Electromagnetic Probes in Ultrarelativistic Heavy-Ion Collisions Hendrik van Hees Justus-Liebig Universität Gießen September 1, 2009 Institut für Theoretische Physik JUSTUS-LIEBIG- UNIVERSITÄT GIESSEN Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 1 / 27
Outline 1 Electromagnetic probes in heavy-ion collisions Vector mesons and electromagnetic probes Sources of dileptons 2 Comparison data (SPS+RHIC) Invariant-mass spectra m T spectra and slope analysis 3 Conclusions and Outlook Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 2 / 27
Electromagnetic probes in heavy-ion collisions γ, l ± : no strong interactions reflect whole history of collision: from pre-equilibrium phase from thermalized medium and hot hadron gas from VM decays after thermal freezeout π,... ρ/ω γ e dn/(dy dm) π 0, η Dalitz decays ρ/ω Φ DD J/ Ψ ψ Drell Yan Low Intermediate High Mass Region > 10 fm > 1 fm < 0.1 fm e + 0 1 2 3 4 5 M [GeV/c] Fig. by A. Drees γ Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 3 / 27
Vector Mesons and electromagnetic Probes q 0 photon and dilepton thermal emission rates given by same electromagnetic-current-correlation function (J µ = f Q f ψ f γ µ ψ f ) [L. McLerran, T. Toimela 85, H. A. Weldon 90, C. Gale, J.I. Kapusta 91] Π < µν(q) = d 4 x exp(iq x) J µ (0)J ν (x) T = 2f B (q 0 ) Im Π (ret) µν (q) α 2π 2 gµν Im Π (ret) µν (q) f B(q 0 ) q0 = q dn γ d 4 xd 3 q = dn e + e d 4 xd 4 q = α 2 gµν 3q 2 Im Π(ret) π3 µν (q) f B (q 0 ) q 2 =M 2 e + e to lowest order in α: e 2 Π µν Σ (γ) µν vector-meson dominance model: Σ γ µν = G ρ derivable from partition sum Z(V, T, µ, Φ)! Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 4 / 27
Hadronic many-body theory HMBT for vector mesons [Ko et al, Chanfray et al, Herrmann et al, Rapp et al,...] ππ interactions and baryonic excitations ρ π B *, a 1, K 1,... ρ ρ ρ π N, K, π,... +corresponding vertex corrections gauge invariance Baryon (resonances) important, even at RHIC with low net baryon density n B n B reason: n B +n B relevant (CP inv. of strong interactions) Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 5 / 27
In-medium spectral functions and baryon effects [R. Rapp, J. Wambach 99] baryon effects important large contribution to broadening of the peak responsible for most of the strength at small M Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 6 / 27
Dilepton rates: Hadron gas in-medium hadron gas matches with similar results also for γ rates quark-hadron duality!? indirect evidence for chiral-symmetry restoration Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 7 / 27
Sources of dilepton emission in heavy-ion collisions 1 initial hard processes: Drell Yan 2 core emission from thermal source [McLerran, Toimela 1985] 1 dn (thermal) = d 4 dn (thermal) x dy Mdϕ q T dmdq T d 4 xd 4 Acc(M, q T, y) q 3 corona emission from primordial mesons (jet-quenching) 4 after thermal freeze-out emission from freeze-out mesons [Cooper, Frye 1975] d N (fo) 3 q = q 0 q µ dσ µ f B (u µ q µ /T ) Γ meson l + l Acc Γ meson additional factor γ = q 0 /M compared to thermal emission physical reason thermal source rate τ med Γ meson l + l γ decay of mesons after fo: rate Γ meson l + l Γ meson Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 8 / 27
Intermediate masses: hadronic 4π contributions e.m. current-current correlator τ 2nπ -Im Π V,A /(π M 2 ) 0.08 0.06 0.04 V [τ 2n π ν τ ] V [τ 2 π ν τ ] ρ + cont. 2π (ρ) 4π (fit) -Im Π V,A /(π M 2 ) 0.04 0.03 0.02 A [τ (2n+1) π ν τ ] A [τ 3 π ν τ ] a 1 + cont. 3π (a 1 ) 5π (fit) 0.02 0.01 0 1.6 1.8 2 M 2 (GeV 2 ) 0 1.6 1.8 2 M 2 (GeV 2 ) 4π contributions : π + ω, a 1 µ + + µ leading-order virial expansion for four-pion piece additional strength through chiral mixing Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 9 / 27
Radiation from thermal sources: Meson t-channel exchange motivation: q T spectra too soft compared to data thermal contributions not included in models so far l π ρ γ l + ω ρ π also for π, a 1 Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 10 / 27
Excess spectra dn µµ /dm (counts) 2000 1500 1000 500 Fireball with standard EoS-A (T c = T chem = 175 MeV) overall normalization fireball lifetime relative normalization of thermal radiation fixed by rates rates integrated over time, volume, q including acceptance semicentral In-In all q T T c =T ch =175 MeV +DD FO + sum dn µµ /dm (counts) 1200 1000 800 600 400 200 central In-In all q T T c =T ch =175 MeV +DD FO+ sum 0 good description of data 0 Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 11 / 27
Excess spectra: IMR and multi-pion contributions dn µµ /dm (counts) 1200 1000 800 600 400 semicentral In-In all q T DD sum () 4π vac sum (4π vac) T c =T ch =175 MeV 200 0 1 1.2 1.4 4π contributions (π + ω, a 1 µ + + µ ) slightly enhanced by VA mixing Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 12 / 27
CERES/NA45 dielectron spectra (dn ee /dm) / <N ch > [100 MeV/c 2 ] -1 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 good agreement also for dielectron spectra in 158 GeV Pb-Au allows further check of low-mass tail from baryon effects down to M 2m e 10 6 1.8 Pb(158 AGeV)+Au (7% central) <N ch >=335 CERES 00 +ω+φ (no bar ρ) p t >0.2GeV 2.1<η<2.65 Θ ee >35mrad 0 0.2 0.4 0.6 0.8 1 1.2 M ee [GeV] (d 2 N ee /dηdm) / (dn ch /dη) [100MeV] -1 10-5 10-6 10-7 35% Central Pb(158AGeV)+Au <N ch >=250 Cocktail CERES 95+ 96 free ρ +ω+φ++4π p t >0.2GeV 2.1<η<2.65 Θ ee >35mrad 4π 10-8 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 M ee [GeV] Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 13 / 27
Excess spectra: acceptance-corrected mass spectra ZD: [K. Dusling, D. Teaney, I. Zahed 2007], RR: [J. Ruppert, T. Renk et al 2008], RH: [HvH, R. Rapp 2008] Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 14 / 27
Importance of baryon effects Baryonic interactions important! in-medium broadening low-mass tail! (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm all q T FO ρ φ ω ω-t ex +ω-t ex (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm all q T FO ρ φ ω ω-t ex +ω-t ex Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 15 / 27
Sensitivity to T c and hadro-chemistry recent lattice QCD: T c 190-200 MeV or T c 150-160 MeV? thermal-model fits to hadron ratios: T chem 150-160 MeV T (GeV) 0.2 0.19 0.18 0.17 0.16 0.15 0.14 0.13 0.12 0.11 t fo =6.7 fm/c EoS-A EoS-B EoS-C 0 1 2 3 4 5 6 7 t (fm/c) µ π (GeV) 0.1 0.08 0.06 0.04 0.02 EoS-A EoS-B EoS-C t 0 fo =6.7 fm/c 0 1 2 3 4 5 6 7 t (fm/c) EoS-A: T c = T chem = 175 MeV EoS-B: T c = T chem = 160 MeV EoS-C: T c = 190 MeV, T chem = 160 MeV T c T T chem : hadron gas in chemical equilibrium keep fireball parameters the same (including life time) Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 16 / 27
EoS-B dn µµ /dm (counts) 2000 1500 1000 semicentral In-In all q T T c =T ch =160 MeV +DD FO + sum dn µµ /dm (counts) 1000 800 600 400 semicentral In-In q T <0.5 GeV T c =T ch =160 MeV +DD FO + sum dn µµ /dm (counts) 1000 800 600 400 semicentral In-In q T >1.0 GeV T c =T ch =160 MeV +DD FO + sum 500 200 200 1/q T dn µµ /(dq T dy) (counts) 10 8 10 7 10 6 10 5 10 4 0 semicentral In-In 0.4 GeV<M<0.6 GeV incl. FO+ 1/q T dn µµ /(dq T dy) (counts) 10 8 10 7 10 6 10 5 10 4 0 semicentral In-In 0.6 GeV<M<0.9 GeV incl. FO ρ 1/q T dn µµ /(dq T dy) (counts) 10 7 10 6 10 5 10 4 10 3 0 semicentral In-In 1.0 GeV<M<1.4 GeV incl. FO ρ T c =T ch =160 MeV 10 3 0 0.5 1 1.5 2 2.5 m T - T c =T ch =160 MeV 10 3 0 0.5 1 1.5 2 2.5 m T - T c =T ch =160 MeV 10 2 0 0.5 1 1.5 2 m T - mass spectra comparable to EoS-A slight enhancement of fireball lifetime in IMR > multi-pion contribution higher hadronic temperatures slightly harder q T spectra not enough to resolve discrepancy with data Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 17 / 27
EoS-C dn µµ /dm (counts) 2000 1500 1000 semicentral In-In all q T T c =190 MeV T ch =160 MeV +DD FO + sum dn µµ /dm (counts) 1000 800 600 400 semicentral In-In q T <0.5 GeV T c =190 MeV T ch =160 MeV +DD FO + sum dn µµ /dm (counts) 1000 800 600 400 semicentral In-In q T >1.0 GeV T c =190 MeV T ch =160 MeV +DD FO + sum 500 200 200 1/q T dn µµ /(dq T dy) (counts) 10 8 10 7 10 6 10 5 10 4 10 3 0 semicentral In-In 0.4 GeV<M<0.6 GeV T c =190 MeV, T ch =160 MeV 0 0.5 1 1.5 2 2.5 m T - incl. FO+ 1/q T dn µµ /(dq T dy) (counts) 10 8 10 7 10 6 10 5 10 4 10 3 0 T c =190 Mev T ch =160 MeV semicentral In-In 0.6 GeV<M<0.9 GeV 0 0.5 1 1.5 2 2.5 m T - incl. FO ρ 1/q T dn µµ /(dq T dy) (counts) 10 7 10 6 10 5 10 4 10 3 10 2 0 semicentral In-In 1.0 GeV<M<1.4 GeV T c =190 MeV, T ch =160 MeV 0 0.5 1 1.5 2 m T - incl. FO ρ mass spectra comparable to EoS-A slight reduction of fireball lifetime in IMR multi-pion contribution higher hadronic temperatures + high-density hadronic phase harder q T spectra better agreement with data Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 18 / 27
standard fireball acceleration: too soft q T spectra lower T c in EoS-B and EoS-C helps (higher hadronic temperatures) NB: here, Drell Yan contribution taken out Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 19 / 27 Inverse-slope analysis to extract T eff fit to 1 dn = 1 q T dq T m T ( dn = C exp m ) T dm T T eff fit of theoretical q T spectra: 1 GeV < q T < 1.8 GeV 300 300 250 250 T eff (MeV) 200 T eff (MeV) 200 EoS-A 150 EoS-B NA 60 NA 60 LMR NA 60 IMR no 100 EoS-A 150 EoS-C NA 60 NA 60 LMR NA 60 IMR no 100
Inverse-slope analysis 300 300 250 250 T eff (MeV) 200 T eff (MeV) 200 EoS-B 150 EoS-B, a =0.1c 2 /fm NA 60 NA 60 LMR NA 60 IMR no 100 EoS-C 150 EoS-C, a =0.1c 2 /fm NA 60 NA 60 LMR NA 60 IMR no 100 enhance fireball acceleration to a = 0.1c 2 /fm effective at all stages of fireball evolution agreement in IMR not spoiled dominated from earlier stages EoS-B harder relative contribution of harder freezeout ρ decays vs. thermal ρ s larger Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 20 / 27
Inverse-slope analysis 300 300 250 250 T eff (MeV) 200 EoS-C, a =0.1c 2 /fm EoS-C, a =0.1c 2 /fm + ω-t-ex (med ρ) 150 EoS-C, a =0.1c 2 /fm + ω-t-ex (vac ρ) NA 60 NA 60 LMR NA 60 IMR no 100 T eff (MeV) 200 EoS-C, a =0.1c 2 /fm 150 EoS-C, a =0.1c 2 /fm + NA 60 NA 60 LMR NA 60 IMR no 100 sensitivity to contributions from meson t-channel exchange hardens low-mass region using vacuum ρ in t-channel contribution: enhances slope in ρ region sensitivity to Drell-Yan contribution for IMR: describes effect seen in data (open vs. solid square data point) in LMR: too high around muon threshold due to uncertainties in extrapolation to low M?!? Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 21 / 27
IMR: vs. multi-pion radiation dn µµ /dm (counts) 200 150 100 semicentral In-In all p T (EoS-A) (EoS-B) (EoS-C) 4π (EoS-A) 4π (EoS-B) 4π (EoS-C) 50 0 0.9 1 1.1 1.2 1.3 1.4 EoS-B: dominates over multi-pion radiation opposite in EoS-A and EoS-C multi-pion radiation dominantly from high-density hadronic phase reason :dn ll /dmdt Im Π em (M, T ) exp( M/T ) T 5.5 radiation maximal for T = T max = M/5.5 hadronic and partonic radiation dual for T T c compatible with chiral-symmetry restoration! Hendrik inconclusive van Hees (JLU Gießen) whether hadronic Em Probes or inpartonic URHICs emission inseptember IMR! 1, 2009 22 / 27
Hadron spectra analysis of cocktail : hadron-m T spectra comparison to fireball evolution sequential freeze-out due to different coupling strength (1/m T ) dn/dm T (a.u.) 10 6 10 5 prim+fo fo prim ρ 10 4 ρ (semicentral) (0.6 GeV<M<0.9 GeV) T c =T ch =175 MeV, a T =0.1 c 2 /fm 10 3 0 0.5 1 1.5 2 2.5 m T - Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 23 / 27
Hadron spectra analysis of cocktail : hadron-m T spectra comparison to fireball evolution sequential freeze-out due to different coupling strength (1/m T ) dn/dm T (a.u.) 10 7 10 6 10 5 prim+fo fo prim ω 10 4 ω (semicentral) T c =T ch =175 MeV a T =0.1 c 2 /fm, t fo =5 fm 10 3 0 0.5 1 1.5 2 2.5 m T - (1/m T ) dn/dm T (a.u.) 10 7 10 6 10 5 prim+fo fo prim φ 10 4 φ (semicentral) T c =T ch =175 MeV a T =0.1 c 2 /fm, t fo =4 fm 10 3 0 0.5 1 1.5 2 2.5 m T - Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 23 / 27
M spectra (in p T slices) EoS-A: T c = T ch = 175 MeV transverse acceleration: a = 0.1c 2 /fm (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm all q T FO ρ φ ω ω-t ex +ω-t ex (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 0.2 GeV<q t <2.4 GeV FO ρ φ ω ω-t ex +ω-t ex norm corrected by 3% due to centrality correction (min-bias data: N ch = 120, calculation N ch = 140) Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 24 / 27
M spectra (in p T slices) EoS-A: T c = T ch = 175 MeV transverse acceleration: a = 0.1c 2 /fm (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 0<q t <0.2 GeV FO ρ φ ω ω-t ex +ω-t ex (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 0.2 GeV<q t <0.4 GeV FO ρ φ ω ω-t ex +ω-t ex norm corrected by 3% due to centrality correction (min-bias data: N ch = 120, calculation N ch = 140) Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 24 / 27
M spectra (in p T slices) EoS-A: T c = T ch = 175 MeV transverse acceleration: a = 0.1c 2 /fm (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 0.4 GeV<q t <0.6 GeV FO ρ φ ω ω-t ex +ω-t ex (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 0.6 GeV<q t <0.8 GeV FO ρ φ ω ω-t ex +ω-t ex norm corrected by 3% due to centrality correction (min-bias data: N ch = 120, calculation N ch = 140) Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 24 / 27
M spectra (in p T slices) EoS-A: T c = T ch = 175 MeV transverse acceleration: a = 0.1c 2 /fm (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 0.8 GeV<q t <1.0 GeV FO ρ φ ω ω-t ex +ω-t ex (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 1.0 GeV<q t <1.2 GeV FO ρ φ ω ω-t ex +ω-t ex norm corrected by 3% due to centrality correction (min-bias data: N ch = 120, calculation N ch = 140) Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 24 / 27
M spectra (in p T slices) EoS-A: T c = T ch = 175 MeV transverse acceleration: a = 0.1c 2 /fm (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 2.0 GeV<q t <2.2 GeV FO ρ φ ω ω-t ex +ω-t ex (1/N ch ) d 2 N µµ /(dm dη) (20 MeV) -1 10-5 10-6 10-7 10-8 10-9 T c =T ch =175 MeV, a t =0.1 c 2 /fm 2.2 GeV<q t <2.4 GeV FO ρ φ ω ω-t ex +ω-t ex norm corrected by 3% due to centrality correction (min-bias data: N ch = 120, calculation N ch = 140) Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 24 / 27
Dileptons@RHIC: New Puzzle? huge enhancement in the LMR unexplained yet! for details see previous talk! /GeV) IN PHENIX ACCEPTANCE 2 (c dn/dm ee 10-1 min. bias Au+Au s = 200 GeV NN DATA cc ee (PYTHIA) R.Rapp & H.vanHees y < 0.35-2 10-3 10-4 10 e p T > 0.2 GeV/c cc ee (random correlation) cocktail sum w/ ρ vacuum sum w/ ρ broadening sum/ ρ dropping) partonic yield (PY) /GeV) IN PHENIX ACCEPTANCE 2 (c dn/dm ee 10-1 min. bias Au+Au s = 200 GeV NN DATA cc ee (PYTHIA) y < 0.35 HSD e p > 0.2 GeV/c cc ee (random correlation) T cocktail -2 10 sum w/ ρ broadening -3 10-4 10 sum w/ ρ broad. + drop. -5 10-5 10 0 0.2 0.4 0.6 0.8 1 1.2 2 m ee (GeV/c ) model: Rapp, HvH [A. Adare et al (PHENIX), arxiv:0912.0244 [nucl-ex]] 0 0.2 0.4 0.6 0.8 1 1.2 2 m ee (GeV/c ) model: HSD Bratkovskaya, Cassing [A. Adare et al (PHENIX), arxiv:0912.0244 [nucl-ex]] Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 25 / 27
Conclusions and Outlook dilepton spectra in-medium em. current correlator model for dilepton sources radiation from thermal sources:, ρ, ω, φ ρ-decay after thermal freeze-out decays of non-thermalized primordial ρ s Drell-Yan annihilation, correlated D D decays invariant-mass spectra and medium effects excess yield dominated by radiation from thermal sources baryons essential for in-medium properties of vector mesons melting ρ with little mass shift robust signal! (independent of T c ) IMR well described by scenarios with radiation dominated either by or multi-pion processes (depending on EoS) Reason: mostly from thermal radiation around 160 MeV T 190 MeV parton-hadron duality of rates compatible with chiral-symmetry restoration! dimuons in In-In (), Pb-Au (CERES/NA45), γ in Pb-Pb (WA98) recent review: [R. Rapp, J. Wambach, HvH, Landolt-Brnstein, 1-23A, arxiv: 0901.3289 [hep-ph]] Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 26 / 27
Conclusions and Outlook fireball/freeze-out dynamics m T spectra and effective slopes non-thermal sources important for q T 1 GeV lower T c higher hadronic temperatures harder q T spectra to describe measured effective slopes a = 0.085c 2 /fm 0.1c 2 /fm off-equilibrium effects (viscous hydro)? Further developments understand recent PHENIX results (large dilepton excess in LMR) understand DLS puzzle (exp. confirmed by HADES) N N (np!) bremsstrahlung! vector- should be complemented with axial-vector-spectral functions (a 1 as chiral partner of ρ) constrained with lqcd via in-medium Weinberg chiral sum rules direct connection to chiral phase transition! Hendrik van Hees (JLU Gießen) Em Probes in URHICs September 1, 2009 27 / 27