1 Heavy-Flavour Production in Nucleus-Nucleus Collisions: From RHIC to LHC André Mischke...

1

Heavy-Flavour Production in Nucleus-Nucleus Collisions:

From RHIC to LHCAndré Mischke

ERC-Starting Independent Research Group QGP - Utrecht

Hirschegg 2010Hirschegg 2010Strongly Interacting Matter under Extreme ConditionsStrongly Interacting Matter under Extreme Conditions

3838ndnd International Workshop on Gross Properties of Nuclei and Nuclear Excitations International Workshop on Gross Properties of Nuclei and Nuclear ExcitationsHirschegg, Kleinwalsertal, Austria, January 17 - 23, 2010Hirschegg, Kleinwalsertal, Austria, January 17 - 23, 2010

Outline

• Introduction

- heavy-flavour production and energy loss in QCD matter

• Total charm production cross section

• Nuclear modification factor

• Heavy-flavour azimuthal correlations

• Summary

I will not talk about collectivity and Quarkonia

Hirschegg - 22. Jan. 2010 2Andre Mischke (UU)

3

• Simplest way to establish the properties of a system

- calibrated probe

- calibrated interaction- suppression pattern tells about density profile

• Heavy-ion collisions- hard processes serve as calibrated probe (pQCD)

- traversing through the medium and interact strongly

- suppression provides density measure

- General picture: energy loss via medium induced gluon radiation (Bremsstrahlung)

p+p collisionAu+Au collision

Quark-GluonPlasma

after the collision

Probing hot and dense QCD matter

collTAA NppYield

AAYieldpR

)(

)()(

Nuclear modification factor:

Hirschegg - 22. Jan. 2010Andre Mischke (UU)

central RAA data

K.J. Eskola et al., Nucl. Phys. A747 (2005) 511

increasing density

light charged hadrons,neutral pions

photons

Light hadron spectra at RHIC• Strong suppression in central Au+Au collisions

• Suppression well described by energy loss models

• Medium density 30-50 times normal nuclear matter

• Surface bias effectively leads to saturation of RAA with density

• Limited sensitivity to the region of highest energy density

4

~1 ~10time scale (fm)

“Thermalizationof QGP”

Heavy-flavour production ~ h/2mQ

Quarkonia meltsand flow develops

• Primarily produced by gluon fusion in early stage of collision: production rates calculable in pQCD• Sensitivity to initial state gluon distribution M. Gyulassy and Z. Lin, Phys. Rev. C51, 2177 (1995)

• Heavy quarks provide information about the hottest initial phase of the collision• Higher penetrating power: mQ>> Tc, QCD

1015~0.1

Hadronisation

Parton energy loss

Charm,PYTHIA 6.208

Heavy quark production

5

6

Wicks et al, NPA784, 426 (2007)

Energy loss of heavy quarks

BottomCharm

Gluon radiation probability:

parton

hot and dense medium• Probe deeper into the medium

Dead-cone effect: gluon radiation suppressed at small angles ( < mQ/EQ)Dokshitzer & Kharzeev, PLB 519, 199 (2001), hep-ph/0106202

• Less energy loss: Eg > ELQ > EHQ

1

1

dd

d

d2

2

2

Q

Q

LIGHT

HEAVY

E

m

II

Hirschegg - 22. Jan. 2010Andre Mischke (UU)

Semileptonic decay of charm and bottom mesons- first evidence for heavy-flavour particles

F.W. Buesser et al. (CCRS), Nucl. Phys. B113, 189 (1976)

- robust electron trigger- needs handle on photonic background

Full reconstruction of open charmed mesons- direct clean probe: signal in invariant mass

distribution

- difficulty: large combinatorial background; especially in a high multiplicity environment

- event-mixing and/or vertex tracker needed

single or non-photonic electron

Detection of heavy-flavour particles

Hirschegg - 22. Jan. 2010 7Andre Mischke (UU)

• Spectra measured up to 10 GeV/c

• Integrated yield follows binary collision scaling

• Yield strongly suppressed at high pT for central Au+Au

Au+Au0-5%

40-80%

p+p

d+Au

10-40%

Phys. Rev. Lett. 98 (2007) 192301

p+p

Phys. Rev. Lett. 98, 172301 (2007)

Au+Au

Single electron spectra

Hirschegg - 22. Jan. 2010 8Andre Mischke (UU)

d+Au 200 GeVPRL 94 (2005) 062301

Cu+Cu 200 GeVA. Shabetai, QM 2008

Au+Au 200 GeVarXiv:0805.0364 [nucl-ex]

• First identified open charmed mesons in heavy-ion collisions

• Current measurements limited to low-pT

D0 reconstruction in STAR

Hirschegg - 22. Jan. 2010 9Andre Mischke (UU)

• Use all possible signals- D0 mesons

- electrons

- muons

• Charm cross section is well constrained- 90% of total cross section

- direct measurement

- D0 mesons and muons constrain the low-pT region

1.40 0.11( .) 0.39( .) mb

in 0-12% central Au+Au

NNcc stat syst

Charm cross section in STAR

D0

electron

muon

Hirschegg - 22. Jan. 2010 10Andre Mischke (UU)

• “Cocktail” of backgrounds constructed from measured background sources• Comparison to charm, bottom and Drell-Yan from PYTHIA

• In good agreement with single electrons (PRL 103, 082002 (2009))

c dominant

b dominant

after subtraction of cocktail

cc= 518 ± 47(stat) ± 135(sys) ± 190(model) b

bb= 3.9 ± 2.4(stat) +3/-2(sys) b

Di-electron measurement in PHENIXPhys. Lett. B670, 313 (2009)

Hirschegg - 22. Jan. 2010 11Andre Mischke (UU)

Inclusive charm production cross section

x4.5

x2R. Vogt, Eur. Phys. J. Spec. Topic 155, 213 (2008)

• Factor ~2 difference between STAR and PHENIX; but consistency with NLO pQCD (large uncertainties; primarily from scale choice and parton density functions)

• Checks- removal of silicon vertex detectors (STAR)

- better control over background contributions (PHENIX): 1/3 of single electrons are from J/ decays for pT > 5 GeV/c up to 16% decrease in open heavy

• Cross section follows binary collision scalingHirschegg - 22. Jan. 2010 12Andre Mischke (UU)

Open-charmed meson spectra from CDF

• Deviation of 50-100% at moderate and high-pT, but consistent within errors• Theoretically not fully understood …even in pp collisions

pp @ 1.96 TeV5.8 pb-1

¯

Hirschegg - 22. Jan. 2010 13Andre Mischke (UU)

Charm production cross section (cont’d)

• Large uncertainties more data needed to constrain model parameters

• Parton spectra from pQCD input for energy loss models

LHC: 7-10 TeV

NLO pQCD, CTEQ6M parton densitiesR. Vogt, private communication, 2009

Hirschegg - 22. Jan. 2010 14Andre Mischke (UU)

• First promising decay channels

- D* D0s, D0 K, D+ K-++

- D,B e + X

• ALICE has the capability to measure open charm down to pT = 0 in pp and p-Pb (1 GeV/c in Pb-Pb)

• ITS: impact parameter resolution better than 50 m for pT > 1.5 GeV/c

Charm cross section in ALICE

Hirschegg - 22. Jan. 2010 15Andre Mischke (UU)

bbRHIC

bbLHC

ccRHIC

ccLHC

100

25

MC@NLO

LHC ?

c

cbar

g

g

flavor creation (LO) gluon splitting (NLO)

g

g

g

g

c

cbar

0

MC@NLOLO PYTHIA

like-sign e-K pairs3 < pT < 7 GeV/c

NLO processes: gluon splitting A. M., PLB 671, 361 (2009)

Gluon jet energy (GeV)

• e-D0 correlations at 200 GeV- away-side peak: Good agreement of peak shape between LO PYTHIA and MC@NLO

- near-side: small gluon splitting contribution (6.5%)

• Gluon splitting rate at RHIC consistent with MC@NLO

Andre Mischke (UU)

D* - jet azimuthal correlations at LHC

gluon splitting +flavour creation

Full Pythia simulation, pp@10 TeV

• Different fragmentation characteristic: soft charm FF for gluon jets

• Results promising; more statistic needed

jet axis

D*

gluon splitting(z < ~0.5)

300k jets, <Ejet> ~ 30 GeVcandidatesbackground☐ signal

Hirschegg - 22. Jan. 2010 17Andre Mischke (UU)

Nuclear modification factor

Hirschegg - 22. Jan. 2010 18Andre Mischke (UU)

STARPHENIX

light hadrons

Single electron RAAOne of the most surprising results from RHIC

Hirschegg - 22. Jan. 2010 19Andre Mischke (UU)

• Electron yield at high-pT stronger suppressed than expected

• Models implying D and B energy loss are inconclusive yet

• Large suppression requires extreme conditions; DGLV: dNg/dy = 3500

Hope? – AdS/CFT

• Heavy quarks lose momentum according to dp/dt = −p/Q + stochastic

• Equilibration times Q are roughly consistent with single electron RAA

Hirschegg - 22. Jan. 2010 20Andre Mischke (UU)

Open heavy-flavour in Pb+Pb in ALICE

mb = 4.8 GeV

D0 K B e+X

1 year @ nominal luminosity107(109) central Pb+Pb(p+p) events

Open bottom reconstruction through displaced electrons

S/B ≈ 10 % S/(S+B) ≈ 40

D0 K

Hirschegg - 22. Jan. 2010 21Andre Mischke (UU)

)()()(/ thAAt

DAAthD pRpRpR )()()( D from eB from e

/ tAAtAAtDB pRpRpR

Mass hierarchyMass hierarchyColour charge dependenceColour charge dependence

• More details on heavy-flavour quenching mechanism

• RcAA/Rb

AA ratio different for pQCD and AdS/CFT

Heavy-flavour energy loss at LHC

Hirschegg - 22. Jan. 2010 22Andre Mischke (UU)

Single electron – D0 correlations

Hirschegg - 22. Jan. 2010 23Andre Mischke (UU)

compilation by A.M. 2009

• FONLL has large uncertainty around the Be / De crossing point (RHIC: 3 < pT < 10 GeV/c)

• Crossing point at LHC similar to that at RHIC

• Separate De and Be contribution experimentally

D and B contribution to single electrons

M. Cacciari et al., PRL 95, 122001 (2005)

− c e± + X (BR = 9.6%)-- b e± + X (BR = 10.86%)

Hirschegg - 22. Jan. 2010 24Andre Mischke (UU)

25

trigger

Away side: study in-medium D/B lose energy? conical emission?

Near side: study D/B decay contribution to single electrons?

Single electron-tagged correlationsNLO pQCD: D/B meson crossing point is largely unknown

M. Cacciari et al., PRL 95, 122001 (2005)

Hirschegg - 22. Jan. 2010Andre Mischke (UU)

A. M., PLB 671, 361 (2009)

PYTHIA, pp@200 GeV

bottom dominant

charm dominant

Electron-hadron correlations in STAR

26Andre Mischke (UU)

• B and D contributions comparable at pT > 5 GeV/c and consistent with FONLL

• Similar result from PHENIX (PRL 103, 082002)

• Bottom stronger suppressed than expected?

Single electron RAA at LHC

T0 = 1 GeV0 = 0.15 fm/c# quark flavours: 2

Pyquen: Pb+Pb(5%)@5.5 TeV

H. van Hees and R. Rapp, 2007

I. Vitev, A. Adil & H. van Hees, 2007Pyquen

• Pythia afterburner

• Radiative (generalisation of BDMPS) and collisional energy loss (high-pT approximation)

Hirschegg - 22. Jan. 2010 27Andre Mischke (UU)

Single electron – D0 angular correlations2 < pT

trigger ele < 4 GeV/c

• Near side: B decays + gluon splitting charm• Away side: charm flavour creation 900M events

PythiaPyquen

Hirschegg - 22. Jan. 2010 28Andre Mischke (UU)

NLO processes

bottom

charm E. N

orrbin and T. S

jostrand, Eur. P

hys. J. C17, 1

37 (2000)

Hirschegg - 22. Jan. 2010 29Andre Mischke (UU)

GS charm + B decaysGS charm only

NLO processes (such as gluon splitting) become important at LHC

(e, D0): Near-side width and IAA

2 < pTtrigger ele

< 4 GeV/c

IAA of near-side yield

Hirschegg - 22. Jan. 2010 30Andre Mischke (UU)

• Broader peak for Pyquen than Pythia• Suppression of D0 yield for Pyquen• Next: fragmentation function

Summary• Heavy quarks are particularly good probes to study the properties of hot QCD matter (especially transport properties)

• RHIC (200 GeV)

- energy loss of heavy quarks in the medium larger than expected energy loss mechanism not fully understood yet (bottom stronger suppressed as expected ?)

• LHC (14 and 5.5 TeV)

- pp data are important baseline measurements- inclusive charm production cross section & spectral shape: test pQCD- relevant for suppression measurements of Quarkonia states

- NLO processes (like gluon splitting) become important: accessible by “charm content in jets” measurements

- Jet-like heavy-flavour particle correlations: modification of the fragmentation function

- First heavy-ion collisions anticipate in fall 2010

• Exciting time ahead of us…Hirschegg - 22. Jan. 2010

31Andre Mischke (UU)

Pythia ppPyquen PbPb

Backup

Hirschegg - 22. Jan. 2010 32Andre Mischke (UU)

Ke3 decays (MC)

0 Dalitz and conversion (MC)

MC (0+Ke3)

DataPHENIX

• Electromagnetic calorimeter and RICH at mid rapidity

pT < 5 GeV/c

STAR

• ToF + TPC

pT < 4 GeV/c

• EMCal + TPC

pT > 1.5 GeV/c

E/p

Electron identification

Hirschegg - 22. Jan. 2010 33Andre Mischke (UU)

• Photonic electron background

- e+ + e- (small for Phenix)

-0 + e+ + e-

-, etc.

• Phenix is almost material free their background is highly reduced compared to STAR

• Background is subtracted by two independent techniques - very good consistency between them

- converter method (1.68% X0)

- cocktail method

• STAR determines photonic background using invariant mass

Phenix

mass (GeV/c2)

e-

e+

e-dca

Electron background sources

STAR

Hirschegg - 22. Jan. 2010 34Andre Mischke (UU)

Electron-hadron correlations in PHENIX

arXiv:0903.4851

• Use e-K invariant mass

• Statistics limited

• Mid-rapidity electron and forward-rapidity muon -> promissing…

p+p

Hirschegg - 22. Jan. 2010 35Andre Mischke (UU)

Charmonium contribution to single electrons

• New study takes J/ e±+X contribution into account

• 1/3 of single electrons are from J/ decays for pT > 5 GeV/c up to 16% decrease in open heavy

• But what is RAA of high-pT J/ ?

• Background contribution from K0

s decays may also play a role (especially at low-pT) under investigation

1/3 of e from J/ decays

Hirschegg - 22. Jan. 2010 36Andre Mischke (UU)

Charm and bottom RAA

• Knowing RAA of single electrons and relative B contribution rB in p+p collisions, one can study RAA(b) as a function of RAA(c):

• Exclude original radiative calculation

• D and B measurements in A+A necessary

pT > 5 GeV/c

RAA = rB RAA(b) + (1-rB) RAA(c)

I: Djordjevic, Gyulassy, Vogt and Wicks, PLB 632 (2006) 81; dNg/dy = 1000II: Adil and Vitev, PLB 649 (2007) 139III: Hees, Mannarelli, Greco and Rapp, PRL 100 (2008) 192301

Hirschegg - 22. Jan. 2010 37Andre Mischke (UU)

Single electron-hadron correlations in Au+Au

• Away-side modification?

• Improved statistics and better background rejection needed

• Similar analysis in PHENIX

STAR preliminary

Hirschegg - 22. Jan. 2010 38Andre Mischke (UU)

UA1 jet reconstruction - cone size 0.4 - Ethr = 10 GeV/c2

D* in jets

Hirschegg - 22. Jan. 2010 39Andre Mischke (UU)

Heavy-flavour production at LHC

• Heavy flavour copiously produced

• Charm and bottom yields (NLO pQCD predictions using MNR PDF)

system, sNN

pp @ 14 TeVPb+Pb (0-5%)

@ 5.5 TeV

11.2 / 0.5 4.3 / 0.2

0.16 / 0.006 115 / 4.6

[mb] QQNN

QQtotN

bbRHIC

bbLHC

ccRHIC

ccLHC

100

25

Hirschegg - 22. Jan. 2010 40Andre Mischke (UU)

Kinematics for c-cbar pairs

Δη

Δϕ

ALICE (central barrel) acceptance

Gluon splitting

Hirschegg - 22. Jan. 2010 41Andre Mischke (UU)

(e,D0) correlations for like-sign pairs

Hirschegg - 22. Jan. 2010 42Andre Mischke (UU)

Near-side width and yield

Hirschegg - 22. Jan. 2010 43Andre Mischke (UU)

Near-side IAA

Hirschegg - 22. Jan. 2010 44Andre Mischke (UU)

Relevant heavy-flavour decay modes• Charm charmed meson modes

c D0 + X (BR = 56.5%)c D+ + X (BR = 23.2%)c D*+ + X (BR = 23.5%)

semileptonic channelc e+ + X (BR = 9.6%) single (non-photonic) electron continuum

• Bottom charmed meson modes

b D0 + X (BR = 59.6%)b D- + X (BR = 23.5%)b D*+ + X (BR = 17.3%)

semileptonic channelb e+ + X (BR = 10.86%)

Based on e+e- data from CLEO/ARGUS and LEP experiments

Review of Particle Physics, C. Amstler et al. (Particle Data Group), Physics Letters B667, 1 (2008).

Hirschegg - 22. Jan. 2010 45Andre Mischke (UU)

Ldt = 5.1026 cm-2 s-1 x 106 s

5.1032 cm-2 PbPb run, 5.5 TeV

NPbPb collisions = 2 .109 collisions

Ldt dt = 3.1030 cm-2 s-1 x 107 s

5.1037 cm-2 for pp run, 14 TeV

Npp collisions = 2 .1012 collisions

Muon triggers: ~ 100% efficiency, ~ 1kHz Electron triggers: Bandwidth limitation NPbPb central = 2 .108 collisions

Muon triggers: ~ 100% efficiency, < 1kHz

Pb-Pb nominal run pp nominal run

Electron triggers: ~ 50% efficiency of TRD L1 20 physics events per event

Hadron triggers: NPbPb central = 2 .107 collisions

Hadron triggers: Npp minb = 2 .109 collisions

LHC running conditions

Andre Mischke (UU) 46Hirschegg - 22. Jan. 2010