Charm production at RHIC

Transcription

1 1 Charm production at RHIC Charm 2007 Conference Cornell University, Ithaca, NY 5 August 2007

2 2 The Quark Gluon Plasma T c Early universe quark-gluon plasma LHC RHIC Tri-critical point? Quark deconfinement possible with sufficient energy density Transition between hadron gas and QGP T C ~ 160 MeV 175 MeV Temperature nucleon gas hadron gas Heavy-Ion Collision nuclei color superconductor Neutron stars CFL vacuum µ B Baryon Chemical Potential see: Alford, Rajagopal, Reddy, Wilczek Phys. Rev. D64 (2001)

3 3 Relativistic Heavy Ion Collider Long Island 2 concentric rings 3.8 km circumference counter-circulating beams p, p : 5 GeV s 500 GeV d, Cu, Au: 5 GeV s NN 200 GeV

4 4 Collision Geometry Centrality: overlap between nuclei N part : number of participant nucleons N bin : number of binary collisions N bin N part /2 Measured through: charged particle multiplicity (N ch ) number of spectator neutrons (N part and N bin from Glauber model calculations) into screen out of screen Central Small distance between centers of nuclei N part, N bin, N ch large Peripheral Large distance between centers of nuclei N part, N bin, N ch small

5 5 Collision Geometry N ch ~ 4000 STAR: Au+Au 200 GeV into screen out of screen Central Small distance between centers of nuclei N part, N bin, N ch large Peripheral Large distance between centers of nuclei N part, N bin, N ch small

6 6 High-p T Particle Suppression Sensitive to medium properties Partons lose energy in medium Gluon Radiation parton PHENIX: Au + Au 200 GeV, central High-p T (> 2 GeV/c) particles suppressed More suppression expected in central collisions Compare to p + p (no medium) Nuclear Modification Factor (R AA ) N bin -scaled ratio of particle yields No Medium Effect: R AA = 1 (for p T > 2 GeV/c) R AA ( p ) T hot and dense medium = Yield Yield ( A + A) ( p + p) N bin γ not suppressed not strongly interacting Hadrons suppressed in central collisions Consistent with formation of QGP Other Tests: Jet Quenching, Elliptic Flow

7 7 Heavy Flavor and the QGP Heavy quarks produced in initial hard scattering of partons Dominant: gg QQ Production rates from pqcd Sensitive to initial gluon distributions Heavy quark energy loss Prediction: less than light quark parton energy loss (dead cone effect) ENERGY LOSS dead cone light bottom hot and dense medium Sensitive to gluon densities in medium Quarkonium Suppression M.Djordjevic PRL 94 (2004)

8 8 Measuring Heavy Flavor Study hadronic decays: D 0 Kπ, D* D 0 π, D ± Kππ, D s± πφ and semileptonic decays: c µ + + anything (B.R.: ~7%) c e + + anything (B.R.: 9.6%) D 0 e + + anything (B.R.: 6.87%) D ± e ± + anything (B.R.: 17.2%) b e + + anything (B.R.: 10.9%) B ±,B 0 e ± + anything (B.R.: 10.2%) Heavy flavor decays dominate non-photonic (single) e ± spectrum; b decays dominate at high p T Photonic e ± background: γ conversions (π γγ, γ e + e - ) Dalitz decays of π 0, η, η ρ, φ, K e3 decays (small contributions) non-photonic e ±

9 9 Non-photonic e ± Remove Photonic e ± Background Combine e ± with oppositely Simulate background e ± from charged tracks in same event; cocktail of measured e ± is background if M inv < 150 sources (γ,π 0,η, etc.) MeV/c 2 Measure e ± with converter, extrapolate to 0 background STAR: Non-photonic e ±, s NN =200 GeV PHENIX: Non-photonic e ±, s NN =200 GeV Au+Au: 0-5% p+p 10-40% 40-80% d+au STAR: B. I. Abelev et al, Phys. Rev. Lett. 98 (2007) PHENIX: A. Adare et al, Phys. Rev. Lett. 98, (2007) Au+Au: 0-92% 0-10% 10-20% 20-40% 40-60% 60-92% p+p

10 10 Perturbative Calculations Heavy quark production can be calculated perturbatively FONLL: Fixed-Order plus Next- Leading Log resummed At what p T do b decays begin to dominate? Within 3 GeV/c < p T < 10 GeV/c Most Likely Value: ~5 GeV/c FONLL: c and b decay contributions to non-photonic e ± cross-section Fraction of Total (c+b) Cross-Section Cacciari, Nason, and Vogt, Phys. Rev. Lett. 95 (2005)

11 11 Disentangling Charm and Bottom Azimuthal distribution of hadrons w.r.t. non-photonic e ± e ± can get bigger kick from B decay: broader same-side peak Also: e ± correlation w/ identified D 0 h correlations with non-photonic e ± D B p+p 200 GeV X. Lin, Strange Quark Matter Conference (2007)

12 12 Disentangling Charm and Bottom Azimuthal distribution of hadrons w.r.t. non-photonic e ± e ± can get bigger kick from B decay: broader same-side peak Also: e ± correlation w/ identified D 0 h correlations with non-photonic e ± PYTHIA Simulation: D and B contributions Fit determines D vs. B ϕ = 1 R ϕ + R ϕ fit ( ) D B B-fraction consistent with FONLL Fractional Contribution of B B D p+p 200 GeV X. Lin, Strange Quark Matter Conference (2007)

13 13 Charm Cross Section D 0, µ ±, and e ± invariant yields µ ± D 0 PHENIX: σ cc from non-photonic e ± STAR: combined fit: D 0 Kπ µ ± non-photonic e ± (TOF used) covers ~95% of cross-section Scales with N bin : charm produced in initial hard scattering No thermal production STAR σ cc > FONLL calculation e ± σ cc /N bin C. Zhong, J. Phys. G 34 (2007) S741-S744 A. Suaide, Quark Matter Conference (2006)

14 14 Nuclear Modification Factor PHENIX and STAR: R AA for Non-photonic e ± central Au+Au, s NN =200 GeV R AA for non-photonic e ± : PHENIX consistent with STAR Similar to light hadron R AA Models tend to under-predict suppression radiative and/or collisional energy loss are insufficient Light Hadron R AA R AA ( p ) T Yield = Yield ( A + A) ( p + p) N bin PHENIX: PRL 98 (2007) STAR: PRL 98 (2007) DVGL: Djordjevic, Phys. Lett. B 632 (2006) 81 BDMPS: Armesto, Phys. Lett. B 637 (2006) 362

15 15 Nuclear Modification Factor PHENIX and STAR: R AA for Non-photonic e ± central Au+Au, s NN =200 GeV R AA for non-photonic e ± : PHENIX consistent with STAR Similar to light hadron R AA Models tend to under-predict suppression radiative and/or collisional energy loss are insufficient Light Hadron R AA R AA ( p ) T Yield = Yield ( A + A) ( p + p) N bin PHENIX: PRL 98 (2007) STAR: PRL 98 (2007) DVGL: Wicks, nucl-th/ (2005) van Hees, Phys. Rev. C (2006)

16 16 Nuclear Modification Factor PHENIX and STAR: R AA for Non-photonic e ± central Au+Au, s NN =200 GeV R AA for non-photonic e ± : PHENIX consistent with STAR Similar to light hadron R AA Models tend to under-predict suppression radiative and/or collisional energy loss are insufficient b only important at higher p T? Light Hadron R AA R AA ( p ) T Yield = Yield ( A + A) ( p + p) N bin PHENIX: PRL 98 (2007) STAR: PRL 98 (2007) DVGL: Djordjevic, Phys. Lett. B 632 (2006) 81

17 17 Nuclear Modification Factor PHENIX and STAR: R AA for Non-photonic e ± central Au+Au, s NN =200 GeV R AA for non-photonic e ± : PHENIX consistent with STAR Similar to light hadron R AA Models tend to under-predict suppression radiative and/or collisional energy loss are insufficient b only important at higher p T? Collisional dissociation of heavy flavor mesons? Models still being refined Light Hadron R AA R AA ( p ) T Yield = Yield ( A + A) ( p + p) N bin PHENIX: PRL 98 (2007) STAR: PRL 98 (2007) Adil and Vitev, Phys. Lett. B 649 (2007) 139

18 18 Summary and Outlook Charm Cross-Section cc production scales as N bin Indicates charm production in initial state STAR disagrees with PHENIX and FONLL calculations Non-photonic e ± R AA Suppression similar to light hadrons Difficult for models to describe e ± -h correlations b-fraction consistent with FONLL For more information about heavy-ion physics, visit

19 19 Summary and Outlook Charm Cross-Section cc production scales as N bin Indicates charm production in initial state STAR disagrees with PHENIX and FONLL calculations Non-photonic e ± R AA Suppression similar to light hadrons Difficult for models to describe e ± -h correlations b-fraction consistent with FONLL The Future at PHENIX Hadron Blind Detector The Future at STAR Low Material Run Reduce photonic e+ bkg. Improvements in data recording Improved Track Resolution Near Vertex Better heavy flavor decay reconstruction Disentangle charm and bottom The Future at RHIC RHIC II Luminosity Upgrade For more information about heavy-ion physics, visit

20 20 Additional Material

21 21 Experiments at RHIC Silicon Vertex Tracker ( η < 1) Time Projection Chambers φ = 2π ( η < 1.8, 2.5 < η < 4) EM Calorimeters (-1 < η < 2) Time-Of-Flight (-1 < η < 0, φ = π/30) Covers large Ω ~500 Collaborators 2 NIM central A499 arms ( η < (2003) 0.35) Drift Chamber ( φ = 90 2) Time Expansion Chamber ( φ = 90 ) RICH ( φ = 90 2) EM Calorimeters ( φ = 90 2) Time-Of-Flight ( φ = 45 ) 2 forward arms (µ ± ID) < y < -1.15; 1.15 < y < 2.44 ~500 Collaborators

22 22 e ± Identification e - EMC: p/e < 2 SMD: cluster sizes 2 TPC: 3.5 kev/cm < de/dx < 5 kev/cm NIM A (2003) EMC: E/p 1 > -2σ EMC: shower shape DC: tracks match EMC showers RICH: 3 associated hits

23 23 Quarkonium Suppression QGP Debye color screening suppressed heavy quarkonia Different resonances may have different dissociation temperatures: Spectrum of quarkonia could be a thermometer for the medium (eventually) Suppression of heavier resonances (i.e. χ c ) can still suppress J/Ψ yield c Quarkonium Dissociation Temperatures (Digal, Karsch, and Satz) Color Screening T d /T c Ψ (2s) 1.12 χ c (1p) 1.16 Υ(3s) 1.17 χ b (2p) 1.19 Υ(2s) 1.60 χ b (1p) 1.76 J/Ψ 2.10 Υ(1s) >4.0 c Just one calculation; others disagree

24 24 J/Ψ Suppression Suppression (R AA ) similar for: NA50 (SPS) Pb + Pb, s NN =17.2 GeV PHENIX Au + Au, s NN =200 GeV Suppression-only models: describe SPS suppression well over-predict RHIC suppression Regeneration of J/Ψ : Gluon density at RHIC 2-3 times greater than at SPS More cc pairs Quarks from dissociated cc pairs can recombine into new pairs J/Ψ R AA vs. N part NA50 at SPS (0<y<1) PHENIX at RHIC ( y <0.35) Bar: uncorrelated error Bracket : correlated error Global error = 12% is not shown NA50: M. C. Abreu et al, Phys. Lett. B 477 (2000) 28 PHENIX: R. Averbeck, J. Phys. G 34 (2007) S567-S574

25 25 Cold Nuclear Matter Effects HOWEVER, J/Ψ also suppressed due to CNM effects Dissociation/Absorption Cronin Effect (multiple gluon scattering) Shadowing (depletion of lowmomentum gluons) Gluon Saturation (Color Glass Condensate) d + Au: model (absorption + shadowing): σ abs < 3 mb Need more d + Au data 0 mb 3 mb Low x 2 ~ (shadowing region)

26 26 Cold Nuclear Matter Effects HOWEVER, J/Ψ also suppressed due to CNM effects Dissociation/Absorption Cronin Effect (multiple gluon scattering) Shadowing (depletion of lowmomentum gluons) Gluon Saturation (Color Glass Condensate) d + Au: model (absorption + shadowing): σ abs < 3 mb Need more d + Au data Au + Au: same model underpredicts suppression Need QGP? forward rapidity (µ + µ - ) mid-rapidity (e + e - )

27 27 J/Ψ Suppression R AA for Au + Au similar to SPS NA50 (Pb + Pb) Models describe SPS data well, but not RHIC 2-3 times greater gluon density at RHIC: more cc pairs pairs recombine regeneration of J/Ψ All models for y=0 J/ψ,ψ,χ c nucl-ex/ Digal, Fortunato, and Satz: hep-ph/ Capella and Ferreiro: hep-ph/ Grandchamp, Rapp, and Brown: hep-ph/ Rapp Satz Capella

28 28 J/Ψ Suppression R AA for Au + Au similar to SPS NA50 (Pb + Pb) Models describe SPS data well, but not RHIC 2-3 times greater gluon density at RHIC: more cc pairs pairs recombine regeneration of J/Ψ Yan model shows less suppression Models: mid-rapidity suppression greater than forward nucl-ex/ Yan, Zhuang, and Xu: nucl-th/

29 29 Elliptic Flow Yields depend on orientation w.r.t reaction plane Azimuthal anisotropy in medium Unequal pressure gradients in and out of reaction plane Measured by v 2 (from Fourier expansion) Yield + v cos 2 φ Ψ + v cos 4 φ ( ) ( Ψ ) + Κ 1 2 lab plane 4 lab plane φ lab Ψ plane Ψ plane Reaction Plane x

30 30 Elliptic Flow Yields depend on orientation w.r.t reaction plane Azimuthal anisotropy in medium Unequal pressure gradients in and out of reaction plane Measured by v 2 (from Fourier expansion) Yield 1+ v2 cos 2 φlab Ψ plane + v4 cos 4 φ STAR observes large asymmetry ( ) ( Ψ ) + Κ STAR: π ± v 2 for Au + Au 200 GeV (peripheral) (central) lab plane

31 31 Elliptic Flow STAR: light hadron v 2 vs. p T suggest QGP obeys ideal hydrodynamics (0 viscosity) QGP is perfect liquid? STAR: Au + Au 200 GeV, 0-80% most central

32 32 Elliptic Flow STAR: light hadron v 2 vs. p T suggest QGP obeys ideal hydrodynamics (0 viscosity) QGP is perfect liquid? PHENIX: flow of non-photonic e ± Data suggest low viscosity for charm PHENIX: Au + Au 200 GeV, 0-10% most central

33 33 Charm Cross-Section pp πp

34 34 PHENIX Non-photnic e ± N e Electron yield Dalitz 0.8% detector converter 0.4% 1.7% With converter Photonic W/O converter Dalitz decays Non-photonic 0 Dalitz : 0.8% X 0 equivalent radiation length Χ 0 (Material amount, %) F. Kajihara, Quark Matter Conference (2006)