±

38+₋23₂₄pb[13]isconsistent withournewmeasurement.The greater sampledluminosityand decreaseddetectormaterialin2009ledtoimprovedstatisticsand lowersystematicuncertaintiesinthepresentmeasurement.

InFig. 1(b),thegraybandshowstheexpectedsignalfromthe p

+

p datascaled by thenumberofbinarycollisions. Dueto dif-ferencesindetectoroccupancyanddetectorcalibrationsthewidth ofthe

Υ

signal differsbetweencollisionsystemsandcentralities. As discussed in the previous section, a misalignment inthe TPC inthed

+

Au andAu

+

Au datasets ledtoabroadeningofthe

Υ

lineshapescompared to the p

+

p dataset. This can be seen by

Fig. 1. (Color online.)Invariant massdistributionsofelectronpairsinthe region

|yee|<0.5.(a):p+p.(b):d+Au.Unlike-signpairsareshownasredfilledcircles

andlike-signpairsashollowbluecircles.Thegraybandshowstheexpectedyield ifRdAu=1 includingresolutioneffects.Seetextfordescriptionofyieldextraction.

examining the line shapes for the

Υ

states in Fig. 1(a) (p

+

p) andFig. 1(b) (d

+

Au).The averagedetectoroccupancy is compa-rable betweenthe two systems,however the d

+

Au dataset has a noticeably broader line shape due to the aforementioned dif-ferences in calibration. The effects of the broadening of the line shapesare takenintoaccountinsystematicuncertainties(Table 2). Thecomparisonofthegraybandwiththed

+

Au datainpanel(b) indicates a suppression of

Υ

productionwith respect to binary-collisionscaling.

A similar procedure is followedfor the region 0

.

5

<

|

yΥ

|

<

1 in p

+

p collisions.Wecombinetheresultstoobtainthe differen-tialcrosssection:Bee

×

d

σ

/

dy

|

|y|<1

=

58

±

12

(

stat.

+

fit

)

+₋1211pb.In

d

+

Au collisions, weanalyze theyields separately in theregions

−

1

<

yΥ

<

−

0

.

5 and 0

.

5

<

yΥ

<

1 because the d

+

Au system isnot symmetricabout y

=

0.Hence, averagingbetweenforward andreverserapiditiesisnotwarrantedasitisinp

+

p. Through-out this paper,the positive rapidity region is the deuteron-going direction, andthe negative rapidity region is the Au-going direc-tion. Integrating over our measured range (

|

yΥ

|

<

1), the cross section in d

+

Au collisions is found to be Bee

×

d

σ

/

dy

|

|y|<1

=

19

±

3

(

stat.

+

fit

)

±

3

(

syst.

)

nb.

We extractthe

Υ (

1S

)

yielddirectlyby integratingover a nar-rowermasswindow(8

.

8–9

.

8 GeV

/

c2).Thismasswindowwas cho-senduetoitshighacceptanceratefor

Υ (

1S

)

anditshighrejection ratefortheexcitedstates.Toaccountforsensitivitytotheshapeof

Fig. 2. (Color online.)(a) Bee×dσ/dy vs.y forp+p collisions(bluestars)andfor d+Au collisions(red,filledcircles;scaleddownby103_{).Notethatthehollowstar}

aty= −0.75 isareflectionofthefilledoneat y=0.75 sincethesearenot inde-pendentmeasurements.ResultsobtainedbyPHENIXareshownasfilleddiamonds. Systematicerrorsareshownasboxesaroundthedata.Theshadedbandsarefrom next-to-leading-orderpQCDcolorevaporationmodelcalculations.Thed+Au pre-dictionusestheEPS09nPDFwhichincludesshadowing[24].(b) RdAuvs. y forSTAR

(redstars)andPHENIX(greendiamonds)results.Thebandontherightshowsthe overallnormalizationuncertaintyfortheSTARresultsduetosystematic uncertain-tiesinthep+p measurement.TheshadedbandshowsthepredictionforRdAufrom EPS09anditsuncertainty.Thedashedcurveshowssuppressionduetoinitial-state partonenergylossandthedot-dashedcurveshowsthesamemodelwithEPS09 incorporated[26].

the

Υ

signal,wevariedtheparametersofthelineshapeobtained fromsimulationsanddata-drivenmethodsdiscussedpreviouslyby their measured uncertainties and varied the excited states from unsuppressedtocompletelysuppressed.Wethenrecalculatedboth efficiencyandpurity(see Table 2,

Υ (

1S

)

purity). Thosevariations were taken into account as additional systematics when quoting

Υ (

1S

)

results.

Fig. 2(a) showstheextracted

Υ (

1S

+

2S

+

3S

)

crosssectionfor p

+

p andd

+

Au asafunctionofrapidity.Thep

+

p measurements areshownasbluestarsandthed

+

Au measurementsasred cir-cles.Thep

+

p resultintheregion0

.

5

<

|

yΥ

|

<

1

.

0 isdisplayedas

astarat y

=

0

.

75 andalsoasahollowstarat y

= −

0

.

75 to illus-tratethatthelatterisnotanindependentmeasurement.Thedata fromPHENIXatforwardrapidityfor p

+

p (filledbluediamonds) andd

+

Au (hollowreddiamonds)arealsoshown[35].

The cross sections in p

+

p are compared to an NLO pQCD calculation of

Υ

production in the Color Evaporation Model (CEM)[23],whichisconsistentwithourdatawithinthestatistical and systematic uncertainties. The same calculation is performed ford

+

Au includingshadowingeffects[24].TheEPS09setof nu-clear Parton Distribution Functions (nPDF) [36] were used. The model isinagreement withour dataexceptforthe mid-rapidity pointwhichislowerthantheprediction.Tostudythisobservation ford

+

Au further,wemakeaclosercomparisontomodelsandto previousmeasurementsof

Υ

productioninp

+

A collisions.

To focus on expected shadowing effects, we obtain the nu-clearmodificationfactorRdAuasafunctionofrapidity.Thenuclear

modificationfactorisdefinedinnucleus–nucleuscollisionsas

RA A

=

1 σA A σpp

×

1

Ncoll

×

Bee

× (

dσ_dyA A

)

Υ Bee

× (

ddyσpp

)

Υ

wherethefirstfactoraccountsforthedifferenceininelasticcross sectioninp

+

p tod

+

Au orAu

+

Au collisions.Thesecondfactor accountsfortheaveragenumberofnucleoncollisionsinad

+

Au or Au

+

Au collision ascalculated by aGlauber model.The third factor accountsforthe measured

Υ

production in p

+

p, d

+

Au or Au

+

Au. We used thefollowing total inelastic cross sections:

σ

pp

=

42 mb,

σ

dAu

=

2

.

2 b,and

σ

AuAu

=

6 b.

Our results for RdAu are shown in Fig. 2(b) and summarized

inTable 3.Ourdata(redstars) arecomparedtoCEMcalculations with theuncertainty fromthe EPS09nPDFshown astheshaded region. Notethatthispredictionfor RdAu,which includes

modifi-cationofthenuclearPDFsbutdoesnotincludeabsorption,implies a modification factorof RdAu

≈

1

.

1.Acalculation in Ref.[25]

ex-ploredvariousnPDFs(EKS98,EPS08,andnDSg)andalsogave RdAu

values above 1 with enhancements in the range of 5–20%. The modelsareinagreementwiththedataexceptinthe y

∼

0 region. Anadditionaleffectwhichcansuppressthe

Υ

yieldisinitial-state parton energyloss.A calculationby Arleo andPeigné[26] incor-porating this effectis shown asthe dashed line. The calculation for a combination of energy loss and shadowing using EPS09 is shownasthedashed–dottedline.Theenergy-lossmodelisalsoin agreement with the data except for the mid-rapidity point. The model from [26] does not include absorption from interactions withspectatornucleons.However,thoseeffectsonlyplayarolein therapidityregion y



1

.

2,wherethe

Υ

mesonswouldbecloser to the frame of the Au spectators. Therefore, the suppression at mid-rapidityisindicativeofeffectsbeyondshadowing,initial-state partonenergyloss,orabsorptionbyspectatornucleons.

We compare our measurements with results from E772 at

√

sN N

=

40 GeV,wheresuppressionofthe

Υ

statesin p

+

A was

observed.This isillustratedinFig. 3(a),whichshowstheratioof the cross section ind

+

Au collisions for STAR(p

+

A for E772)

to that of p

+

p collisions normalized by the mass number A.

E772 plottedaratioofextractedcrosssectionsnormalizedbythe datawheretheprotonbeamhitaliquiddeuteriumtarget( A

=

2). Assuming that the cross section scales as

σ

p A

=

Aα

σ

pp, and

us-ing their p

+

d result as the baseline, the solid line showsthat the ratio should scale as

(

A

/

2

)

α−1. Our measurement in d

+

Au for the

Υ (

1S

)

state (red star) is consistent with the fit to the E772 data, shown as hollow blue circles for

Υ (

1S

)

and hollow greensquaresfor

Υ (

2S

+

3S

)

.Ourresultscovertherapidityrange

|

y

|

<

1 whereastheE772 measurements werein theforward re-gion 0

<

y

<

1

.

05. To better compare our rapidity coverage, we plotthe

α

valueasafunctionofFeynman-x(xF)inFig. 3(b).The

Table 3

TableofRdAuandRA Aresults.Theresultsarelistedintheforma±b±c±d±e wherea isRdAuorRA A,b isthed+Au orAu+Au statisticaluncertainty,c isthep+p

statisticaluncertainty,d isthed+Au orAu+Au systematicuncertainty,ande isthep+p systematicuncertainty.

System Centrality States Rapidity RA A,d A

d+Au Min. bias 1S+2S+3S −1.0<yΥ<−0.5 0.84±0.62±0.18±0.03±0.10 |yΥ| <0.5 0.48±0.14±0.07±0.02±0.06 0.5<yΥ<1.0 1.42±0.82±0.30±0.05±0.17 |yΥ| <1.0 0.79±0.22±0.10±0.03±0.09 1S −1.0<yΥ<−0.5 0.74±0.43±0.16+₋00..0306±0.09 |yΥ| <0.5 0.63±0.18±0.09+₋00..0205±0.08 0.5<yΥ<1.0 1.31±0.63±0.28+₋00..0511±0.16 |yΥ| <1.0 0.83±0.20±0.11+₋00..0307±0.10 Au+Au 0–10% 1S+2S+3S |yΥ| <0.5 0.46±0.05±0.07±0.02±0.05 |yΥ| <1.0 0.49±0.13±0.07±0.02±0.06 1S |yΥ| <0.5 0.69±0.05±0.10+₋00..0206±0.08 |yΥ| <1.0 0.66±0.13±0.10+₋00..0205±0.08 10–30% 1S+2S+3S |yΥ| <0.5 0.69±0.16±0.10±0.02±0.08 |yΥ| <1.0 0.82±0.20±0.12±0.03±0.10 1S |yΥ| <0.5 0.85±0.16±0.13+−00..0307±0.10 |yΥ| <1.0 1.07±0.20±0.16+₋00..0309±0.13 30–60% 1S+2S+3S |yΥ| <0.5 0.74±0.22±0.11±0.03±0.09 |yΥ| <1.0 0.82±0.22±0.12±0.03±0.10 1S |yΥ| <0.5 1.22±0.22±0.18+₋00..0410±0.15 |yΥ| <1.0 1.19±0.22±0.18+₋00..0410±0.14 0–60% 1S+2S+3S |yΥ| <0.5 0.62±0.11±0.09±0.02±0.07 |yΥ| <1.0 0.66±0.09±0.10±0.02±0.08 1S |yΥ| <0.5 0.85±0.11±0.13+₋00..0307±0.10 |yΥ| <1.0 0.88±0.09±0.13+₋00..0307±0.11 larger suppression we observe at mid-rapidity is also consistent

withthelargersuppressionseeninE772forxF

∼

0.

Wenext turn tothe measurements inAu

+

Au collisions.The Au

+

Au invariantmassspectrum isfitin3centralitybins:30–60% (Fig. 4(a)),10–30%(Fig. 4(b)),and0–10%(Fig. 4(c)).AsinFig. 1we show the fits including, in succession, combinatorial background (dashedblueline),thephysicsbackgroundfromDrell–Yanandbb

¯

pairs (dot-dashed green line), and the

Υ

contribution (solid red line).TheabsenceoftheL2triggerintheAu

+

Au datasetremoves thecut-offeffect.Onecanthereforeseethebackground(modeled asthesumoftwoexponentials),dominatedbythecombinatorial component,risingatlowerinvariantmass.Measuredcrosssections aresummarizedinTable 4.The graybandsinthe Au

+

Au figure illustrate the expectedsignal fromthe p

+

p data scaled by the number of binary collisions. There is a clear suppression of the expectedyieldinAu

+

Au collisions.

This suppression is quantified in Fig. 5, which displays the nuclear modification factor, RA A, plotted as a function of Npart

withthe0–10%most-centralcollisions correspondingto

Npart

=

326

±

4.Fig. 5(a) showsthedataforallthreestatesintherapidity range

|

y

|

<

1, whileFig. 5(b) isforthe narrower

|

y

|

<

0

.

5 range. Fig. 5(c) shows RA A and RdAu forthe groundstate

Υ (

1S

)

inthe

range

|

y

|

<

1

.

0.Thedataconfirmthatbottomoniaareindeed sup-pressedind

+

Au and inAu

+

Au collisions.Ford

+

Au collisions, wefindRdAu

(

1S

+

2S

+

3S

)

=

0

.

79

±

0

.

22

(

d

+

Au stat.

)

±

0

.

10

(

p

+

p

stat.

)

±

0

.

03

(

d

+

Au syst.

)

±

0

.

09

(

p

+

p syst.

)

intherange

|

y

|

<

1. Weuseatotalinelasticcrosssectionforp

+

p collisionsof42 mb, ford

+

Au collisionsof2.2 b,and

Ncoll

=

7

.

5

±

0

.

4 forcalculating

RdAu. Inthe same rapidityrange andforthe 0–10%most-central

Au

+

Au collisions,wefindRA A

(

1S

+

2S

+

3S

)

=

0

.

49

±

0

.

13

(

Au

+

Au

stat.

)

±

0

.

07

(

p

+

p stat.

)

±

0

.

02

(

Au

+

Au syst.

)

±

0

.

06

(

p

+

p syst.

)

, whichis

≈

4

.

5

σ

away fromunity.The resultsaresummarizedin Table 3.

Inthenarrowerrapidityrange(Fig. 5(b)),we seean indication ofalower RdAu asdiscussedearlier.Our dataandtheE772 data

show alarger suppressionat y

∼

0 or xF

∼

0 thanthat expected

fromshadowing.Thelevelofsuppressionweobservefor

|

y

|

<

0

.

5 staysapproximatelyconstantfromd

+

Au uptocentralAu

+

Au col-lisions.Thissuggeststhatsuppressionind

+

Au inthiskinematic rangeneedsto beunderstoodbeforeinterpretingthesuppression inAu

+

Au.

For d

+

Au collisions we find RdAu

(

1S

)

=

0

.

83

±

0

.

20

(

d

+

Au

stat.

)

±

0

.

11

(

p

+

p stat.

)

₋+₀0._.03₀₇

(

d

+

Au syst.

)

±

0

.

10

(

p

+

p syst.

)

in the range

|

y

|

<

1

.

0. For the 0–10% most-central collisions we find RA A

(

1S

)

=

0

.

66

±

0

.

13

(

Au

+

Au stat.

)

±

0

.

10

(

p

+

p

stat.

)

+₋0₀._.02₀₅

(

Au

+

Au syst.

)

±

0

.

08

(

p

+

p syst.

)

.Similar suppression is found by CMS in Pb

+

Pb collisions (RA A

(

1S

)

≈

0

.

45 at

simi-lar Npart)[37–39].We observethenuclearmodificationfactorfor

the

Υ (

1S

)

as a function of Npart to be consistent withunity in

d

+

Au throughmid-centralAu

+

Au collisions(seeFig. 5(c)).Inthe mostcentral Au

+

Au collisions,wesee anindication of suppres-sionofthe

Υ (

1S

)

atthe2

.

7

σ

level.Inthecontextofsuppression of theexcited states,ifthe feed-down fractionremains

∼

49% as measured at higher energies and high-p_⊥, it is possible that an RA A

(

1S

)

aslowas0

.

51 couldbeduesolelytosuppressionofthe

excitedstates[43].

One canrelate the RA A ofthe combinedstates tothat ofthe

ground state via the equation RA A

(

1S

+

2S

+

3S

)

=

RA A

(

1S

)

×

(

1

+

NA A

(

2S

+

3S

)/

NA A

(

1S

))/(

1

+

Npp

(

2S

+

3S

)/

Npp

(

1S

))

.The

ra-tiooftheexcited statestothegroundstatecanbeobtainedfrom measurements by CMSandFermilab experiments[40,41] and al-ternativelyfromcombiningtheoreticalcalculations[23]with mea-sured branching ratios from the PDG [42]. In the case where NA A

(

2S

+

3S

)

=

0,RA A

(

1S

+

2S

+

3S

)

≈

RA A

(

1S

)

×

0

.

7.Thisis

con-sistentwithourobserved RA A values,andcanalsobeinferredby

examiningthe massrange10–11 GeV

/

c2 _{inFig. 4,whereno}

sig-nificant2Sor3Ssignalsareseen.

By applying the methods described in [44], we can calculate an upper limit on the RA A of the combined 2S and 3S states.

Fig. 3. (Color online.)Comparisonofourd+Au measurementstothepA measure-mentsfromE772.(a):RatioofΥ productioninpAtoppscaledbymassnumber asafunctionofmassnumber.Shownarethe1S(hollowbluecircles)and2S+3S (hollowgreensquares)Υ measurementsfromE772andour1Smeasurement(red star).Also shownis themodel usedbyE772whereσp A=Aασpp.E772 found

α=0.962±0.006[21].(b):ExponentαasafunctionofxF.Thevertical,dashed redlinesatthebottomoftheplotdenotethewidthofthexF binsfortheSTAR measurements.NotethattheSTARdatapoints areoffsetwithinthebinsforclarity.

background,wefindan upperlimit of29signalcountswith95% confidence in the mass range 10–11 GeV

/

c2 _{for 0–60%}

central-ity collisions. To transform this upper limit into an upper limit on RA A

(

2S

+

3S

)

,we assumedthat thepurityofexcited statesin

thismass rangeisthe same asinthe p

+

p case.While the ex-cited states are likely more suppressed than the groundstate in theAu

+

Au case,usingthep

+

p puritygivesusanupperlimitin theAu

+

Au puritywhichcanbe usedtocalculatean upperlimit ontheRA A.The2S

+

3S crosssectioninp

+

p wasextractedfrom

thefull crosssection,assuming thepuritycan beobtainedbased onthePDGbranchingratios[42]andtherelativeproductioncross sectionsof thethree states.In thecentrality rangeof0–60%, we thusobtain a 95%-confidenceupperlimit of RA A

(

2S

+

3S

)

<

0

.

32

(seeFig. 6).

Ourdata are alsocompared to model calculations incorporat-ing hot-nuclear-mattereffects forAu

+

Au [27–29].These aimto incorporatelattice-QCDresults pertinentto screening and

broad-Fig. 4. (Color online.)Invariantmassdistributionsofelectronpairsintheregion

|yee|<1.0 forthe centrality selections30–60% (a), 10–30% (b),and 0–10%(c).

Unlike-signpairsareshownasfilledredcirclesandlike-signpairsashollowblue circles.Fitsaredescribedinthetext.Thegraybandshowstheexpectedsignal as-sumingscalingofthe p+p yieldwiththenumberofbinarycollisionsincluding resolutioneffects.

eningofbottomoniumandtomodelthedynamicalpropagationof the

Υ

meson inthe colored medium.Both modelsare in agree-ment with the level of suppression seen in Au

+

Au. The model proposed by Emerick, Zhao, and Rapp (EZR), Ref. [28], includes possible CNM effects, modeled as an absorption cross section of up to 3mb whichcan account foravalue of RA A aslow as0.7.

In this model the additional suppression to bring RA A down to

Table 4

ΥproductioncrosssectionsinAu+Au collisions.Thefirstuncertaintylistedisthe combinationofthestatisticalandfituncertaintiesandthesecondisthesystematic uncertainty. Centrality Rapidity dσ/dy (nb) 0–60% |yΥ| <0.5 2170±357±349 |yΥ| <1.0 2180±250±351 0–10% |yΥ| <0.5 3950±416±636 |yΥ| <1.0 3990±1020±642 10–30% |yΥ| <0.5 3040±676±489 |yΥ| <1.0 3430±827±552 30–60% |yΥ| <0.5 905±225±146 |yΥ| <1.0 950±198±153

etal. [29] in Fig. 5(c)is for the inclusive

Υ (

1S

)

RA A, using the

internalenergyastheheavy-quarkpotentialandaninitial temper-atureofthefireballofT

=

340 MeV, whichgiventheinput from latticeQCDresults,isnothotenoughtomeltthedirectlyproduced

Υ (

1S

)

.Hence,thesuppressionismostlydrivenbythedissociation oftheexcitedstates(boththeS-statesandtheP-states).Theinitial temperatureusedintheEZRmodelis330 MeV(withaformation timeof0

.

6 fm

/

c).ThetemperaturesoftheQGPneededin Strick-land’smodel,Ref.[27],areintherange428–442 MeV.However,it shouldbenotedthatneithertheStrickland model,northe calcu-lationfromLiuetal. includeanyCNMeffects.

Consideringtwopossiblesourcesofsuppression,CNMandQGP effects,weusedaMonteCarlopseudoexperimenttocompareour results to different possible sources of suppression. We investi-gated four possible scenarios: (1) No suppression compared to p

+

p;(2) SuppressionduetoCNMeffectsonly;(3) QGP suppres-sion only; (4) Suppression from both CNM and QGP effects. We simulated

Υ

productioninp

+

p,d

+

Au,andAu

+

Au collisionsvia a Poissongenerator. CNM effects were included via the suppres-sionparametrizationusedbyE772[21]andpresentedinFig. 3(a). We used the predictions fromthe Strickland model [27] to esti-matesuppressionfromQGPeffects.Forscenario(4),theexpected suppressionissimply takentobe theproduct ofthesuppression fromscenario(2) andscenario(3).Forthispseudoexperimentwe assumedaflatpriorbasedontheallowedRA A giveninStrickland–

Bazow [27], depicted as the band for this calculation in Fig. 5, stemmingfromthechoiceof1

<

4

π η

/

S

<

3.

AsummaryofthepseudoexperimentresultsisshowninFig. 7. Panel(a) showsour result for RdAu in the range

|

y

|

<

1

.

0

com-pared to scenarios (1) and (2), shown as the solid line and dottedhistogram,respectively.The‘no-cold-suppression-scenarios’ (1 and 3)areexcludedwhiletheCNMeffectsfromE772 parame-terizationareconsistentwithourobservation.Panel(b)showsRA A

forthemost-centralAu

+

Au binintherange

|

y

|

<

1

.

0.By compar-ingtheresultsofthepseudoexperimentswithourmeasurements, weareabletoexcludescenario(1) ata

∼

5

σ

confidencelevel. Fi-nally,weseethathypothesis(4) (dot-dashedcurve),includingboth hotandcoldnucleareffects,isconsistentwithourmeasurements whenboththed

+

Au andAu

+

Au resultsaretakenintoaccount. We repeated this procedure for the rapidity range

|

y

|

<

0

.

5. The results are shown in Figs. 7(c) and (d). In the mid-rapidity rangewefindalargeramountofsuppressionind

+

Au thanwhat we observe in the range

|

y

|

<

1

.

0. Furthermore, RdAu is

compa-rableto RA A in0–10%forthisrapidity range.Thiscould indicate

that suppression ofbottomonium already occursin d

+

Au colli-sions.However, giventhe uncertainties inour currentresults,no particularmodelof

Υ

suppressionind

+

Au isfavored.Hence, fur-therinvestigation ofcold-nuclear-mattereffects on

Υ

production ishighlywarranted.Thesuppressioneffectsseenind

+

Au,which

Fig. 5. (Color online.)NuclearmodificationfactorforΥ (1S+2S+3S),in|y|<1.0 (a) andin|y|<0.5 (b),andΥ (1S)in|y|<1.0 (c),ind+Au (greensquare)andAu+Au (blackcircles)collisionsasafunctionofNpart.Theboxesaroundunityshowthe

statistical(shaded)andsystematic(filled)uncertaintyfromthep+p measurement.

Thegraybandsaroundthedatapointsarethesystematicuncertainties.Thedata arecomparedtocalculationsfromRefs.[27–29].

are not explained by the modelsdiscussed here, still need to be understoodbeforetheAu

+

Au resultscanbefullyinterpreted.

4. Conclusions

Inconclusionwestudied

Υ (

1S

+

2S

+

3S

)

productionin p

+

p, d

+

Au, and Au

+

Au collisions at

√

s

=

200 GeV. We measured the cross section in p

+

p collisions to be Bee

×

d

σ

/

dy

|

|y|<1

=

Fig. 6. (Color online.)Nuclearmodificationfactorofquarkoniumstatesasafunction ofbindingenergyasmeasuredbySTAR.ThehorizontallineoftheΥ (2S+3S)upper limitspanstherangefromthe3Sto2Sbindingenergy; thearrowisplacedat theweightedaverageofthebindingenergies.Thehigh-pT J/ψ resultsarefrom

Ref.[45].

61

±

8

(

stat.

+

fit

)

₋+₁₂13

(

syst.

)

pb andfindittobeconsistentwithin er-rorswithNLOcalculations.Thecrosssectionind

+

Au collisionsis foundtobe Bee

×

d

σ

/

dy

|

|y|<1

=

19

±

3

(

stat.

+

fit

)

±

3

(

syst.

)

nb.We

obtainanuclearmodificationfactorinthisrapidityregion(

|

y

|

<

1) of RdAu

(

1S

+

2S

+

3S

)

=

0

.

79

±

0

.

22

(

d

+

Au stat.

)

±

0

.

10

(

p

+

p

stat.

)

±

0

.

03

(

d

+

Au syst.

)

±

0

.

09

(

p

+

p syst.

)

.Modelsof

Υ

produc-tionincold nuclearmatter, whichincludeshadowingand initial-state partonic energy loss, are consistent with the cross-sections we observein ourd

+

Au data. Higher statistics d

+

Au dataare required to further investigate the 3

σ

deviation we observe at

|

y

|

<

0

.

5.Wemeasured the

Υ (

1S

+

2S

+

3S

)

nuclearmodification factorin Au

+

Au collisions at

√

sN N

=

200 GeV as a function of

centrality. In the range

|

y

|

<

1 and in 0–10% most-central colli-sions we find RA A

(

1S

+

2S

+

3S

)

=

0

.

49

±

0

.

13

(

Au

+

Au stat.

)

±

0

.

07

(

p

+

p stat.

)

±

0

.

02

(

Au

+

Au syst.

)

±

0

.

06

(

p

+

p syst.

)

, indicat-ing additional

Υ

suppression inhot nuclear matter compared to coldnuclearmatter.In0–60%centralitywe finda95%-confidence upper limit on the nuclear modification of the excited states of RA A

(

2S

+

3S

)

<

0

.

32. Calculations of the centrality dependence

of

Υ

RA A usingmodels basedonlatticeQCD calculationsof

bot-tomonium melting in a hot medium are found to be consistent withourdata.Therefore,thesuppressionseenincentral Au

+

Au collisions is indicative of the presence of deconfined matter in heavy-ioncollisions. Itwouldbedesirabletohavea higher statis-ticsd

+

Au datasetinordertostrengthentheconclusionsregarding cold-nuclearmodificationsto

Υ

productionbeforeastronger con-nectionbetweenpartondeconfinement,Debye screening,andthe observed

Υ

suppressioninAu

+

Au canbemade.

Acknowledgements

WethankR.Vogt,M.Strickland,R.Rapp,F.Arleo,andJ.P. Lans-bergforproviding uscalculationsin theSTAR kinematicregions. WethanktheRHICOperations GroupandRCFatBNL,theNERSC Center atLBNL, theKISTI Center in Korea, andtheOpen Science Grid consortium for providing resources and support. This work was supported in part by the Offices of NP andHEP within the U.S.DOEOfficeofScience,theU.S.NSF,CNRS/IN2P3,FAPESP–CNPq ofBrazil,theMinistryofEducationandScienceoftheRussian Fed-eration,NNSFC,CAS,MoSTandMoEofChina,theKoreanResearch Foundation, GA and MSMT of the Czech Republic, FIAS of Ger-many,DAE,DST, andCSIRofIndia,theNationalScienceCentre of Poland,NationalResearchFoundation(NRF-2012004024),the

Min-Fig. 7. (Color online.)Summaryoftheresultsoffourdifferentpseudoexperiments: NoSuppression(solidgold),CMNeffectsonly(dashedred),QGPeffectsonly(dotted green),andbothCMNandQGPeffects(dot-dashed blue).Weshowourresultsfor twosystemsandtworapidityranges:(a) d+Au|y|<1.0,(b) Au+Au|y|<1.0, (c) d+Au|y|<0.5,(d) Au+Au|y|<0.5.Ourdataisshownasaredverticalline withsystematicsshownbythepinkbox.TheQGPeffectsaremodeledin[27].

istry ofScience, Education andSports ofthe Republicof Croatia, andRosAtomofRussia.

References

[1]T.Matsui,H.Satz,Phys.Lett.B178(1986)416.

[2]N.Brambilla,S.Eidelman,B.K.Heltsley,R.Vogt,G.T.Bodwin,E.Eichten,A.D. Frawley,A.B.Meyer, et al.,Eur. Phys.J. C71(2011) 1534, arXiv:1010.5827 [hep-ph].

[3]F.Karsch,E.Laermann,in:R.C.Hwa,etal.(Eds.),QuarkGluonPlasma3,World Scientific,2004,pp. 1–59,arXiv:hep-lat/0305025.

[4]M.Cheng,S.Ejiri,P.Hegde,F.Karsch,O.Kaczmarek,E.Laermann,R.D. Mawhin-ney,C.Miao,etal.,Phys.Rev.D81(2010)054504,arXiv:0911.2215[hep-lat].

[5]S.Borsanyi,etal.,Wuppertal–BudapestCollaboration,J.HighEnergyPhys.1009 (2010)073,arXiv:1005.3508[hep-lat].

[6]S.Digal,P.Petreczky,H.Satz,Phys.Rev.D64(2001)094015,arXiv:hep-ph/ 0106017.

[7]M.Laine,O.Philipsen,P.Romatschke,M. Tassler,J. HighEnergy Phys.0703 (2007)054,arXiv:hep-ph/0611300.

[8]P.Petreczky,C.Miao,A.Mocsy,Nucl.Phys.A855(2011)125,arXiv:1012.4433 [hep-ph].

[9]A.Dumitru,Y.Guo,M.Strickland,Phys.Rev.D79(2009)114003,arXiv:0903. 4703[hep-ph].

[10]A.Rothkopf,T.Hatsuda,S.Sasaki,Phys.Rev.Lett.108(2012)162001,arXiv: 1108.1579[hep-lat].

[11]A.Adare,etal.,PHENIXCollaboration,Phys.Rev.Lett.98(2007)232002,arXiv: hep-ex/0611020.

[12]A.Adare,et al.,PHENIXCollaboration,Phys.Rev.D82(2010)012001,arXiv: 0912.2082[hep-ex].

[13]B.I.Abelev,et al.,STARCollaboration,Phys. Rev.D82(2010)012004,arXiv: 1001.2745[nucl-ex].

[14]H.Agakishiev,etal.,STARCollaboration,Phys.Rev.D83(2011)052006,arXiv: 1102.2611[nucl-ex].

[15]M.Cacciari,P.Nason,R.Vogt,Phys.Rev.Lett.95(2005)122001,arXiv:hep-ph/ 0502203.

[16]J.Adams,etal.,STARCollaboration,Phys.Rev.Lett.94(2005)062301,arXiv: nucl-ex/0407006.

[17]A.Adare,et al.,PHENIXCollaboration,Phys.Rev.C84(2011)044905,arXiv: 1005.1627[nucl-ex].

[18]C.Gerschel,J.Hufner,Phys.Lett.B207(1988)253.

[19]C.Gerschel,J.Hufner,Annu.Rev.Nucl.Part.Sci.49(1999)255,arXiv:hep-ph/ 9802245.

[20]Z.-w.Lin,C.M.Ko,Phys.Lett.B503(2001)104,arXiv:nucl-th/0007027.

[21]D.M.Alde,H.W.Baer,T.A.Carey,G.T.Garvey,A.Klein,C.Lee,M.J.Leitch,J.W. Lillberg,etal.,Phys.Rev.Lett.66(1991)2285.

[22]B.Alessandro,etal.,NA50Collaboration,Eur.Phys.J.C48(2006)329,arXiv: nucl-ex/0612012.

[23]A.D.Frawley,T.Ullrich,R.Vogt,Phys.Rep.462(2008)125,arXiv:0806.1013 [nucl-ex].

[24]CalculationsforcharmoniumwithshadowingdoneinR.Vogt,R.E.Nelson,A.D. Frawley,in:PoSConfinementX,vol. 203,2012.

Bottomoniumcalculationsbasedonasimilarprocedure(private communica-tion).

[25]E.G.Ferreiro, F.Fleuret,J.P.Lansberg,N. Matagne, A.Rakotozafindrabe,Eur. Phys.J.C73(2013)2427,arXiv:1110.5047[hep-ph].

[26]F. Arleo,S. Peigné, J. HighEnergy Phys. 1303 (2013) 122,arXiv:1212.0434 [hep-ph],andprivatecommunication.

[27]M.Strickland,D.Bazow,Nucl.Phys.A879(2012)25,arXiv:1112.2761[nucl-th].

[28]A.Emerick, X.Zhao,R.Rapp,Eur.Phys.J. A48(2012) 72,arXiv:1111.6537 [hep-ph].

[29]Y.Liu,B.Chen,N.Xu,P.Zhuang,Phys.Lett.B697(2011)32,arXiv:1009.2585 [nucl-th].

[30]M.Anderson,etal.,Nucl.Instrum.MethodsPhys.Res.,Sect.A499(2003)659.

[31]M.Beddo,etal.,Nucl.Instrum.MethodsPhys.Res.,Sect.A499(2003)725.

[32] R.Vogt,RHIC-IIScienceWorkshopApril28–30,2005,http://rhicii-science.bnl. gov/heavy/doc/April05Meeting/.

[33]T.Sjostrand,arXiv:0809.0303[hep-ph].

[34] J.Gaiser,PhD thesis,SLAC-R-255,AppendixF,p. 178.

[35]A.Adare,etal.,PHENIXCollaboration,Phys.Rev.Lett.109(2012)242301,arXiv: 1211.4017[nucl-ex].

[36]K.J.Eskola,H.Paukkunen,C.A.Salgado,J.HighEnergyPhys.0904(2009)065, arXiv:0902.4154[hep-ph].

[37]S.Chatrchyan,etal.,CMSCollaboration,Phys.Rev.Lett.107(2011)052302.

[38]S.Chatrchyan,etal.,CMSCollaboration,J.HighEnergyPhys.1205(2012)063, arXiv:1201.5069[nucl-ex].

[39]S. Chatrchyan,et al.,CMSCollaboration,Phys.Rev.Lett.109(2012)222301, arXiv:1208.2826[nucl-ex].

[40]V. Khachatryan,et al., CMSCollaboration, Phys. Rev.D 83(2011) 112004, arXiv:1012.5545[hep-ex].

[41]G.Moreno,C.N.Brown,W.E.Cooper,D.Finley,Y.B.Hsiung,A.M.Jonckheere,H. Jostlein,D.M.Kaplan,etal.,Phys.Rev.D43(1991)2815.

[42]J.Beringer,etal.,ParticleDataGroup,Phys.Rev.D86(2012)010001.

[43]T. Affolder,etal., CDFCollaboration, Phys.Rev.Lett.84(2000)2094,arXiv: hep-ex/9910025.

[44]W.A.Rolke,A.M.López,J.Conrad,Nucl.Instrum.MethodsPhys.Res.,Sect.A 551(2005)493.

[45]L. Adamczyk,etal., STARCollaboration, Phys. Lett.B722(2013)55, arXiv: 1208.2736[nucl-ex].