1 Joint meeting IPNO-LAL LUA9-AFTER Orsay, Novembre 2013 Roberta Arnaldi INFN Torino (Italy)

2 History of heavy-ion quarkonium studies Quarkonium suppression is, since 25 years, one of the most striking signatures for QGP formation in AA collisions s GeV/c ~ SPS Year RHIC LHC AA LHC pA

3 From suppression…to (re)combination Differences in the binding energies of the quarkonium states lead to a sequential melting of the states with increasing temperature (Digal,Petrecki,Satz PRD 64(2001) )  thermometer of the initial QGP temperature Increasing the energy of the collision the cc pair multiplicity increases An enhancement via (re)combination of cc pairs producing quarkonia can take place at hadronization or during QGP stage P. Braun-Muzinger and J. Stachel, Phys. Lett. B490(2000) 196, R. Thews et al, Phys.ReV.C63:054905(2001) (2S) cc T>>T c TcTc (2S) ϒ (1S) T<T c TcTc J/ (2S) ϒ (1S) T~T c TcTc J/ (2S) ϒ (1S) T~3T c TcTc J/

4 How can we measure medium effects? If there are medium effects  R AA  1 If yield scales with the number of binary collisions  R AA = 1 Nuclear modification factor R AA : Hot Medium effects: quarkonium suppression enhancement due to recombination Cold Nuclear Matter effects (CNM): Nuclear parton shadowing Parton energy loss cc in medium dissociation knowledge of CNM effects fundamental to disentangle genuine QGP induced suppression in AA need infos on quarkonium production in pA collisions! compare quarkonium yield in AA with the pp one, scaled by a geometrical factor (from Glauber model)

5 From “low energy” experiments… Charmonium production deeply investigated at RHIC: stronger suppression at forward rapidities SPS vs. RHIC: similar R AA pattern versus centrality Puzzles from SPS and RHIC Observation of J/ suppression No final theoretical explanation Decisive inputs expected from LHC results, having access to: higher energy larger cc multiplicity other quarkonium states (bottomonium) SPS (NA50, NA60) s NN = 17 GeV RHIC (PHENIX,STAR) s NN =39,62.4,200GeV Hint for (re)combination at RHIC? Eur.Phys.J.C71:1534,2011

6 … to LHC data! ALICE CMS ATLAS J/, (2S),    +  - 2.5<y<4 J/  e + e - |y|<0.9 p T coverage down to p T ~0 p T J/>6.5GeV/c J/   +  - |y|<2.4 J/, (2S)   +  - |y|<2.4 p T >6.5GeV/c LHCb J/   +  - 2<y<4.5 p T coverage down to p T ~0 (no heavy ion physics program) Complementary quarkonium results from LHC experiments    +  - |y|<2.4 p T >0 Currently available AA and pA results: (J/ p T coverage depends on y)

Clear J/ suppression with almost no centrality dependence above N part ~100 Less J/ suppression at mid-rapidity wrt forward y for central events 7 ALICE: low p T J/ Is this the expected signature for (re)combination ? Comparison with PHENIX: ALICE results show weaker centrality dependence and smaller suppression for central events ALICE 2.5<y J/ <4 PHENIX 1.2<|y J/ |<2.2 ALICE |y J/ |<0.9 PHENIX |y J/ |<0.35 ALICE Coll. arXiv:

8 ALICE R AA vs p T Models: ~50% of low-p T J/ are produced via (re)combination, while at high p T the contribution is negligible J/ production via (re)combination should be more important at low transverse momentum recombination primordial Different suppression for low and high p T J/ Smaller R AA for high p T J/ p T region accessible by ALICE

9 CMS: high p T J/ Good agreement with ALICE (at high p T ) in spite of the different rapidity range High p T : stronger J/ suppression at LHC wrt to RHIC (re-combination should not play a role) The high p T region can be investigated by CMS!

10 Hint for J/ flow in heavy-ion collisions (LHC), contrary to v 2 ~0 observed at RHIC! The contribution of J/ from (re)combination should lead to a significant elliptic flow signal at LHC energy ALICE: qualitative agreement with transport models including regeneration J/ flow ALICE PRL111, (2013) arXiv: STAR D.Moon, HP2013 CMS: path-length dependence?

11 (2S) studied by both CMS and ALICE (different kinematics) comparing the (2S) yield to the J/ one in Pb-Pb and in pp (2S) in Pb-Pb Study of other charmonium states can help constraining theoretical models At SPS, (2S) is more suppressed than J/ and the suppression increases with centrality Difference trend in ALICE and CMS: large statistics and systematic errors prevent a firm conclusion on the (2S) enhancement or suppression versus centrality ALICE excludes a large enhancement Nucl.Phys.A 19 (2013), pp CMS PAS HIN (2S) much less bound than J/

12 The  family LHC is the machine for studying bottomonium in AA collisions Main features of bottomonium production: no B hadron feed-down gluon shadowing effect are smaller (re)combination is less important Clear suppression of (nS) in PbPb wrt pp PRL 109, (2012)

13 Clear (2S) suppression even in peripheral collisions (1S) suppression might be compatible with excited state suppression (~50% feed-down)? Compatible with STAR (1S+2S+3S) (within large errors) (arXiv: ) (nS) R AA (nS) suppression increases with centrality STAR R AA ((1S+2S+3S)) = 0.56  CMS R AA ((1S+2S+3S)) ~0.32 Sequential suppression of  states?

14 Comparison  and J/ Similar R AA for low p T inclusive J/ and (1S) Sequential suppression observed for prompt J/ and (nS) at high p T interplay of the competing mechanisms for J/ and  can be different and dependent on kinematics!

15 27 years after first suppression prediction, this is finally observed also in the  sector with very good accuracy! R AA vs binding energy: looser bound states more suppressed than the tighter ones Where are we? Two main mechanisms at play: can qualitatively explain the main features of the results 1.Suppression in a deconfined medium 2.Re-combination (for charmonium) at high s however hot and cold effects not yet disentangled…need pA results! Nucl.Phys.A (2013) 194c-201c

investigate initial/final state Cold Nuclear Matter effects on J/: shadowing, energy loss, parton saturation effects, cc dissociation in the medium… 16 What can we learn from pA? New results from p-Pb data at s NN = 5.02 TeV provide a reference for AA collisions to disentangle genuine QGP effects

17 J/ nuclear modification factor R pA vs y Theoretical predictions: reasonable agreement with shadowing EPS09 NLO calculations (R. Vogt) or EPS09 LO (E. Ferreiro et al) models including coherent parton energy loss contribution (F. Arleo et al) CGC description (H. Fujii et al) seems not to be favoured R pA decreases towards forward y Very good agreement between ALICE vs LHCb results

18 J/ nuclear modification factor R pA vs p T Theoretical predictions: reasonable agreement with shadowing EPS09 NLO calculations (R. Vogt) models including coherent parton energy loss contribution (F. Arleo et al) CGC description (H. Fujii et al) seems not to be favoured Forward y: R pA increases towards high p T Mid-rapidity: R pA tends to increase vs p T Backward y: R pA is rather flat and close to unity Forward rapidityMid-rapidity Backward rapidity

19 J/ R pPb (p T ) vs R PbPb (p T ) Hypothesis: 2  1 kinematics for J/ production  similar x g in spite of different s and y factorization of shadowing effects in p-Pb and Pb-Pb: R PbPb enhanced when corrected by this shadowing evaluation Forward rapidity Mid-rapidity

20 (1S) measurements in p-A EPS09 shadowing models, CGC and coherent energy loss in fair agreement with (1S) R pA result Hint for (1S) R pPb suppression at forward rapidity. Smaller effect at backward y R pPb comparable for J/ and (1S)

21 (2S) measurements in p-A Forward rapidity (2S) is clearly suppressed in p-A wrt pp (at s=7TeV) (2S)/J/ suppression is observed also in d-Au at s=200GeV Shadowing and/or coherent energy loss don’t explain the stronger (2S) suppression. Hot medium effects?

22 L. Benhabib, HP2013 p-Pb vs PbPb: additional final-state effects in Pb-Pb affecting the excited states more than the ground state p-Pb vs pp: excited states more suppressed than the ground states. Suppression increases with increase of charged particle multiplicity (2S) & (3S) measurements in p-A

23 Conclusions LHC charmonium and bottomonium results are now complementing the large wealth of data from SPS, RHIC! the picture is quite complicated because of the interplay of many mechanisms! …but, hopefully, putting together all the pieces of the puzzle, the quarkonium production in AA collisions will be clarified! Quarkonia study in heavy ion collisions is already a 25 years long story! SPS RHIC J/ from B CNM feed down regeneration sequential melting suppression LHC

24 Backup

25 Quarkonium production Quarkonium production can proceed: directly in the interaction of the initial partons via the decay of heavier hadrons (feed-down) For J/ (at CDF/LHC energies) the contributing mechanisms are: Direct production Feed-down from higher charmonium states: ~ 8% from (2S), ~25% from  c B decay contribution is p T dependent ~10% at p T ~1.5GeV/c Prompt Displaced Feed down and J/ from B, if not properly taken into account, may affect physics conclusions Direct 60% B decay 10% Feed Down 30%

26 ALICE and ATLAS J/

27 (2S)/ 

28 CMS:  theoretical comparison

29 ALICE:  theoretical comparison

30 CMS: high p T J/ Small hint of p T dependent suppression even in the CMS p T range Good agreement with ALICE (high p T ) in spite of the different rapidity range High p T : stronger J/ suppression at LHC wrt to RHIC (re-combination should not play a role) The high p T region can be investigated by CMS!

31 J/ R AA vs rapidity Suppression beyond the current shadowing estimates. Important to quantify cold nuclear matter effects in p-A collisions At low p T (ALICE) R AA decreases by 40% from y=2.5 to y=4 At high p T (CMS) almost no y dependence in the range |y|<2.4 R AA y pattern depends on the J/ p T PRL 109 (2012) , arXiv: JHEP05 (2012) 063, CMS PAS HIN togliere

32 (1S) ALICE vs CMS ALICE has measured the inclusive (1S) R AA in the forward y region Centrality dependence of CMS and ALICE (1S) R AA is comparable Suppression factor is rather constant in the y range covered by ALICE and CMS togliere

33 Ratio forward to backward yields: R FB Less stringent comparison to theory wrt R pA : however theoretical predictions including energy loss show strong nuclear effects at low p T, in fair agreement with the data The R FB ratio shows a rather flat y dependence and a p T dependence with stronger forward to backward suppression at low p T R FB : free from uncertainties on the pp reference

34 (2S) studied by both CMS and ALICE, different kinematics (2S) in Pb-Pb Study of other charmonium resonances can help constraining theoretical models (2S) much less bound than J/ Results from the SPS showed a suppression larger than the J/ one

35 (2S) in Pb-Pb The (2S) yield is compared to the J/ one in Pb-Pb and in pp At SPS, the (2S)/J/ suppression increased with centrality Overall interpretation is challenging Difference trend in ALICE and CMS: large statistics and systematic errors prevent a firm conclusion on the (2S) enhancement or suppression versus centrality ALICE excludes a large enhancement Nucl.Phys.A 19 (2013), pp CMS PAS HIN

36 L. Benhabib, HP2013 p-Pb vs Pb-Pb: additional final state effects in Pb-Pb affecting the excited states more than the ground state p-Pb vs pp: suppression increases with increase of charged particle multiplicity Excited quarkonia states in p-A Excited states suppressed relative to ground states Similar suppression observed also in d-Au at s=200GeV Shadowing and/or coherent energy loss don’t explain the stronger (2S) suppression. Hot medium effects? (2S) is clearly suppressed in p-A wrt pp (at s=7TeV) (2S)(2S) & (3S)