Anisotropic Flow: from RHIC to the LHC

Transcription

1 Anisotropic Flow: from RHIC to the LHC Raimond Snellings The 2 nd Asian Triangle Heavy Ion Conference 13 th - 15 th October, 28 University of Tsukuba, Tsukuba, Japan

2 arxiv: [nucl-ex] 2

3 Elliptic Flow A unique hadronic probe of the early stage ε = y2 x 2 y 2 + x 2 = cos 2φ In non central collisions coordinate space configuration is anisotropic (almond shape). However, initial momentum distribution isotropic (spherically symmetric) y x py px Interactions among constituents generate a pressure gradient which transforms the initial coordinate space anisotropy into the observed momentum space anisotropy anisotropic flow Time y Time py self-quenching sensitive to early stage x px see talk ShinIchi Esumi and Maya Shimomura 3

4 Time Schematic Time Evolution freeze-out lo n ex git pa ud ns ina io l n hadronization QGP? thermalization? hard scattering z initial conditions Flow builds up during all phases, probably even contributions from initial conditions: CGC, pqcd, parton cascade, viscous hydro, hadronization, hadron cascade 4 T. Hirano; arxiv: [nucl-th] C. Nonaka; arxiv:nucl-th/7282 talk Yasushi Nara

5 Ideal hydro does well STAR Phys. Rev. Lett. 86, (21) n ch / n max (p t ) Hydro curves Huovinen EOS with phase transition Hadron gas EOS! + +! - p + p [GeV/c] STAR Phys. Rev. Lett. 87, (21) p t Ideal hydro gets ~ magnitude and species dependence The species dependence is more sensitive to the EoS 5

6 The Bulk Matter The observed particles can be characterized by a single freeze-out temperature and a common azimuthal dependent boost velocity (p t ) centrality: -8%! + +! - KS p + p " + " Cascade RHIC preliminary T = 1 MeV, # =.54c, # a =.4c and s 2 =.4 Common freeze-out curves What about early freezout? What about importance of hadronic phase? Fits from STAR Phys. Rev. Lett. 87, (21) p t [GeV/c] 6

7 RHIC Scientists Serve Up Perfect Liquid New state of matter more remarkable than predicted -- raising many new questions April 18, 25 7

8 Hydro at RHIC significant effects due to shear viscosity Derek Teaney significant uncertainties due to initial conditions ideal hydro agreement somewhat accidental ch / STAR Phys. Rev. Lett. 86, (21) n n max 8 (percent) CGC STAR non-flow corrected (est). STAR event-plane!/s= p T [GeV]!/s=.8!/s=.16!/s=.24 Matthew Luzum, Paul Romatschke arxiv: hydro+cascade, CGC.16 hydro+cascade, Glauber.14 PHOBOS(hit).12 PHOBOS(track) N part T. Hirano et al., Phys. Lett. B (26) T. Hirano et al., J.Phys.G34:S ,27

9 The Perfect Liquid? In the Low Density Limit (LDL): ɛ σ 1 S dn dy H. Heiselberg and A. M. Levy, Phys. Rev. C 59, 2716 (1999) S. A. Voloshin and A. M. Poskanzer, Phys. Lett. B 474, 27 (2) For ideal hydrodynamics: ɛ = h! / NA49: C. Alt et al., Phys. Rev. C68, 3493 (23) HYDRO limits E lab /A=11.8 GeV, E877 E lab /A=4 GeV, NA49 E lab /A=158 GeV, NA49 s NN =13 GeV, STAR s NN =2 GeV, STAR Prelim (1/S) dn ch /dy Hydro limits from P.F. Kolb, J. Sollfrank, U.W. Heinz; Phys.Rev.C62:5499,2. This figure is not understood in ideal hydrodynamics! Corrections needed: parton cascade, viscous hydro, hadron cascade... 9

10 The Hydro Limit? Use relativistic Boltzmann calculations to calculate how v2 approaches hydro as function of cross section and density CGC Au-Au Cu-Cu fit "=5.5 mb Based on these calculations the v2/ε dependence has been parameterized as function of K Try to use the data to constrain the hydro limit! v2/ε= h/(1+1.4k) h: hydro limit Knudsen number: K = λ/r The number of collisions per particle: 1/K =(σ/s)(dn/dy)cs σ = partonic cross section /! (1/S)(dN/dy)[mb -1 ] R.S. Bhalerao, J-P. Blaizot, N. Borghini and J-Y. Ollitrault; Phys.Lett.B627:49-54,25. C. Gombeaud and J-Y. Ollitrault; arxiv:nucl-th/7275 H-J. Drescher, A. Dumitru, C. Gombeaud and J-Y. Ollitrault; Phys.Rev.C76:2495,27. 1

11 1 ɛ = h 1+1.4K K = R λ = c sσ 1 S K K From LDL to Hydro where K is the Knudsen number From LDL to Hydro: dn dy ɛ = h K + K 2.. ɛ = c s σ 1 S dn dy /! hydro LDL reach about 7% of ideal hydro limit! K {4}/N wounded! {4}/CGC! {ZDCSMD}/N wounded {ZDCSMD}/CGC!! preliminary Standard! CGC! AuAu 2 GeV Glauber CGC /S dn/dy (fm ) for STAR results see talks Aihong Tang and Hiroshi Masui (D. Teaney: Boltzmann η/s =.316 T/cσn) η/s = 2-3 AdS/CFT limit H-J. Drescher, A. Dumitru, C. Gombeaud and J-Y. Ollitrault; Phys.Rev.C76:2495,27. 11

12 From LDL to Hydro /! #/s =.8 #/s =.16 #/s =.24 K K Au+Au, e = 3 GeV/fm 3, "s~2 GeV Au+Au, e = 15 GeV/fm 3, "s ~62.5 GeV Cu+Cu, e = 15 GeV/fm 3, "s ~2 GeV viscous hydro, full I-S, EOS L $ % = 3#/sT (1/S) dn ch /dy (fm -2 ) H. Song and U. Heinz; arxiv: [nucl-th] ɛ = h K + K 2.. ɛ = c s σ 1 S dn dy 12 /! K {4}/N wounded! {4}/CGC! {ZDCSMD}/N wounded {ZDCSMD}/CGC!! preliminary Standard! CGC! AuAu 2 GeV Glauber CGC /S dn/dy (fm ) H-J. Drescher, A. Dumitru, C. Gombeaud and J-Y. Ollitrault; Phys.Rev.C76:2495,27. CGC ε brings centrality dependence measured already rather close to viscous hydro

13 Viscous hydro 25 CGC!/s=1-4 data described using viscous hydro with η/s between AdS/CFT lower bound.8 and.16 (percent) STAR non-flow corrected (est). STAR event-plane!/s=.8!/s=.16 need to compare with matching ideal hydro limit (EoS) p T [GeV]!/s=.24 Matthew Luzum, Paul Romatschke arxiv: Compatible estimates of η/s from viscous hydro and Boltzmann cascade! 13

14 From LDL to Hydro see talks Aihong Tang and Hiroshi Masui STAR Preliminary Different species reach same fraction of ideal hydro limit! different species decouple on average at the same time? 14

15 ALICE at the LHC first collision 12/9/28 15

16 LHC: the multiplicity /2! 8 6 AGS: Au+Au SPS: Pb+Pb PHOBOS: Au+Au LHC LHC part dn ch /d#/"n 4 2 PHOBOS: Cu+Cu $ln(s) fit PHOBOS data pp RHIC s NN dn for d# s=5.5tev Extrapolated from Lower Energies AuAu 2 GeV AuAu 13 GeV N. Armesto, C.A. Salgado and U.A. Wiedemann 1 8 N part =36 AuAu 19.6 GeV CuCu 2 GeV CuCu 62 GeV CuCu 22 GeV multiplicities: dn d# 6 N part =25 longitudinal scaling ~ 12, 4 2 N part =1 N part = # W Busza saturation ~15 and HIJING/BB ~ 35 16

17 Ideal hydro at the LHC? SPS (EOS Q) (EOS H)!! v " # # (EOS Q) Pb+Pb, b=7fm RHIC total pion multiplicity density at y= P.F. Kolb, J. Sollfrank, U.W. Heinz; Phys.Rev.C62:5499,2. LHC 1!! v " # # /! Au+Au Charged, b=6.3fm 1 T. Hirano CGC+hydro, T =1MeV th CGC+hydro, T =169MeV CGC+hydro+cascade 2 1 RHIC 1 3 th s NN LHC 1 [GeV] 4 v2 increase of ~ -2% careful do not connect the multiplicity dependence in the hydro plot to centrality dependence at fixed collision energy! 17

18 LDL at the LHC?.6 PbPb Extrapolated to 5.5TeV (4% Most Central) AGS (E877) NA49 cumul NA49 std/mod CERES Phobos STAR Phenix LHC GeV 62.4 GeV 13. GeV V GeV W Busza η s v2 increase of ~ 5% 18

19 Centrality dependence! hydro cs 2 =.22 hydro2 cs 2 = hydro2 LDL hydro dn ch /d! E. Simili, PhD thesis Only centrality dependence can determine which picture is more correct! 19

20 The LHC hydro limit? /! K {4}/N wounded! {4}/CGC! {ZDCSMD}/N wounded {ZDCSMD}/CGC!! preliminary Standard! CGC! AuAu 2 GeV Glauber CGC LHC /S dn/dy (fm ) Fraction of Hydro Limit N wounded N wounded! constituent quark!.2 CGC! N wounded + N bin (Hydro)! Hydro /S dn/dy only centrality dependence is predicted (knowing ε) not the absolute magnitude (magnitude depends on effective cs) Measurement of the centrality dependence will allow us to discriminate between this, ideal hydro and LDL 2

21 Fluctuations and nonflow M. Miller and RS, arxiv:nucl-ex/3128 {2} = ( 2 + σ 2 v + δ) two-particle correlation methods (v2{ep}, v2{2}) measure flow in participant plane (+ nonflow) PHOBOS QM25: Nucl. Phys. A774: 523 (26) multi-particle methods ({4} and higher, v2{lyz}) and methods using the spectators measure flow in the reaction plane and in addition remove the nonflow R.S. Bhalerao, J-Y. Ollitrault Phys.Lett.B641:26-264,26 S.A. Voloshin, A.M. Poskanzer, A. Tang, G. Wang Phys.Lett.B659: ,28 21 (%) of all particles (! <2.5) (%) (! <2.5) {2} {4} {6} UrQMD(2.2) Au+Au s 1/2 =2AGeV % Most Central UrQMD(2.2) Au+Au s 1/2 =2AGeV (2-3%) {2} {4} {6} P T (GeV/c) X Zhu, M. Bleicher, H. Stoeker, Phys.Rev.C72:64911(25)

22 Flow and nonflow compare twoparticle azimuthal correlations in pp, da and AA in most peripheral and most central collisions and at high-pt the correlation is very similar in the three systems ))! i - " cos (2(" pt # i $ 1.1 Au+Au 1%-8% d+au Minbias p+p Minbias Au+Au 6%-2% d+au Minbias p+p Minbias Au+Au top 5% d+au Minbias p+p Minbias p t (GeV/c) J. Adams et al. [STAR Collaboration], Phys. Rev. Lett. 93, (24) 22

23 Flow and nonflow centrality 5-6% CuCu centrality 4-5% centrality 3-4% (GeV/c) centrality 2-3%.35.3 p t (GeV/c) centrality 1-2%.35.3 preliminary p t (GeV/c) centrality -1%.35.3 p t S.Voloshin QM (GeV/c) p t (GeV/c) p t (GeV/c) p t A rapidity gap (Δη=2-3) is also not always sufficient! 23

24 ALICE detectors at forward rapidity PMD, FMD (2 < η < 5, -1.7 < η <-3.4) ZDCs multiplicity large enough for multiparticle correlations cumulants Lee Yang Zeroes Lee Yang Zeroes + event plane 24

25 RHIC Scientists Serve Up Perfect Liquid New state of matter more remarkable than predicted -- raising many new questions April 18, 25 now better quantitative understanding viscous correction are important! in early hydro phase or hadronic phase initial conditions (e.g. ε, initial flow fields) not sufficiently constrained! current estimates of η/s are ~ 2x the conjectured lower bound from AdS/ CFT at the LHC elliptic flow will again be one of the first results coming from heavyions centrality dependence of elliptic flow at the LHC can easily kill or confirm the perfect fluid picture discovered at RHIC the answer to the question how much of the viscosity comes from the hadronic or partonic phase can come from identified particles v2 like the Φ 25

26 Arigatou gozaimasu Thank you 26

27 Initial Conditions? initial flow fields x radial flow Peter F. Kolb, Ralf Rapp Phys.Rev. C67 (23) 4493 z faster initial expansion hydrodynamics + statistical hadronization Mikolaj Chojnacki, Wojciech Florkowski, Wojciech Broniowski, Adam Kisiel x z F. Becattini, F. Piccinini, J. Rizzo arxiv:

28 Initial Conditions? directed flow as function of collision energy and system size reduces to a single distribution when plotted versus η-ybeam none of the available models so far provides a description for this connection to the initial conditions? (%) v y CM = 3% - 6% Data AMPT 2 GeV: Au +Au 2 GeV: Cu +Cu 62.4 GeV: Au +Au 62.4 GeV: Cu +Cu ! - y beam 28

29 25 2 CGC STAR non-flow corrected (est). STAR event-plane!/s=1-4!/s= Glauber STAR non-flow corrected (est.) STAR event-plane!/s=1-4!/s=.8 (percent) 15 1!/s=.16 (percent) 15 1!/s=.16 5!/s= p T [GeV] ideal (22), 1/4π ~ 3% reduction (16), 2/4π ~ 45% reduction (12) p T [GeV] ideal (15.5), 1/4π ~ 25% reduction (12), 2/4π ~ 4% reduction (9.5) 29

30 Fluctuations and Planes y y RP the reaction plane, defined by the impact parameter PP the participant plane, defined by the major axis of the created system! PP x x! RP PHOBOS QM25: Nucl. Phys. A774: 523 (26) Fluctuations change the angle of the symmetry plane 3

31 Initial Eccentricity? eccentricity N part N part scaling N coll scaling KLN (running! s ) KLN " (KLN) (running! s ) " (KLN) DHJ Dipole.2.18 hydro+cascade, CGC.16 hydro+cascade, Glauber.14 PHOBOS(hit).12 PHOBOS(track) N part H-J. Drescher et al., Phys.Rev.C74:4495,26 CGC initial conditions give a much (up to 5%) larger ε reduces /ε below the ideal hydro limit The perfect liquid conclusion depends on the choice of ε! 31 T. Hirano et al., Phys. Lett. B (26) T. Hirano et al., J.Phys.G34:S ,27

32 Higher Harmonics Ratio v 4/v2 2 is sensitive to the degree of thermalization Updated analysis from QM6 with smaller systematic uncertainties The Boltzmann curves are from the same calculation as used for v2/ε, which for most central collisions reached K~.5 Also these measured ratios indicate that ideal hydro limit is not yet reached 2 v 4 / preliminary 2 v 4 / Hydro Boltzmann K=.1 Boltzmann K=.5 Boltzmann K= 1. AuAu 2 GeV 2-3% (GeV/c) Yuting Bai PhD thesis, Jocelyn Mlynarz [ Boltzmann curves J-Y. Ollitrault [ p t 32

33 Elliptic Flow at RHIC strong elliptic flow constituent quark degrees of freedom large energy loss ch / n Hydro curves Huovinen EOS with phase transition.8 Hadron gas EOS.7.6! + +! -.5 p + p n max [GeV/c] (p t ) p t minimum-bias PRELIMINARY Run-4 Run-7.1 (a)! + +! - (PHENIX) + - K +K (PHENIX) K (STAR) S p+p (PHENIX) "+" (STAR) + # - +# (STAR) (b) (%) 2 v 25 2 preliminary /n q Rapp & van Hees, PRC 71, 3497 (25) p T /n q (GeV/c) KE T /n q (GeV) 5 centrality: 2-6% AuAu 2 GeV v {EP} 2 v {2} 2 v {4} p t (GeV/c) 33