Charmonium suppression

Transcription

1 Charmonium suppression from SPS to RHIC Raphaël Granier de Cassagnac LLR École polytechnique / IN2P3 PHENIX & CMS experiments Seattle, 2009, June 17 th 1/36

2 Introduction overview talk on experimental status of charmonium suppression. Setting up the firm ground of observations Leaving on the side Production mechanisms, polarization, feed downs, fermilab and hera b Some redundancy with at least Rapp, Redlich, Stachel, Frawley, Lansberg, Tuchin, Faccioli, Leitch, Woehri, Linden Levy, Lourenço 2/36

3 The normal introduction Matsui & Satz, PLB178 (1986) 416 In 1986, Matsui & Satz predicted an unambiguous signature of QGP Onset of quarkonia melting above a certain temperature / energy density threshold Exampleof assumed T d (but theorists still working on it, remember yesterday): 3/36

4 0. The published situation before new SPS and RHIC preliminary references shown at QM09 4/36

5 Cold and hot SPS Normal nuclear absorption alone does a splendid job describing pa, SU and peripheral InIn and PbPb: σ abs = 418± mb 0.35 Scaled from 400 to 158 AGeV Beyond is anomalous suppression InIn looks like an onset ± 10% global systematics J/ψ L NA60 preliminary NA50, EPJ C39 (2005) 335 NA60, PRL99 (2007) /36

6 R AuAu (y 0 in PHENIX) R PbPb (@ SPS) Lower rapidity R AA looks surprisingly similar, while there are obvious differences: At a given N part, different energy densities Cold nuclear matter effects (x Bjorken, σ abs ) REDO Without PHENIX, PRL98 (2007) forward SPS from QM06 6/36

7 R AuAu (y 1.7) < R AuAu (y 0) in RHIC, more J/ψ suppression at forward rapidity! While energy density should ldbe smaller REDO 60% PHENIX, PRL98 (2007) R AA (y~1.7) R AA (y~0) What is this further 40% suppression due to? 7/36

8 @ RHIC, more suppression at forward rapidity! Twopossible theoretical explanations A. One hot: coalescence, regeneration B. One cold: saturation, shadowing 8/36

9 A. Hot coalescence, regeneration Large variety of approaches, all justify: R AA (y=0) > R AA (y=1.7) (more c quarks to recombine aty=0) As an example Latest references R. Thews et al, EPJ C43, 97 (2005) Yan, Zhuang, Xu, PRL97, (2006) A. Andronic et al., NPA789, 334 (2007) Ravagli, Rapp, PLB655, 126 (2007) Zhao, Rapp, PLB664, 253 (2008) A. Capella et al., EPJ C58, (2008) O. Linnyk et al., NPA807, 79 (2008) Rapp and Stachel at Seattle (Apologies if I forgot somebody) Capella, Ferreiro, Tywoniuk et al. Fitting Cu+Cu, Au+Au, Mid and forward rapidity 9/36

10 B. Cold matter Assuming two shadowing schemes, derive breakup cross sections from R da (y) σ EKS mb σ NDSG mb Extrapolate to AuAu (Also available for CuCu) Mid and forward are correlated through shadowing scheme If you believe this shadowing + absorption, there is anomalous suppression, larger at forward rapidity. PHENIX, PRC 77 (2008) & erratum: PRC79 (2009) Midrapidity Forward rapidity 10/36

11 B. Cold matter RGdC, HP06 and QM06 PHENIX, PRC 77, (2008) More model independent Ina Glauber data driven driven model, propagate what we know from R da (y,centrality) R AA (y,b) = Σ i R da ( y,b i 1) x R da (+y,b i 2) No shadowing nor absorption schemes Mid and forward are not correlated, Midrapidity Survival = % less model dependent larger uncertainties (especially y@ y 0) Survival = % Anomalous suppression, at least at forward rapidity! Anomalous suppression could be identical at different rapidities Forward rapidity (No dcu, so no CuCu) 11/36

12 B. Recent CGC paper Kharzeev, Levin, Nardi, Tuchin Gluon saturation could further suppress forward J/ψ in AuAu First numerical estimate Absolute amount of suppression is fitted to the AuAu data! Waiting forward to new dau data to fit them first However, rapidity dependence should be ok But it does not reproduce peripheral data Anyway PRL102 (2009) and arxiv: dn/dy Again, J/ψ anomalous suppression could be identical at mid and forward rapidity! 12/36

13 How to disentangle these two scenarios experimentally? (now moving to preliminary material) Twopossible theoretical explanations A. One hot: coalescence, regeneration B. One cold: saturation, shadowing 13/36

14 How to move forward experimentally? RGdC, Quarkonia in hot and cold matters, Quark Matter Calm down? (Better pa/da reference) 2. Be more open? (Measure cc to constrain regen.) 3. Get excited? (ψ, χ c ) 4. Get high? (inmass mass, looking atupsilons) 5. Broaden interest? (in transverse momentum) 6. Let it flow? (elliptically) i 7. Be upset? (and search for onset) 8. Give up? And move to the LHC? Some progress on all these points at this meeting! 14/36

15 1. New preliminary SPS NA GeV New absorption, considering also antishadowing (EKS98) In In 158 GeV (NA60) Pb Pb 158 GeV (NA50) ± 11% global uncertainty Arnaldi, QM09 Still anomalous suppression in Pb+Pb (of ψ and χ c?) Not sizeable in In+In, but such a peculiar shape Considering shadowing 75% With new absorption 68% With old reference 58% 15/36

16 1. New preliminary RHIC EKS σ = 0,1,2,3,4, 15 EKS σ = 0,1,2,3,4, 15,,,, R da not yet available No datadriven driven method Shadowing & σ abs do not describe data anymore One way is to use various (effective) σ abs vs rapidity Linden Levy, QM09 Frawley, Trento & Seattle 16/36

17 Now, a little word of caution Tony s extrapolation of the dau Phenixpreliminary i R CP does reinforce the cold hypothesis to explain the R AA rapidity dff difference However, need different σ abs for different rapidity also seen at SPS, Fermilab So, something else or different is going on there Something for which the extrapolation to R AA may or may not be appropriate So let us think! (and wait for final R da ) Lourenço, Vogt, Woehri, JHEP 0902:014 Frawley, Trento & Seattle 17/36

18 What else could bring information? 2. Open charm? 18/36

19 2. Measuring open charm could constrain both? Regeneration α (N cc ) 2??? 10 to 20 cc pairs Initial state effect (shadowing ) shared with J/ψ A factor of 2 difference between experiments 25% systematic error But binary scaling (within uncertainties ) i At SPS too: 1.16 ±0.16 PHENIX, PRL98 (2007) NA60, EPJ.C59 (2009) /36

20 2. Open charm vs rapidity Only pp, and very poorly known T k b t To know more about open charm, wait for silicon upgrades in Phenix and Star D. Hornback, PHENIX, QM08 20/36

21 Look at other quarkonia 3. Excited charmonia 4. Upsilons 21/36

22 3. Excited states (feed down to J/ψ) Excited states should ψ from ψ = 8.6 ±2.5% A. melt if J/ψ suppression is cold effects + sequential melting B. also regenerate if J/ψ do (and maybe even more) AtRHIC, unfortunately only p+p for now Feeddown ratio ψ dau is feasible ψ AuAu needs new run M. PANIC08 ψ from χ c < 42% (90%CL) Beauty cross section ψ from B = % 22/36

23 ψ at SPS NA50, EPJC48 (2006) 329 NA50, EPJC49 (2007) 559 Trento ψ pa / A at 400 AGeV ψ /Drell Yan J/ψ ψ 450, 400 and 200 GeV points rescaled to 158 GeV close to a factor of 10 Preliminary ψ are more absorbed and suppressed than J/ψ σ abs (158 AGeV) not available 23/36

24 4. Beginning of a bottomonia story Atomssa, Liu, Conesa del QM09 R dau = 098± ± (from STAR) R AuAu < 064@ % CL (from PHENIX ) Could be cold effects No continuum subtraction (but < 15% from pp) Feeddown of χ b important b 50% for p T > 8 GeV/c at CDF Stay tuned [8.5,11.5] GeV/c 2 24/36

25 Look at other observables 5. p T (broadening) 6. Elliptic flow 25/36

26 5. p T SPS? Tested on many systems NA60 pa slope is unique Cronin goes like: <p 2 2 T2 > AB = <p T2 > pp + α x L SPS Corte ese (NA60 0), Hard probes 08 + ho omemade powerpoi int fits 26/36

27 5. p T RHIC? vs N part? Widely unknowninitial charm production: Recombined R AA are poorly constrained Instead look at p T : Hot: Inherited p T should be lower than initial Cold: Cronin effect should broaden initial p T Cronin goes like: <p T2 > AB = <p T2 > pp + α x L PHENIX, PRL 101, (2008) Mid rapidity Forward rapidity No strong <p T2 > dependence Modest rise at forward rapidity Could be broadening No need for recombination here 27/36

28 5. p T RHIC? vs thickness? Widely unknowninitial Mid rapidity F d idit charm production: Recombined R AA are poorly constrained Instead look at p T : Hot: Inherited p T should be lower than initial Cold: Cronin effect should broaden initial p T 1.2< y <2.2 Forward rapidity y <0.35 Cronin goes like: No strong <p T2 > dependence <p T2 > AB = <p T2 > pp + α x L Modest rise at forward rapidity Could be broadening No need for recombination here 28/36

29 5. Reaching higher p T STAR, arxiv: Rapp & Zhao, arxiv: STAR s R CuCu (high p T ) 1 Hot wind scenario 0 Screening length from AdS/CFT But several reasons for R AA to grow at high p T Cronin effect, Bottom contribution, Leakage, (Anti)shadowing Already seen at SPS! R CP of Pb Pb data IMHO, not a big deal NA50, L. Ra amello, QM M05 29/36

30 6. J/ψ elliptic flow in PHENIX Ifrecombined, J/ψ should inherit the (rather large) charm quark elliptic flow. First measurement: C. QM08 E.T. HP08 Various levels l of recombination Will require RHIC2 for a discriminating measurement 30/36

31 6. But also J/ψ elliptic SPS R. QM08 Eccentric In+In collisions F. HP08 MinBias Pb+Pb collisions Cannot tbe due to recombination ( 0.05 cc pairs in In+In) Needs confirmation and understanding 31/36

32 What else? 7. Look foronsetsonsets 8. Go to LHC, the uncharted territory 32/36

33 7. Common onset at SPS and RHIC? Clear misalignment wrt N part Better wrt dn/dη R. Arnaldi + T. Frawley, Trento More next week Getting closer to energy density, more next week 33/36

34 8. J/ψ at LHC? A new story will begin More J/ψ melting Larger shadowing / saturation effects Larger recombination (maybe 200 cc pairs) If recombination prevails golden signal If not, expect same or worse difficulties as at RHIC Example of prediction RHIC A. Andronic et al., NPA789 (2007) 334 J. Stachel, Seattle 34/36

35 8. More LHC A lot of Upsilons Y and Y should ldbe suppressed Y shouldn t (apart from 50% χ b feeddown) CMS 3,700 (10 σ)? Pb b Pb 0,5 nb 1 Frawley, Ullrich, Vogt, PhysRept 462 (2008) /36

36 One strong conclusion: Anomalous conclusions J/ψ suppression isnot (well and yet) understood Cold and hot effects going in the same direction Forward/mid rapidity difference could be due to: A. Regeneration / coalescence of cc pairs? B. Cold effects? more and more favored (next talk) However, conservative cold matter approaches still give significant anomalous suppression at least at SPS and at forward rapidity The hot matter is deconfining some quarkonia Could be only ψ and χ c at SPS New data helps further understanding dauand pa data, ψ and Upsilon, RHIC2 and LHC 36/36

37 That s all folks 37/36

38 1. R dau versus rapidity 38/36

39 8. Quick look at shadowing on J/ψ (emitted gluons Pb/p and p T are neglected) < y A factor of 2x2 < C uncertainty on -4 M < y < < y < 0.9 charm production -2.5 S from current shadowing ALICE knowledge 39/36

40 Lourenço et al, arxiv: E866 : flat with x F if no shadowing is assumed Even with no shadowing, little s dependence of σ abs 40/36

41 Still open questions at SPS Interplay shadowing absorption s dependence of absorption? Lourenço et al, arxiv:0901:3054. NA60, 168 GeV? HP08? QM09? Unexplained rapidity dependence d in pa? Eur.Phys.J.C48:329,2006 σ nds (mb) σ EKS Tram & Arleo, EPJ. C 55, (2008) 41/36

42 J/ψ in pa, NA50 In pa, an unsolved rapidity dependence EPJ +50% +30% 42/36

43 R AuAu (run 4) =R AuAu (run 7) Forward rapidity only (for now) More bins at higher centrality Confirm the trend R AA(y (y 1.7) < R AA(y (y 0) 43/36

44 R AuAu vs R Final CuCu analysis Slightly below 1 in CuCu R AA (y~1.7) R AA (y~0) PHENIX, arxiv: /36

45 R CuCu (STAR, high p T ) 1 2 sigma J/ψ signal in Cu+Cu STAR = PHENIX charm spectra R CuCu raising with p T * * * * These are not phenix results yet, but could become as soon as the two experiments talk to each others 45/36

46 Various R XY (p T ) 1 2 Several (hints of) raising R AA (p T ) 1. R CP PbPb (NA50) 2. R AuAu (PHENIX) 3. R dau (PHENIX) Several potential reasons: Leakage effect, J/ψ escape High p T J/ψ forming beyond QGP Cronin effect Raising x Bj = less shadowing 0.02 to 0.05 from 0 to 9 GeV/c See discussion in Think about it PHENIX, arxiv: compared to Ferreiro, Fleuret, Rakotozafindrabe, arxiv: /36

47 2. Cold matter again? Fitting an effective break up cross section (depending on y) and etrapolateto extrapolate to CuCuC and AuAu STOP Do you agree that we have poor handle on the cold nuclear matter effect? PHENIX, arxiv: /36

48 7. Search for an onset? Onset curves fit the midrapidity AuAu data Chaudhury, nucl th/ Gunji et al, hep ph/ (after CNM subtraction) But so do smooth curves! Nagle nucl ex/0705 ex/ Density y=0 is incompatible with SPS onset Linnyk & al, nucl th/ No y=1.7? Wait for run7 analysis & CNM constraints! STOP Gunji et al, PRC 76 (2007) AK A.K. Chaudhury, nucl-th/ /36

49 Onset curves fit the midrapidity data Density threshold? No! Chaudhury, nucl-th/ Gunji et al, hep-ph/ p (after CNM subtraction) So do smooth curves! Nagle nucl-ex/ Density y=0 is incompatible with SPS onset or larger suppression y=1.7 Linnyk & al, nucl-th/ J. Nagle, nucl-ex/ /36

50 From da to RHIC For a given A+A collision at b AA,, Glauber b 1 b 2 provides a set of N+N collisions occurring at b i1 and b 2 i b AA One minimal assumption is rapidity factorization: R AA ( y,b AA ) = Σ collisions [R da ( y,b i1 ) x R da (+y,b i2 )] / N coll J/ψ = x Works (at least) for absorption & shadowing since production ~ pdf1 x pdf2 x exp ρσ(l 1 +L 2 ) RGdC, hep-ph/ J/ψ = x J/ψ 50/36

51 Heavy flavor elliptic flow Also a surprise! Now, do bees fly? Need the b/c+b in AA to properly estimate the b flow (todo : average the 2 datasets cause they have different stat/syst balance) 51/36

52 NA60 In+In new reference curve (σ abs =7.6mb) old reference curve (σ abs =4.2mb) 52/36

53 53/36