Quarkonium results: lessons from LHC run-1

Quarkonium results: lessons from LHC run-1 E. Scomparin (INFN-Torino) Trento, March 16-20 2015 Introduction, pre-lhc summary LHC run-1 substantial progress on charmonium/bottomonium studies! Open points and prospects for run-2 Conclusions

A lively topic Quarkonium suppression as a QGP thermometer My own thermometer of the interest of the community in quarkonium studies Number of citations of the seminal Matsui s and Satz s paper 14/03/07 1176 09/11/09 1279 T c 06/12/10 1344 22/11/11 1447 09/11/12 1644 17/11/14 2041 Citations doubled in the last ~8 years Still a very hot topic! 2

C. Baglin et al. (NA38), Phys. Lett. B220(1989) 471 (peripheral) (central) ~30 years of experiments CMS Coll., PRL 109(2011) 222301 From the 80s SPS (NA38-NA50-NA60) RHIC (PHENIX-STAR) LHC (ALICE-CMS) until today! 3

HI studies: reference processes Choice of suitable reference process for the modification of the quarkonium yields proved to be crucial (and much debated!) SPS energy Drell-Yan process Cancellation of syst. uncertainties No initial/final state effects in the explored kinematic region Small statistics q q + - RHIC, LHC energy R AA Limited cancellation of uncertainties Does not account for initial/final state effects not related to the medium Precise reference data can be collected Ideal reference Normalization to open charm Most natural reference initial state effects cancel out Differential comparisons not straightforward Charm energy loss shifts D-meson p T J/ suppression removes yield Open charm data have non-negligible uncertainties (especially at low p T ) 4

HI studies: role of CNM Considered as a crucial step for the understanding of A-A results However Description of pa data in terms of various CNM effects is difficult Extrapolation pa AA can be effect-dependent and/or model dependent p c c g J/, c,... At SPS Use of effective quantity abs (break-up cross section) tuned on pa data and then directly extrapolated to A-A (via L-variable) At RHIC Combination of shadowing (factorized out using nuclear PDFs parameterizations) and break-up Centrality dependence exhibits surprising (understood?) effects At LHC Break-up cross section negligible (Coherent) energy loss effects become important First attempts (see later) at an extrapolation pa AA 5

Low energy results: J/ from SPS & RHIC SPS (NA38, NA50, NA60) s NN = 17 GeV RHIC (PHENIX, STAR) s NN = 39, 62.4, 200 GeV R.Arnaldi et al.(na60) NPA830 (2009) 345c A. Adare et al. (PHENIX) PRC84(2011) 054912 First evidence of anomalous suppression (i.e. beyond CNM expectations) in Pb-Pb collisions ~30% suppression compatible with (2S) and c decays suppression, with strong rapidity dependence, in Au-Au at 6 s= 200 GeV

Low energy results: J/ from SPS & RHIC Comparison of SPS and RHIC results N.Brambilla et al. (QWG) EPJC71 (2011) 1534 Good agreement between SPS and RHIC patterns if cold nuclear matter effects are taken into account Compensation of suppression/recombination effects? Suppression of c and (2S) w/o recombination? Understanding cold nuclear matter effects and feed-down is essential for a quantitative assessment of charmonium physics 7

Low energy results: (2S) from SPS & RHIC SPS (NA50) pa, AA @ s NN = 17 GeV NA50 Coll., Eur. Phys. J. C 49, 559 (2007) p-a RHIC (PHENIX) d-au @ s NN = 200 GeV PHENIX Coll., PRL 111, 202301 (2013) S-U Pb-Pb (2S) is more suppressed than J/ already in pa collisions and the suppression increases in Pb-Pb unexpected (2S) suppression (forms outside nucleus) stronger than the J/ one in central d-au 8

Low energy results: from SPS & RHIC SPS (NA50) pa, s NN =29 GeV RHIC (PHENIX, STAR) dau, Au-Au s NN = 200 GeV B. Alessandro (NA50 Coll), PLB 635(2006) 260 A. Adare (PHENIX Coll.), 1404.2246 L. Adamczyk (STAR Coll.) PLB 735 (2014) 127 First measurement at SPS energies. Hint for no strong medium effects on (1S+2S+3S) in pa R AA compatible with suppression of excited states, with large uncertainties 9

Lessons from low-energy A-A Suppression effect on J/ beyond CNM undisputable at both SPS and RHIC Common interpretation: mainly related to screening/dissociation in hot (deconfined) matter Role of J/ regeneration tiny (if any) at SPS energy Quantitatively much debated at RHIC energy Energy scan very interesting in principle (onset of suppression), but only top energy was explored at the SPS (new NA60+ experiment?) Results at s=39 and 62.4 GeV suffer from large uncertainties (absence of proper reference data) (2S) largely suppressed in A-A compared to J/ at SPS energy Commonly seen as an effect of its weak binding resonances out of SPS reach Intriguing results at RHIC, first recent attempt to separate 1S (STAR) Suppression compatible with complete 2S+3S melting, 10 1S suppression only for central events

and questions for LHC 1) Evidence for charmonia (re)combination: now or never! Do we see enhancement vs centrality? Do we see J/ flow? Do we see softer p T distributions? 2) A detailed study of bottomonium suppression (3S) (2S) b (2P) b (1P) Do we see sequential suppression? (as recombination does not play a role) (1S) 11

The main actors ALICE Access mid- and forwardrapidity (e + e - and + respectively) Good mass resolution for J/ (~70 MeV for muons, ~30 MeV for electrons) Full p T acceptance in the whole y-range Prompt vs non-prompt at y=0 J/ ATLAS CMS LHCb ALICE ALICE CMS (high p T ) CMS Excellent mass resolution for muons(35 MeV for J/ ) Prompt vs non-prompt Cut low-p T charmonia 12

The low p T region: ALICE Centrality dependence of the nuclear modification factor studied at both central and forward rapidities Global syst: 13% e + e - 15% + - Inclusive J/ R AA Small effect of non-prompt contribution on the inclusive R AA B. Abelev et al., ALICE Phys. Lett. B 734 (2014) 314 At forward y, R AA flattens for N part 100 Central and forward rapidity suppressions compatible within uncertainties Forward y: No B suppression R AA prompt ~0.94R AA incl Full B suppression R AA prompt ~1.07R AA incl Central y: No B suppression R AA prompt ~0.91R AA incl Full B suppression R AA prompt ~1.17R AA incl 13

Low p T : comparison ALICE vs PHENIX Comparison with PHENIX Stronger centrality dependence at lower energy Systematically larger R AA values for central events in ALICE Behaviour qualitatively expected in a (re)generation scenario Look at the p T dependence of the suppression 14

J/ R AA vs centrality: theory comparison Comparison to theory calculations: Models including a large fraction (> 50% in central collisions) of J/ produced from (re)combination or models with all J/ produced at hadronization provide a reasonable description of ALICE results Still rather large theory uncertainties: models will benefit from a precise measurement of cc and from cold nuclear matter evaluation 15

A (re)generation signature : the p T dependence of R AA Global syst: 8% ALICE 10% PHENIX At low p T, for central events, the suppression is up to 4 times larger at PHENIX, compared to ALICE Strong indication for (re)generation 16

Moving to higher p T : CMS vs ALICE B. Abelev et al., ALICE Phys. Lett. B 734 (2014) 314 Complementary y-coverage: 2.5<y<4 (ALICE) vs 1) 1.6< y <2.4 (CMS, left) 2) y <2.4 (CMS, right) Qualitative agreement in the common p T range 17

CMS results: prompt J/ at high p T CMS PAS HIN-2012-014 CMS-PAS HIN-12-2014 Striking difference with respect to ALICE No saturation of the suppression vs centrality High-p T RHIC results show weaker suppression No significant p T dependence from 6.5 GeV/c onwards (Re)generation processes expected to be negligible 18

CMS results: prompt J/ at high p T CMS PAS HIN-2012-014 CMS-PAS HIN-12-2014 Striking difference with respect to ALICE No saturation of the suppression vs centrality High-p T RHIC results show weaker suppression No significant p T dependence from 6.5 GeV/c onwards (Re)generation processes expected to be negligible 19

J/ flow The contribution of J/ from (re)combination should lead to a significant elliptic flow signal at LHC energy STAR found v 2 consistent with 0 ALICE measures v 2 (with a significance up to 3 for chosen kinematic/centrality selections) in agreement with transport models including (re)combination ALICE Coll., Phys. Rev. Lett. 111 (2013) 162301 b diffusion CMS measures a significant v 2 in a region where (re)combination should be negligible due to path-length dependence 20 of J/ suppression

Moving to p-pb J/ results: R ppb vs y LHCb Coll., JHEP 02 (2014) 072 ALICE Coll., JHEP 02 (2014) 073 ALICE and LHCb results in good agreement Strong suppression at forward and mid-y: no suppression at backward y Data are consistent with models including shadowing and/or energy loss Color Glass Condensates (CGC) inspired models underestimate data Dissociation cross section abs <2 mb cannot be excluded 21

R ppb vs p T backward-y mid-y forward-y The p T dependence of J/ R ppb has been studied in the three y ranges backward-y: negligible p T dependence, R pa compatible with unity mid-y: small p T dependence, R pa compatible with unity for p T >3GeV/c forward-y: strong R pa increase with p T Comparison with theory: Data consistent with pure shadowing calculations and with coherent energy loss models (overestimating J/ suppression at low p T, forward-y) CGC calculation overestimate suppression at forward-y 22 22

Event activity dependence: Q ppb Q J / pa T Y pa J pa J / pp At forward-y, strong J/ Q pa decrease from low to high event activity At backward-y, Q pa consistent with unity, event activity dependence not very significant 23 23

CNM effects: from p-pb to Pb-Pb x-values in Pb-Pb s NN =2.76 TeV, 2.5<y cms <4 2 10-5 < x < 9 10-5 1 10-2 < x < 6 10-2 x-values in p-pb s NN =5.02 TeV, 2.03 < y cms < 3.53 2 10-5 < x < 8 10-5 x-values in p-pb s NN =5.02 TeV, -4.46 < y cms < -2.96 1 10-2 < x < 5 10-2 Partial compensation between s NN shift and y-shift If CNM effects are dominated by shadowing R PbPb CNM = R ppb R Pbp = 0.75 ± 0.10 ± 0.12 R PbPb meas = 0.57 ± 0.01 ± 0.09 compatible within 1- Same kind of agreement in the energy loss approach which does not exclude hot matter effects which partly compensate each other F. Arleo and S. Peigne, arxiv:1407.5054 24

p T -dependence Perform the extrapolation as a function of p T pa p-pb Pb-Pb AA p-pb Pb-Pb No more agreement between Pb-Pb and CNM extrapolations High-p T suppression is not related to CNM effects At low p T CNM suppression is of the same size of the effects observed in Pb-Pb: recombination? 25

Comparing charmonia and open charm: p-pb ALICE p-pb results, mid-rapidity, p T integrated ALICE Coll., PRL 113 (2014) 232301 R ppbd = 0.85 0.05 0.11 (weighted average of p T differential points using FONLL cross section (no FONLL unc.) and R ppb (0-1)=R ppb (1-2) ) Assuming R ppb (0-1) = 0.4 R ppbd = 0.82 0.05 0.11 R ppb J/ = 0.73 0.08 0.15 Within uncertainties (and with reasonable extrapolations to p T =0), CNM effects on integrated J/ and D-mesons production 26 have the same size

Comparing charmonia and open charm: p-pb ALICE p-pb results, mid-rapidity, p T differential ALICE Coll., PRL 113 (2014) 232301 Bin-to-bin comparison less straightforward g g D D At fixed p T, gluon kinematics can be (very) different for single D and J/ g g J/

Comparing charmonia and open charm: p-pb p-going direction Single muon results available for p T > 2 GeV/c (More) difficult to extract an integrated R ppb Bin-to-bin comparison not straightforward 28

Comparing charmonia and open charm: p-pb Pb-going direction Single muon results available for p T > 2 GeV/c (More) difficult to extract an integrated R ppb Bin-to-bin comparison not straightforward 29

Comparing charmonia and open charm: Pb-Pb 0-10% R PbPbD = 0.51 0.08 0.09 (weighted average of p T differential points using FONLL cross section (no FONLL unc.) and R ppb (0-1)=R ppb (1-2) ) Assuming R ppb (0-1) = 1 R PbPbD = 0.63 0.08 0.10 ALICE, PLB 734 (2014) 314 R PbPb J/ = 0.73 ± 0.09 ± 0.06 ± 0.09 Good compatibility (especially assuming R ppb (0-1) = 1) between D and J/ Suppression and regeneration balance Warning: D S, c (not included) may 30 be enhanced in Pb-Pb

Comparing charmonia and open charm: Pb-Pb 0-10% ALICE, PLB 734 (2014) 314 Forward rapidity: results for muons with p T >4 GeV/c Extrapolation to all p T problematic R PbPb J/ = 0.56 ± 0.02 ± 0.02 ± 0.08 31

J/ in Pb-Pb: run-1 summary Evidence for smaller suppression compared to RHIC Occurrence of recombination is at present the only explanation p T -dependence of R PbPb also compatible with recombination Although qualitative interpretation looks unambiguous, the quantitative assessment of the effects at play needs refinement Values for d cc /dy evolved. At present, in the forw.-y ALICE domain: SHM 0.15 0.25 mb (y=4 and y=2.5) no shadowing Zhao and Rapp 0.5 mb empirical shad. vs no shad. Zhuang et al. 0.4 0.5 mb EKS98 shadowing Ferreiro et al. 0.4 0.6 mb + Glauber-Gribov shad. ~ ndsg(min.) > EKS98 LHC run-2 (almost) a factor 2 gain in s would it be possible to extract d cc /dy which gives the best fit to run-1 results, extrapolate to run-2 energy (FONLL?) and give predictions? Suppression persists up to the largest investigated p T Higher p T reach in run-2 increase of R PbPb? Predictions? Interesting indication for azimuthal anisotropies. Run-2 needs Experiment (much) larger statistics Theory solid predictions 32

J/ in p-pb: run-1 summary p-pb data: characterization of CNM effects in terms of shadowing plus coherent energy loss (no break-up) looks satisfactory Uncertainties on shadowing calculations are large, could one use the LHC data to better constrain shadowing? Effects are strong, R ppb ~ 0.6 at low p T and central to forward rapidity Strong influence of CNM effects in Pb-Pb in the corresponding kinematic region The simple estimate R PbPb CNM =R ppb R Pbp (inspired to a shadowing scenario) leads, once this effect is factorized out, to an even steeper p T -dependence of R PbPb Also for p-pb, run-2 energy predictions ( s~8 TeV), with parameters TUNED on run-1 results, would allow a crucial test of our understanding of the involved mechanisms 33

(2S)/J/ in Pb-Pb The (2S) yield is compared to the J/ one in Pb-Pb and in pp CMS (central events) p T >3 GeV/c & 1.6< y <2.4 (2S) less suppressed than J/ p T >6.5 GeV/c & y <1.6 (2S) more suppressed than J/ Improved agreement between ALICE and CMS data (wrt preliminary) 34

(2S) R ppb vs y cms 2S J 2S J pa pp RpA RpA J 2S pa pp ALICE Coll., JHEP12(2014)073 (2S) suppression is stronger than the J/ one and reaches a factor ~2 wrt pp Same initial state CNM effects (shadowing and coherent energy loss) expected for both J/ and (2S) Theoretical predictions in disagreement with (2S) result Other mechanisms needed to explain (2S) behaviour? 35 Final state effects related to the (hadronic) medium created in the p-pb collisions? N.B.: crossing times smaller than formation time, no nuclear break-up (Forward-y: c ~10-4 fm/c, backward-y: c ~7 10-2 fm/c)

(2S) Q ppb vs event activity The (2S) Q pa is evaluated as a function of the event activity Q pa instead of R pa due to potential bias from the centrality estimator, which are not related to nuclear effects 2S J 2S J pa pp QpA QpA J 2S pa pp with Q J pa T Y J pa mult J pa pp Rather similar (2S) suppression, increasing with N coll, for both ALICE and PHENIX results 36

J/ and (2S) Q ppb vs event activity J/ and (2S) Q pa are compared vs event activity forward-y: J/ and (2S) show a similar decreasing pattern vs event activity backward-y: the J/ and (2S) behaviour is different, with the (2S) significantly more suppressed for largest event activity classes Another hint for (2S) suppression in the (hadronic) medium? 37

(2S): run-1 summary In Pb-Pb collisions the CMS results show an enhancement of the (2S) yield, compared to J/, at intermediate p T, and a suppression at low p T The ALICE preliminary results are marginally compatible with this observation (large uncertainties, low S/B) A convincing explanation of the Pb-Pb results is still lacking In p-pb collisions a significant suppression, compared to J/, is observed The effect becomes very strong at backward rapidity, and implies sizeable final-state effects on the (2S) Formation-time vs crossing-time arguments imply that the suppression may be related to the (hadronic?) medium created in p-pb collision First theory calculations support this interpretation Run-2 is expected to yield large luminosity, mandatory for a meaningful study of (2S) in Pb-Pb 38

suppression in Pb-Pb collisions LHC is the machine for studying bottomonium in AA collisions CMS Coll., PRL 109, 222301 (2012) Main features of bottomonium production wrt charmonia: no B hadron feed-down gluon shadowing effect are smaller (re)combination expected to be smaller theoretical predictions more robust due to the higher mass of b quark with a drawback smaller production cross-section PbPb pp Clear suppression of states in PbPb with respect to pp collisions 39

suppression in Pb-Pb collisions Strong suppression of (2S) (1S) suppression compatible with suppression of excited states (50% feed-down) Sequential suppression of the three states according to their binding energy: R AA ( (1S)) = 0.56 ± 0.08 (stat) ± 0.07 (syst) R AA ( (2S)) = 0.12 ± 0.04 (stat) ± 0.02 (syst) R AA ( (3S)) <0.1 (at 95% C.L) CMS Coll., PRL 109, 222301 (2012) Suppression at LHC is stronger than at RHIC 40

Comparison of ALICE vs CMS results Comparison of ALICE (forward-y) and CMS (mid-y) results CMS CMS ALICE ALICE CMS Coll., PRL 109 (2012) 222301 ALICE Coll., PLB 738 (2014) 361 Stronger suppression at forward rapidity than at mid-rapidity, in particular for central collisions 41

MODEL Comparison with theory Evolving QGP described via a dynamical model including suppression of bottomonium states, but not CNM nor recombination 2 different initial temperature y profiles: boost invariant or Gaussian (3 tested shear viscosity) The model underestimates the measured (1S) suppression at forward-y, while it is in fair agreement with mid-y data 42

MODEL Comparison with theory Transport model accounting for both regeneration and suppression CNM effects included via an effective absorption cross section (0-2 mb) The measured R AA vs centrality is slightly overestimated by the model at forward-y, while it reproduces CMS results Constant R AA behavior vs y is not 43 supported by the data

(1S) Production in p-pb (1S) measured at forward-y by both ALICE and LHCb Compatible R pa results within uncertainties (but LHCb systematically higher) Hint for stronger suppression at forward-y (similarly to J/ ) Theoretical calculations based on initial state effects seem not to describe simultaneously forward and backward y ALICE Coll., PLB 740 (2015) 105-117 LHCb Coll., JHEP 07(2014)094 44

(ns)/ (1S) Production in p-pb Initial state effects similar for the three states p-pb Pb-Pb p-pb vs pp @mid-y: final states effects in p-pb affecting the excited states p-pb vs PbPb @mid-y : even stronger suppression of excited states in PbPb CMS Coll., JHEP 04 (2014) 103 CMS Coll., PRL 109 (2012) ALICE (and LHCb) observes: (2S)/ (1S) (ALICE) 2.03<y<3.53: 0.27±0.08±0.04-4.46<y<-2.96: 0.26±0.09±0.04 Compatible with pp results 0.26±0.08 (ALICE, pp@7tev) CMS analyses the double ratio [ (2S)/ (1S)]/[ (ns)/ (1S)] pp and finds 0.83±0.05±0.05 45

(ns)/ (1S) vs event activity Strong decrease with increasing charged particle multiplicity both in pp and p-pb production affects multiplicity? (1S) produced with more particles than excited states or multiplicity affects the? activity around the breaks the state CMS Coll., JHEP 04 (2014) 103 46 Weaker dependence when the activity estimator is in a different kinematic region with respect to the

: run-1 summary First detailed study of bottomonia in HI collisions Suppression of 1S, 2S, 3S states clearly observed More weakly bound states are more suppressed Evidence for sequential suppression Suppression of 1S state at mid-rapidity consistent with feed-down effects Rapidity dependence of 1S suppression exhibits surprising features Still not satisfactorily reproduced by models p-pb results, still significant uncertainties at forward-y, no sharp conclusions At central rapidity, evidence for final-state effects on 2S and 3S states CMS R ppb results still to be delivered Intriguing features on yields vs event activity 47

Conclusions (1) LHC run-1 has led to a very significant advance of our understanding of charmonia/bottomonia in hot matter Charmonium highlight evidence for a new mechanism which enhances the J/ yield, in particular at low p T, with respect to low-energy experiments In addition Indications for J/ azimuthal anisotropy (non-zero v 2 ) Significant final state effects on (2S) in p-pb, likely related to the (hadronic) medium created in the collision Bottomonium highlight evidence for a stronger suppression of 2S and 3S states compared to 1S. Effect not related to CNM and compatible with sequential suppression of bottomonium states In addition 1S is also suppressed (~50%). Feed-down effect only? y-dependence of 1S suppression to be understood 48

Conclusions (2) Prospects for run-2 Collect a ~1 order of magnitude larger integrated luminosity High-statistics J/ sample Comparison with run-1 AND with theoretical predictions crucial to confirm/quantify our understanding in terms of regeneration Significant (2S) sample Crucial: run-1 results exploratory (and interpretation not clear) High-statistics (1S) sample A significant increase in 1S suppression with respect to run-1 might imply that a high-t QGP is formed ( threshold scenario) Differential (2S) and (3S) results from run-1 are limited by statistics Centrality and p T -dependent studies important to assess details of sequential suppression 49

Backup 50

51

52

53

RHIC: suppression vs recombination Did we reach a consensus on the role played by recombination at RHIC? One should in principle observe J/ elliptic flow J/ should inherit the heavy quark flow J/ p T distribution should be softer (<p T2 > ) wrt pp Evidence not compelling Could weaker suppression at y=0 be due to other effects (CNM, for example)?

CMS, focus on high p T Muons need to overcome the magnetic field and energy loss in the absorber Minimum total momentum p~3-5 GeV/c to reach the muon stations Limits J/ acceptance Midrapidity: p T >6.5 GeV/c Forward rapidity: p T >3 GeV/c..but not the one (p T > 0 everywhere)

Non-zero v 2 for J/ at the LHC CMS HIN-2012-001 E.Abbas et al. (ALICE), PRL111(2013) 162301 The contribution of J/ from (re)combination should lead to a significant elliptic flow signal at LHC energy A significant v 2 signal is observed by BOTH ALICE and CMS The signal remains visible even in the region where the contribution of (re)generation should be negligible Due to path length dependence of energy loss? Expected for J/? In contrast to these observations STAR measures v 2 =0 56

Finally, the LHC is really the machine for studying bottomonium in AA collisions (and CMS the best suited experiment to do that!) 57

First accurate determination of suppression Suppression increases with centrality First determination of (2S) R AA : already suppressed in peripheral collisions (1S) (see also ALICE) compatible with only feed-down suppression? Probably yes, also taking into account the normalization uncertainty Compatible with STAR (1S+2S+3S)(but large uncorrelated errors): expected 58? Is (1S) dissoc. threshold still beyond LHC reach? Full energy

(1S) vs y and p T from CMS+ALICE Start to investigate the kinematic dependence of the suppression Suppression concentrated at low p T (opposite than for J/, no recombination here!) Suppression extends to large rapidity (puzzling y-dependence?) 59

Do not forget CNM Also in the sector, the influence of CNM is not negligible With respect to 1S, the 2S and 3S states are more suppressed than in pp but less than in Pb-Pb confirm Pb-Pb suppression as hot matter effect As a function of event activity, loosely related to centrality in ppb (and 60 surely not in pp!) smooth behaviour: to be understood!

RHIC: energy scan System size and energy dependence of R AA No appreciable dependence on both energy and system size Not trivial! Requires counterbalancing of suppression+regeneration effects over a large s-region (note however large global systematics) Warning: CNM effects (shadowing) expected to vary with s 61

Quarkonia where are we? Two main mechanisms at play 1) Suppression in a deconfined medium 2) Re-generation (for charmonium only!) at high s can qualitatively explain the main features of the results ALICE is fully exploiting the physics potential in the charmonium sector (optimal coverage at low p T and reaching 8-10 GeV/c) R AA weak centrality dependence at all y, larger than at RHIC Less suppression at low p T with respect to high p T CNM effects non-negligible but cannot explain Pb-Pb observations CMS is fully exploiting the physics potential in the bottomonium sector (excellent resolution, all p T coverage) Clear ordering of the suppression of the three states with their binding energy as expected from sequential melting (1S) suppression consistent with excited state suppression (50% feed-down) 62

Conclusions LHC: first round of observations EXTREMELY fruitful Many (most) of the heavy-quark/quarkonia related observables were investigated, no showstoppers, first physics extracted Many (most) of the heavy-quark/quarkonia related observables would benefit from more data to sharpen the conclusions full energy run, 2015-2017 upgrades, 2018 onwards RHIC: still a main actor, with upgraded detectors Lower energies: SPS, FAIR Serious experimental challenge High- B region of the phase diagram unexplored for what concerns heavy quark/quarkonia below 158 GeV/c 63

Assume R pa non-prompt = 1 From R pa incl to R pa prompt The value of R pa prompt can differ significantly from R pa prompt at large f b 64

Is the difference significant for ALICE? Exercise 1) Assume R ppb non-prompt =1 2) Plot R ppb prompt vs f b for the values of R ppb inclusive measured by ALICE 3) Plot the ALICE point at the f B value corresponding to the p T where the measurement is performed Result For ALL the p T range accessible to ALICE, the difference between R inclusive ppb and the calculated R prompt ppb is smaller than the uncertainties 65

PHENIX new systems/energies New system (Cu-Au) at old energy: Cu-going finally different! (probably not a CNM effect) A challenge to theory SPS went the other way round (from S-U to Pb-Pb ) Old system (Au-Au) at new energy: still a balancing of suppression and regeneration? Theory seems to say so. 66

PHENIX CNM p T dependence of R dau Increase vs p T at central/forward y Reminds SPS observation But different behaviour at backward rapidity Not easy to reproduce in models! First study of a charmonium excited state at collider energy Seems contradicting our previous knowledge Overall picture still not clear! 67

STAR - Bottomonium: the clean probe 3 states with very different binding energies No complications from recombination But not that easy at RHIC! and this has been split into 3 centrality bins. Compatible with 3S melting and 2S partial melting 68

Hints from theory Theory is on the data! Fair agreement, but. one model has no CNM, no regeneration the other one has both CNM and regeneration (which would be responsible for all (2S) in central events) Still too early to claim a satisfactory understanding? 69

(2S) R ppb vs y cms The (2S) suppression with respect to binary scaled pp yield can be quantified with the nuclear modification factor Can the stronger suppression of the weakly bound (2S) be due to break-up of the fully formed resonance in CNM? possible if formation time ( f ~0.05-0.15fm/c) < crossing time ( c ) arxiv:1405.3796 forward-y: c ~10-4 fm/c backward-y: c ~7 10-2 fm/c c L z D. McGlinchey, A. Frawley and R.Vogt, PRC 87,054910 (2013) break-up effects excluded at forward-y at backward-y, since f ~ c, break-up in CNM can hardly explain the very strong difference between J/ and (2S) suppressions Final state effects related to the (hadronic) medium created in the p-pb collisions? 70

Charmonia data samples ALICE L int (2011) = ~70 b -1 (2.5<y<4), ~28 b -1 ( y <0.9) Trigger: MB + 2 tracks in the muon trigger chambers (p T > 1 GeV/c) B. Abelev et al., ALICE arxiv:1311.0214. Background subtraction via like-sign or mixed-event techniques 71

Charmonia data samples CMS L int (2011) = ~150 b -1 ( y <2.4) Trigger: dimuon events at L1 (no constraints on muon momentum) CMS PAS HIN-2012-014 Use pseudo-proper decay length to estimate the b-hadron decay length N.B.: discuss only prompt production in this talk 72