Low-mass dileptons: A thermometer for the hottest stuff in the universe

Transcription

1 Low-mass dileptons: A thermometer for the hottest stuff in the universe orsten Dahms Excellence Cluster Universe - U München! 53 rd International Winter Meeting on Nuclear Physics in Bormio January 30 th, 015

2 What is its temperature? measure thermal photons Properties of the QGP Does it restore chiral symmetry? modification of the vector mesons How does it affect heavy quarks? modification of the intermediate mass region All these questions can be answered by measuring dileptons (e + e or µ + µ no strong final state interactions: leave collision system unperturbed emitted at all stages: need to disentangle contributions

3 Properties of the QGP What is its temperature? measure thermal photons Does it restore chiral symmetry? modification of the vector mesons How does it affect heavy quarks? modification of the intermediate mass region Possible modifications Chiral symmetry restoration! continuum enhancement! modification of vector mesons Modification of open charm! hermal radiation! Exotic bound states All these questions can be answered by measuring dileptons (e + e or µ + µ no strong final state interactions: leave collision system unperturbed emitted at all stages: need to disentangle contributions Quarkonia suppression (enhancement

4 Dilepton Signals: p vs. mass 3

5 Dilepton Signals: p vs. mass Low Mass Region: mee < 1. GeV/c Dalitz decays of pseudo-scalar mesons Direct decays of vector mesons In-medium decay of ρ mesons in the hadronic gas phase 3

6 Dilepton Signals: p vs. mass Low Mass Region: mee < 1. GeV/c Dalitz decays of pseudo-scalar mesons Direct decays of vector mesons In-medium decay of ρ mesons in the hadronic gas phase LMR: mee < 1. GeV/c! LMR I (p mee quasi-real virtual photon region. Low mass pairs produced by higher order QED correction to the real photon emission! LMR II (p < 1 GeV Enhancement of dileptons discovered at SPS (CERES, NA60 3

7 Dilepton Signals: p vs. mass Low Mass Region: mee < 1. GeV/c Dalitz decays of pseudo-scalar mesons Direct decays of vector mesons In-medium decay of ρ mesons in the hadronic gas phase Intermediate Mass Region:1. < mee <.9 GeV/c Correlated semi-leptonic decays of charm quark pairs Dileptons from the QGP LMR: mee < 1. GeV/c! LMR I (p mee quasi-real virtual photon region. Low mass pairs produced by higher order QED correction to the real photon emission! LMR II (p < 1 GeV Enhancement of dileptons discovered at SPS (CERES, NA60 3

8 Dilepton Signals: p vs. mass Low Mass Region: mee < 1. GeV/c Dalitz decays of pseudo-scalar mesons Direct decays of vector mesons In-medium decay of ρ mesons in the hadronic gas phase Intermediate Mass Region:1. < mee <.9 GeV/c Correlated semi-leptonic decays of charm quark pairs Dileptons from the QGP High Mass Region: mee >.9 GeV/c Dileptons from hard processes LMR: mee < 1. GeV/c! LMR I (p mee quasi-real virtual photon region. Low mass pairs produced by higher order QED correction to the real photon emission! LMR II (p < 1 GeV Enhancement of dileptons discovered at SPS (CERES, NA60 Drell-Yan process Correlated semi-leptonic decays of heavy quark pairs Charmonium Upsilon HMR probe the initial stage Little contribution from thermal radiation 3

9 Combinatorial and Correlated Background Same-sign spectrum contains all background sources /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA comb ±± /5 MeV (c counts/n evt -1 (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb +- 4

10 Combinatorial and Correlated Background Same-sign spectrum contains all background sources /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA comb ±± /5 MeV (c counts/n evt -1 (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb +- 4

11 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA comb ±± /5 MeV (c counts/n evt -1 (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb +- 4

12 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events statistics limited (stat. error increases /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA comb ±± /5 MeV (c counts/n evt -1 (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb +- 4

13 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events statistics limited (stat. error increases at LHC energies: b-mixing is a physics source to same-sign spectrum /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA comb ±± /5 MeV (c counts/n evt -1 (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb +- 4

14 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events statistics limited (stat. error increases at LHC energies: b-mixing is a physics source to same-sign spectrum Combinatorial Background from mixed events normalized to N++N statistics only limited by memory /5 MeV (c counts/n evt /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA comb ±± comb m ee (GeV/c 4

15 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events statistics limited (stat. error increases at LHC energies: b-mixing is a physics source to same-sign spectrum Combinatorial Background from mixed events normalized to N++N statistics only limited by memory Correlated Background /5 MeV (c counts/n evt /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA comb ±± comb m ee (GeV/c 4

16 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events statistics limited (stat. error increases at LHC energies: b-mixing is a physics source to same-sign spectrum Combinatorial Background from mixed events normalized to N++N statistics only limited by memory Correlated Background Cross pairs simulated with decay generator EXODUS /5 MeV (c counts/n evt /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb ±± comb +- 4

17 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events statistics limited (stat. error increases at LHC energies: b-mixing is a physics source to same-sign spectrum Combinatorial Background from mixed events normalized to N++N statistics only limited by memory Correlated Background Cross pairs simulated with decay generator EXODUS Jet pairs simulated with PYHIA /5 MeV (c counts/n evt /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb ±± comb +- 4

18 Combinatorial and Correlated Background Same-sign spectrum contains all background sources need to corrected for different acceptance (acceptance difference from mixed events statistics limited (stat. error increases at LHC energies: b-mixing is a physics source to same-sign spectrum Combinatorial Background from mixed events normalized to N++N statistics only limited by memory Correlated Background Cross pairs simulated with decay generator EXODUS Jet pairs simulated with PYHIA /5 MeV (c counts/n evt /5 MeV (c counts/n evt (a p+p s = 00 GeV all like sign pairs(n -7 ±± combinatorial background (B corr correlated pairs (B ±± cross pairs (EXODUS jet pairs (PYHIA (b p+p s = 00 GeV + - all e e pairs (N combinatorial background (B corr correlated pairs (S + B cross pairs (EXODUS jet pairs (PYHIA m ee (GeV/c comb normalized to same-sign data and use same normalization for opposite-sign background 4 ±± comb +-

19 Data absolutely normalized Excellent agreement with cocktail Filtered in PHENIX acceptance p+p at s = 00 GeV PHENIX, PLB 670 (

20 Data absolutely normalized Excellent agreement with cocktail Filtered in PHENIX acceptance p+p at s = 00 GeV PHENIX, PLB 670 ( Light hadron contributions subtracted Extract heavy quark cross sections 5

21 Data absolutely normalized Excellent agreement with cocktail Filtered in PHENIX acceptance p+p at s = 00 GeV PHENIX, PLB 670 ( Light hadron contributions subtracted Extract heavy quark cross sections cc = 518 ± 47(stat ± 135(syst ± 00(model µb bb =3.9 ±.4(stat+3 (syst µb 5

22 d+au at snn = 00 GeV Data consistent with cocktail No significant cold nuclear matter effects PHENIX, PRC 91 ( /GeV] IN PHENIX ACCEPANCE [c dn/dm ee 1/N evt Data/Cocktail -4 - d+au s = 00 GeV e y < 0.35 p e > 0. GeV/c DAA cc(pyhia bb(pyhia DY Υ /GeV] IN PHENIX ACCEPANCE [c dn/dm ee 1/N evt π 0 γee η γee η' γee ρ ee 0 ω ee & π ee φ ee & ηee J/ψ ee ψ' ee cc + bb + DY [GeV/c [GeV/c 14 ] FIG. 3: Inclusive e + e pair yield from minimum bias d+au collisions as a function of mass. he data are compared to our model of expected sources. he inset shows in detail the mass range up to 4.5 GeV/c. In the lower panel, the ratio of data to expected sources is shown with systematic uncertainties. sum m ee m e e ] 6

23 d+au at snn = 00 GeV Data consistent with cocktail Extract beauty from fit in mass & p 0.0<p <0.5 GeV/c.0<p <.5 GeV/c σbb NN = 3.4 ± 0.8 (a (stat ± 1.1 (syst (e - - ± 0.11 (model μb ] -1 No significant cold nuclear matter effects [(GeV/c dn/dm ee 1/N evt -1 Data cc (PYHIA bb (PYHIA (cc+bb (PYHIA m e ] + e -[GeV/c m e 0.5<p <1.0 GeV/c + e -[GeV/c 9.5<p <3.5 GeV/c (b (f ] ] -1 [(GeV/c dn/dm ee 1/N evt - -1 PHENIX, PRC 91 ( <p <0.5 GeV/c (a (b.0<p <.5 GeV/c - Data cc (MC@NLO bb (MC@NLO (cc+bb (MC@NLO m e ] + e -[GeV/c m e 0.5<p <1.0 GeV/c + e -[GeV/c 9.5<p <3.5 GeV/c (e (f ] - - d+au, s NN = 00 GeV d+au, s NN = 00 GeV /GeV] IN PHENIX ACCEPANCE [c dn/dm ee 1/N evt Data/Cocktail d+au e y < 0.35 > 0. GeV/c s = 00 GeV π m e ] + e -[GeV/c γee p e 3 J/ψ4 5ee6 7 8 m 1.0<p <1.5 GeV/c e + e -[GeV/c 9-3 η γee η' γee ψ' ee 3.5<p <8.0 GeV/c (c m e e [GeV/c m e ] + e -[GeV/c - - m ee [GeV/c ] DAA cc(pyhia bb(pyhia DY Υ 1.5<p /GeV] IN PHENIX ACCEPANCE [c dn/dm ee 1/N evt <.0 GeV/c (d ] m e + e -[GeV/c ] ρ ee 0 ω ee & π ee φ ee & ηee cc + bb + DY (g 0.0<p <8.0 GeV/c (h m e ] + e -[GeV/c m8 ee 9 [GeV/c ] ] m e ] + e -[GeV/c FIG. 3: Inclusive e + e pair yield from minimum bias d+au collisions as a function of mass. he data are compared to our model of expected sources. he inset shows in detail the mass range up to 4.5 GeV/c. In the lower panel, the ratio of data to expected sources is shown with systematic uncertainties. sum m e ] + e -[GeV/c m 1.0<p <1.5 GeV/c e + e -[GeV/c 9 3.5<p <8.0 GeV/c (c m e ] + e -[GeV/c m e 1.5<p <.0 GeV/c 0.0<p <8.0 GeV/c + e -[GeV/c 9 (d m e ] + e -[GeV/c (g (h m e + e -[GeV/c ] ] ]

24 d+au at snn = 00 GeV Data consistent with cocktail Extract beauty from fit in mass & p 0.0<p <0.5 GeV/c.0<p <.5 GeV/c σbb NN = 3.4 ± 0.8 (a (stat ± 1.1 (syst (e - - ± 0.11 (model μb ] -1 No significant cold nuclear matter effects [(GeV/c dn/dm ee Data cc (PYHIA bb (PYHIA (cc+bb (PYHIA m e ] + e -[GeV/c m e 0.5<p <1.0 GeV/c + e -[GeV/c 9.5<p <3.5 GeV/c -1-1 Charm very model dependent! 1/N evt PYHIA: σcc NN = 385±34(stat±119(syst μb (b MC@NLO: σcc NN = 958±96(stat±335(syst 9 d+au, s NN = 00 GeV μb (f ] ] -1 [(GeV/c dn/dm ee 1/N evt - -1 PHENIX, PRC 91 ( <p <0.5 GeV/c (a (b.0<p <.5 GeV/c - Data cc (MC@NLO bb (MC@NLO (cc+bb (MC@NLO m e ] + e -[GeV/c m e 0.5<p <1.0 GeV/c + e -[GeV/c 9.5<p <3.5 GeV/c - d+au, s NN = 00 GeV (e (f ] /GeV] IN PHENIX ACCEPANCE [c dn/dm ee 1/N evt Data/Cocktail d+au e y < 0.35 > 0. GeV/c s = 00 GeV π m e ] + e -[GeV/c γee p e 3 J/ψ4 5ee6 7 8 m 1.0<p <1.5 GeV/c e + e -[GeV/c 9-3 η γee η' γee ψ' ee 3.5<p <8.0 GeV/c (c m e e [GeV/c m e ] + e -[GeV/c - - m ee [GeV/c ] DAA cc(pyhia bb(pyhia DY Υ 1.5<p /GeV] IN PHENIX ACCEPANCE [c dn/dm ee 1/N evt <.0 GeV/c (d ] m e + e -[GeV/c ] ρ ee 0 ω ee & π ee φ ee & ηee cc + bb + DY (g 0.0<p <8.0 GeV/c (h m e ] + e -[GeV/c m8 ee 9 [GeV/c ] ] m e ] + e -[GeV/c FIG. 3: Inclusive e + e pair yield from minimum bias d+au collisions as a function of mass. he data are compared to our model of expected sources. he inset shows in detail the mass range up to 4.5 GeV/c. In the lower panel, the ratio of data to expected sources is shown with systematic uncertainties. sum m e ] + e -[GeV/c m 1.0<p <1.5 GeV/c e + e -[GeV/c 9 3.5<p <8.0 GeV/c (c m e ] + e -[GeV/c m e 1.5<p <.0 GeV/c 0.0<p <8.0 GeV/c + e -[GeV/c 9 (d m e ] + e -[GeV/c (g (h m e + e -[GeV/c ] ] ]

25 Electron-muon correlations!!!!! Open Charm Correlations Direct access to cc and bb No contribution from hadron decays e + K νe Current measurement with rapidity gap Needs modelling to relate to midrapidity dielectron results D 0 c c D 0 _ νµ µ K + n 7

26 Au+Au at snn = 00 GeV /GeV IN PHENIX ACCEPANCE (c dn/dm ee Low Mass Region: enhancement 150 < mee < 750 MeV/c p+p s = 00 GeV π 0 γ ee J/ψ ee 4.7 ± 0.4(stat ± 1.5(syst ± 0.9(model η γee ψ ee DAA y < 0.35 η γee cc ee (PYHIA Intermediate e Mass Region: p > 0. GeV/c ρ ee bb ee (PYHIA dominated by charm 0 (Ncoll DY ee (PYHIA σcc NN ω ee & π φ ee & ηee ee sum Input: dσcc NN /dy= 13±1±37µb PYHIA _ Random cc correlation Single electron measurement: High p suppression of charm Charm flows Expected modifications in the pair invariant mass _ random cc correlation? Room for thermal contribution? FIG. 5. (Color online Inclusive mass spectrum of e + e pairs in e PHENIX acceptance in p+p collisions compared to the expecta- /GeV IN PHENIX ACCEPANCE (c dn/dm ee Data/Cocktail -1 min. bias Au+Au s = 00 GeV NN DAA π 0 γee y < 0.35 e p > 0. GeV/c PHENIX, PRC 81 ( η γee η' γee ρ ee 0 ω ee & π ee φ ee & ηee J/ψ ee ψ' ee cc ee (PYHIA sum cc ee (random correlation bb ee (PYHIA DY ee (PYHIA m ee (GeV/c FIG. 6. (Color online Inclusive mass spectrum of e + e pairs in the PHENIX acceptance in minimum-bias Au+Au compared to expectations from the decays of light hadrons and correlated decays 8

27 Centrality Dependence PHENIX, PRC 81 (

28 Centrality Dependence PHENIX, PRC 81 ( / ( Yield / (N part 4 (b 3 1 Yield (0<m <0 MeV/c / (N / ee part N part FIG. 9. (Color online Dielectron yield per participating nucleon pair (N part / as function of N part for two different mass ranges (a: 0.15 <m ee < 0.75 GeV/c,b:0<m ee < 0.1GeV/c compared to the expected yield from the hadron decay model. he two lines give the systematic uncertainty of the yield from cocktail and charmed 9

29 Centrality Dependence A. ADARE et al. PHENIX, PRC 81 ( Au+Au (a Yield (150<m <750 MeV/c / (N / p+p s NN ee = 00 GeV part / ( Yield / (N part 0 15 Au+Au COCKAIL 5-5 / ( Yield / (N part 0 4 (b 3 1 Yield (0<m ee part <0 MeV/c / (N / N part FIG. 9. (Color online Dielectron yield per participating nucleon pair (N part / as function of N part for two different mass ranges (a: 0.15 <m ee Waiting < 0.75 GeV/cfor,b:0<m HBD ee < results?! 0.1GeV/c compared to the expected yield from the hadron decay model. he two lines give the systematic uncertainty of the yield from cocktail and charmed 9

30 SAR Low Mass Dileptons from SAR SAR observes smaller enhancement than PHENIX In better agreement with models that -1 L 113, 0301 involve (014 broadening of Pρ Hspectral YSICAL function REVIEW LEERS 3 N part Yield Caveat: 1.4x larger MinBias charm cross sections used in cocktail: dσ/dy= 171±6 µb Slower increase with (b centrality: Yield/Npart ~ Npart 0.54±0.18 (c N part b a N part /GeV (c dn/dm ee Ratio to Cocktail /GeV (c dn/dm ee -3-5 A: ρ-like ( GeV/c. 3 (color SAR, online. PRL 113 Integrated ( dielectron B: ω-like yields( in the mass GeV/c ions of (ρ-like, (ω-like, C: φ-like and ( GeV/c (A: GeV/c (B: GeV/c (C: GeV/c (A - cocktail (B and (C N part (a (b Au + Au e s NN p >0. GeV/c e η <1, y <1 ee = 00 GeV (MinBias π 0, φ, J/ψ, ψ η, η, ω, bb, DY cc PYHIA Cocktail Sum.5 Rapp: broadened ρ +QGP (GeV/c M ee 0.01 Data - Cocktail PHSD: broadened ρ +QGP (c Rapp: vacuum ρ +QGP Rapp: broadened ρ +QGP (GeV/c M ee PHSD: broadened ρ +QGP

31 SAR Low Mass Dileptons: BES SAR measured low mass dileptons at five different snn no strong energy dependence of enhancement total baryon density comparable at 0 and 00 GeV AID:0 /FLA [m1+; v 1.196; Prn://014; 16:07] P.5 (1 increases only at even lower snn P. Huck / Nuclear Physics A ( 5 ured BES Phase-I LMR excess normalized to the respective dn/dy π. he grey box represents the average BES Phase-I SAR, projected PRL into the 113 BES Phase-II (014 regime 0301 with the size of the uncertainty taken from 00 GeV. line depicts the expected trend of the LMR excess in the BES Phase-II regime as suggested by the according HSD calculations. SAR (Preliminary 11

32 SAR Low Mass Dileptons: BES Enhancement compared at 19.6 and 00 GeV Fully acceptance corrected Spectrum at low energy consistent with NA60 result In contrast to NA60 did not separate prompt from nonprompt charm decay experimentally subtracted charm and Drell-Yan with PYHIA Indication of energy and centrality dependence of low mass enhancement very similar total baryon density sign of longer medium lifetime SAR, arxiv: /dy (0 MeV 55 FIG. 4. (color online he acceptance-corrected excess dielectron mass spectra, normalized to the charged particle multi Au+Au 00 GeV 00% Au+Au 00 GeV 18 plicity45at mid-rapidity, Au+Au 19.6 in Au+Au GeV collisions at s NN =19.6 (solid circles andin+in GeV (diamonds. Comparison to 16the 40 NA60 data [8, 37] for In+In collisions at s NN =17.3GeV (open circles is also shown. Bars are statistical uncertainties, and 30 systematic uncertainties are shown as grey boxes. 1 A model calculation (solid curve [11, 3] with a broadened ρ 5 spectral function in hadron gas (HG and QGP thermal ra- diation 0 h. lifetime 17.3 GeV is compared with the excess in Au+Au collisions h. lifetime 19.6 GeV 8 at snn 15=19.6GeV.henormalizationuncertaintyfromthe h. lifetime 00 SAR measured dn/dy is about %, which is not shown 6 in 0.4<M ll <0.75 GeV/c the figure /dy ch (dn/dy/(dn ch N/dydM/(dN (d Dielectron excess Au+Au 19.6 GeV 00% Au+Au 00 GeV 00% In+In 17.3 GeV dn ch /dη>30 HG+QGP 1 M ll (GeV/c dn ch /dy lifetime (fm/c

33 ies, respectively. SAR Azimuthal Anisotropy v 0.4 a π b + - measured(e e c + - simulation(e e M ee <0.14 GeV/c 0.14<M ee <0.3 GeV/c 0.5<M ee <0.7 GeV/c p (GeV/c v 0.5 d e φ f <M ee <0.8 GeV/c 0.98<M ee <1.06 GeV/c 1.1<M ee <.9 GeV/c p (GeV/c SAR, arxiv: (submitted to PRC Color online Panels (a-f: he dielectron v as a function of p in minimum-bias Au+Au collisions at p s x di erent Dielectron mass regions: v consistent 0,, charm with + 0, cocktail!,, and charm+qgp thermal radiation. Also shown are th and the meson v [40] measured through the decay channel! K + K. he expected dielectron v m 0, More,! and data decays needed in theto relevant extract mass v regions of are shown in panels (a-e while that from c c contri anel (f. he bars and bands represent statistical and systematic uncertainties, respectively. low mass enhancement 13

34 Virtual Photons Any source of real γ can emit γ* with very low mass If the Q (= m of virtual photon is sufficiently small, the source strength should be the same he ratio of real photon and quasi-real photon can be calculated by QED Real photon yield can be measured from virtual photon yield, which is observed as low mass e + e pairs q Kroll-Wada equation: d N dm ee dn = 3 s 1 4m e M ee! 1+ m e Mee Case of Hadrons:! S = F Mee! 1 M ee S g γ " obviously S = 0 at Mee > Mhadron Case of γ*: if p Mee : S = 1 1 M ee M hadron! 3 q e + e - Possible to separate hadron decay components from real signal in the proper mass window 14

35 Fitting the direct γ* fraction r = direct γ*/inclusive γ* determined by fitting the following function for each p bin f data (M ee =(1 r f cocktail (M ee +r f direct (M ee fdirect given by Kroll-Wada formula with S = 1 fcocktail given by cocktail components Normalized to data for m < 30 MeV/c! Fit in MeV/c (insensitive to π 0 yield Assuming direct γ* mass shape: χ /NDF = 1./6 PHENIX, PRL 4 ( PHENIX, PRC 81 (

36 hermal Photons Direct photon measurements real (p > 4GeV virtual (1 < p < 4 GeV & mee < 300 MeV! p+p: good agreement with NLO pqcd down to p = 1 GeV/c Au+Au = pqcd + exponential ave = 1 ± 19 (stat ± 19 (syst MeV In agreement with hydrodynamic models: Assume formation of a hot QGP with: 300 MeV < init < 600 MeV 0.6 fm/c < τ0 < 0.15 fm/c exponential + AA scaled p+p PHENIX, PRL 4 ( PHENIX, PRC 81 ( well above the critical temperature of 170 MeV predicted by lattice QCD 16

37 Internal vs. External Conversions Identify real photons via conversion pairs ag photons from π 0 decays look for an em. shower in the calorimeter and reconstruct γγ mass conversion probability cancels PHENIX, arxiv: (submitted to PRC 17

38 Internal vs. External Conversions Identify real photons via conversion pairs ag photons from π 0 decays look for an em. shower in the calorimeter and reconstruct γγ mass conversion probability cancels Matches result of internal conversions validates internal conversion method and extrapolation to m = 0 d N dp dy [(GeV/c ] 1 p p p snn = 00GeV N coll -scaled pp fit pp fit Au+Au, min. bias PRL 4, (0 PRL 98, 0100 (0 PRD 86, (01 PRL 9, 1530 (01 PRL 4, ( p [GeV/c] PHENIX, arxiv: (submitted to PRC 17

39 Internal vs. External Conversions Identify real photons via conversion pairs ag photons from π 0 decays look for an em. shower in the calorimeter and reconstruct γγ mass conversion probability cancels Matches result of internal conversions validates internal conversion method and extrapolation to m = 0 Excess scales with ~Npart 1.5 not an artefact of low p points 1 d N dp dy [(GeV/c ] 1 p p p snn = 00GeV N coll -scaled pp fit pp fit Au+Au, min. bias PRL 4, (0 PRL 98, 0100 (0 PRD 86, (01 PRL 9, 1530 (01 PRL 4, ( p [GeV/c] 0 dn dy 1 PHENIX, arxiv: (submitted to PRC p > 0.4GeV/c p > 0.6GeV/c p > 1.0GeV/c p > 1.GeV/c p > 0.8GeV/c p > 1.4GeV/c 3 1 N part 17

40 hermal Photon v Elliptic flow of direct photons at low p similar to π 0 flow direct photons emitted from the late stage hard to combine with large rates which suggest early emission from the hot phase v π 0, γ inc., γ dir. 0.5 π v (Φ (a 0.5 dir. BBC (b γ v (Φ inc. BBC γ v 0. (Φ Min. Bias BBC (c 0.5 (d 0.6 fm/c fm/c ~0 [%] (e 0.5 (f PHENIX, PRL 9 ( ~40 [%] p [GeV/c] 18

41 hermal Photon v3 v 3 0. dir. γ v 3 π 0 v3 E.P. : RxN(I+O 3 0-0(% 0-40(% 400(% Au+Au 00GeV p (GeV/c 0 4 p (GeV/c p (GeV/c Direct photons show also a v 3 similar to π 0 Expect direct photon v3 from fluctuations in the initial geometry viscosity should reduce v3 19

42 hermal photons at the LHC? ALICE measured real photons via conversion pairs Observed excess above hadron decays However: needs comparison to pp! crucial for the quantifying enhancement Large background from π 0 decays Can be avoided by measuring virtual γ Large uncertainty due to material budget Can be avoided by tagging method Similar flow as at RHIC w 0

43 Direct Photons in p+p! Preliminary measurement of direct photon fraction in p+p at 7 ev direct photon ratio!!!!!! via low mass dielectrons observe hint of small signal (GeV/c ALI PREL 6907 ALICE preliminary pp, s=7 ev virtual photons (1+r conversions (γ /π 0 /(γ /π 0 inc decay Extracted direct photon cross section consistent with NLO pqcd p c 3 - (pb GeV p 3 σ/d 3 d E ALI PREL ALICE preliminary pp, s=7 ev 95 % C.L. γ direct γ incl γ direct (W.V. NLO µ=0.5 µ=1.0 µ= p = r (PCM γ incl (GeV/c 1

44 Mass continuum in pp collisions Low Mass Dileptons in ALICE Hadronic cocktail (mb/gev/c dσ dm ee Data/Cocktail cocktail sum pp, s=7 ev π 0 eeγ η eeγ p 1 e >0. GeV/c η eeγ η e <0.8 φ ee & φ eeη ALI PREL ω ee & ω eeπ ρ ee J/ψ ee & J/ψ eeγ cc ee (PYHIA, 8.5 mb m ee (GeV/c -1 ((GeV/c dn /dm ee NSD 1/N evt data/cocktail -1 1 Cocktail sum with uncertainties I sallis ALICE Preliminary parametrization of 0 π γee p-pb NSD s = 5.0 ev η γee 0 NN e,, and J/ 0 p > 0. GeV/c ω ee and ω π ee φ ee and φ ηee η e < 0.8 η' γee I- Other mesons: m scaling ρ ee cc (<N > x pp PYHIA MNR, σ I bb (<N > x pp PYHIA MNR, σ -3 c c dynamics from PYHIA (Like-sign subtracted ppb coll ppb coll I c c normalized to measured -4 cross section -5 I Systematic uncertainty determined by systematic 1 uncertainty of input spectra J/ψ ee and J/ψ γee cc bb = 6.9mb = µb (GeV/c ALI-PREL9715 m ee Dielectron spectra in p+p and p+pb consistent with hadron cocktail I Hadronic cocktail is consistent with data for p -integrated Big uncertainty on open charm cross section dielectron spectrum I Hadronic cocktail is consiste 0 <m ee < 3.5 GeV/c

45 Low Mass Dileptons in ALICE: Pb+Pb -1 raw yield ((GeV/c 7 6 ALICE Preliminary Pb Pb, 0.4 < p 1.0 p e pair s NN =.76 ev, 0-% < 3.5 GeV/c, η e < GeV/c S/B 1 ALICE Preliminary Pb Pb, 0.4 < p 1.0 p e pair s NN =.76 ev, 0-% < 3.5 GeV/c, η e < GeV/c ALI-PREL (GeV/c m ee S/B ratio of few at low p bin: -4 ALI-PREL8533 accurate combinatorial background evaluation needed (GeV/c mall S/B ratio requires a precise determination of the background Focus on virtual photon production hape and normalization! Work in progress m ee 3

46 Low Mass Dileptons in ALICE: Future PC and IS upgrades: allow high data rates reduce charm background with dca cut /<&"1$1+(*'"#$%&"'"(* G-"*.1=*(".*5d9*#(0** s NN = 5.5 ev #0"0*dQ6* 0 - %,.5E9-1 dn/dm ee dy (GeV Dedicated (".*5d9*MHLXIP low B-field e *.1=*S63*%$* run (B=0. -1 dn/dm ee dy (GeV (GeV/c s NN 0 - %,.5E9 y < 0.84 e e p > 0. GeV/c 0.0 < p < 3.0 t,ee = 5.5 ev Rapp Sum Rapp in-medium SF Rapp QGP M ee.5 9 events = 1 year at 50 khz.5e9 'meas.' - cc - cockt. Syst. err. bkg. Syst. err. cc + cocktail $B6$CC%C!$6<DE;% (GeV/c M ee -4-5 epton workshop, rento, May 4, (GeV/c M ee y < 0.84 e e p > 0. GeV/c 0.0 < p < 3.0 t,ee Sum Rapp in-medium SF Rapp QGP cocktail w/o! (± % cc " ee (± 0%.5E9 'measured' Syst. err. bkg. (± 0.5% Simulation Simulation

47 Summary EM probes ideal penetrating probes of dense partonic matter created at RHIC and the LHC! Double differential measurement of dilepton emission rates can provide! emperature of the matter Medium modification of EM spectral function PHENIX and SAR measured dilepton continuum in p+p, d+au and Au+Au:! In p+p and d+au: good agreement between data and hadronic cocktail measured charm and beauty cross section in IMR and HMR In Au+Au: low p and low mass enhancement above hadronic cocktail PHENIX data not reproduced by theoretical models discrepancy to SAR measurement, which is well explained by models involving ρ broadening In Au+Au: very low-mass enhancement for p>1 GeV/c interpreted as thermal photons with inverse slope of ~0 MeV (initial = MeV strong elliptic flow in combination with large yields a challenge for theory At the LHC:! first performance studies from ALICE need IS and PC upgrades for precise measurements (+ low B-field 5

48 Backup

49 From SPS to RHIC SPS profit from high baryon density dropping ρ mass broadening of ρ What to expect at RHIC? increase of centre-of-mass per nucleon-nucleon pair from 17 to 00 GeV!!!!!! SPS (Pb+Pb RHIC (Au+Au _ dn(p/dy produced baryons (p, p, n, n _ p - p _ particpating nucleons (p - pa/z total baryon number 1! Baryon density: almost the same at SPS & RHIC (although the NE baryon density is not! 7

50 Momentum Dependence p+p in agreement with cocktail PHENIX, PRC 81 ( Au+Au low mass enhancement concentrated at low p 8

51 Momentum Dependence: PHENIX vs. SAR A. ADARE et al. PHYSICAL REVIEW C 81,034911(0 dy [(c/gev ] N/dp p+p min. s bias = 00 Au+Au GeV s NN = 00 GeV m ee m< ee 0 < 0 MeV/c 0 0 MeV/c MeV/c m ee ee < 00 MeV/c MeV/c MeV/c m ee < ee 300 MeV/c MeV/c 300 MeV/c m 300 MeV/c ee < 500 MeV/c 500 MeV/c -1 ee 500 MeV/c m 500 MeV/c ee < 750 MeV/c 750 MeV/c - ee 8 MeV/c m ee < 990 MeV/c -3 8 MeV/c m ee < 990 MeV/c min. bias Au+Au < 0 MeV/c m ee s = 00 GeV NN 0 MeV/c m ee < 00 MeV/c 00 MeV/c m ee < 300 MeV/c 300 MeV/c m ee < 500 MeV/c 500 MeV/c m ee < 750 MeV/c MeV/c m ee < 990 MeV/c d p -7 / (1/( π part -9 - (1/(N p p (GeV/c p (GeV/c FIG. PHENIX: 37. (Color online Au+Au p low mass enhancement concentrated at low p n p+p (left and Au+Au (right spectra collisions of for e + edifferent pairs inmass p+p bins, (left which and Au+Au are fully(right collisions for different mass bins, which are fully part/. acceptance he solid corrected. curves Au+Au show thespectra expectations are divided from by thensum part /. of he the hadronic solid curves decay show the expectations from the sum of the hadronic decay e dashed cocktailcurves and the show contribution the sum of from thecharmed cocktail and mesons. charmed he dashed meson contributions curves show the plusum of the cocktail and charmed meson contributions plus rting the the contribution photon yield from from direct Fig. photons 34 the calculated e + e pair byyield converting using Eqs. the photon (31and(B14. yield from Fig. 34 to the e + e pair yield using Eqs. (31and(B14. 9

52 Momentum Dependence: PHENIX vs. SAR A. ADARE et al. PHYSICAL REVIEW C 81,034911(0 dy [(c/gev ] N/dp p+p min. s bias = 00 Au+Au GeV s NN = 00 GeV m ee m< ee 0 < 0 MeV/c 0 0 MeV/c MeV/c m ee ee < 00 MeV/c MeV/c MeV/c m ee < ee 300 MeV/c MeV/c 300 MeV/c m 300 MeV/c ee < 500 MeV/c 500 MeV/c -1 ee 500 MeV/c m 500 MeV/c ee < 750 MeV/c 750 MeV/c - ee 8 MeV/c m ee < 990 MeV/c -3 8 MeV/c m ee < 990 MeV/c min. bias Au+Au < 0 MeV/c m ee s = 00 GeV NN 0 MeV/c m ee < 00 MeV/c 00 MeV/c m ee < 300 MeV/c 300 MeV/c m ee < 500 MeV/c 500 MeV/c m ee < 750 MeV/c MeV/c m ee < 990 MeV/c d p -7 / (1/( π part -9 - (1/(N p p (GeV/c p (GeV/c FIG. PHENIX: 37. (Color online Au+Au p low mass enhancement concentrated at low p n p+p (left and Au+Au (right spectra collisions of for e + edifferent pairs inmass p+p bins, (left which and Au+Au are fully(right collisions for different mass bins, which are fully part/. acceptance he solid corrected. curves Au+Au show thespectra expectations are divided from by thensum part /. of he the hadronic solid curves decay show the expectations from the sum of the hadronic decay e dashed cocktailcurves and the show contribution the sum of from thecharmed cocktail and mesons. charmed he dashed meson contributions curves show the plusum of the cocktail and charmed meson contributions plus rting the the contribution photon yield from from direct Fig. photons 34 the calculated e + e pair byyield converting using Eqs. the photon (31and(B14. yield from Fig. 34 to the e + e pair yield using Eqs. (31and(B14. 9

53 4 P. Huck / Nuclear Physics A ( JID:NUPHA AID:0 /FLA [m1+; v 1.196; Prn://014; 16:07] P.4 (1 4 P. Huck / Nuclear Physics A ( Momentum Dependence: PHENIX vs. SAR A. ADARE et al. PHYSICAL REVIEW C 81,034911(0 dy [(c/gev ] N/dp d p p+p min. s bias = 00 Au+Au GeV s NN = 00 GeV m ee m< ee 0 < 0 MeV/c 0 0 MeV/c MeV/c m ee ee < 00 MeV/c MeV/c MeV/c m ee < ee 300 MeV/c MeV/c 300 MeV/c m 300 MeV/c ee < 500 MeV/c 500 MeV/c -1 ee 500 MeV/c m 500 MeV/c ee < 750 MeV/c 750 MeV/c - ee 8 MeV/c m ee < 990 MeV/c -3 8 MeV/c m ee < 990 MeV/c SAR min. bias Au+Au < 0 MeV/c m ee s = 00 GeV NN 0 MeV/c m ee < 00 MeV/c 00 MeV/c m ee < 300 MeV/c 300 MeV/c m ee < 500 MeV/c 500 MeV/c m ee < 750 MeV/c MeV/c m ee < 990 MeV/c / (1/( π (1/(N part p p (GeV/c p (GeV/c FIG. PHENIX: 37. (Color online Au+Au p low mass enhancement concentrated at low p Fig. 1. Measured spectrainvariant of e + e pairs mass inand p+p LMR (left and p Au+Au (right collisions for different mass bins, which are fully Fig. 1. Measured invariant mass and LMR p dielectron dielectron spectra for BES Phase-I compared to cocktail simulations spectra for BES Phase-I compared to cocktail simulations and model calculations [15,7]. (Left In addition to the total cocktail (grey, the single vector meson contributions are and model SAR: calculations no [15,7]. p spectrum (Left In addition at to the snn = 00 total cocktail GeV (grey, the single vector meson contributions are depicted for 19 GeV. For 39 GeV, the two model contributions from hadronic gas and QGP phase are shown next to the depicted for 19 GeV. For 39 GeV, the two model contributions from hadronic gas and QGP phase are shown next to the total expected yield from cocktail and model (orange. Also, contributions from photon conversions are not removed from 9 n p+p (left and Au+Au (right collisions for different mass bins, which are fully part/. acceptance he solid corrected. curves Au+Au show thespectra expectations are divided from by thensum part /. of he the hadronic solid curves decay show the expectations from the sum of the hadronic decay e dashed cocktailcurves and the show contribution the sum of from thecharmed cocktail and mesons. charmed he dashed meson contributions curves show the plusum of the cocktail and charmed meson contributions plus rting the the contribution photon yield from from direct Fig. photons 34 the calculated e + e pair byyield converting using Eqs. the photon (31and(B14. yield from Fig. 34 to the e + e pair yield using Eqs. (31and(B14.

54 Direct γ*/inclusive γ* p+p µ = 0.5 p! µ = 1.0 p! µ =.0 p Au+Au Base line (curves:! NLO pqcd calculations with different theoretical scales done by W. Vogelsang p+p Consistent with NLO pqcd better agreement with small µ! Au+Au direct inclusive = direct inclusive Clear enhancement above NLO pqcd PHENIX, PRL 4 ( PHENIX, PRC 81 (

55 SAR 1 Direct γ*/inclusive γ* Preliminary result by SAR Also sees enhancement above NLO pqcd at low p / inclusiveγ fraction = directγ Au+Au SAR theory scale theory scale theory scale µ=0.5p µ=1.0p µ=.0p SAR Preliminary p (GeV/c 31

56 SAR 1 Direct γ*/inclusive γ* Preliminary result by SAR Also sees enhancement above NLO pqcd at low p / inclusiveγ fraction = directγ Au+Au SAR theory scale theory scale theory scale µ=0.5p µ=1.0p µ=.0p SAR Preliminary p (GeV/c Below GeV/c: smaller fraction than PHENIX? 31

57 Direct Photons: Model Comparison From data: Au+Au = pqcd + exponential ave = 1 ± 19 (stat ± 19 (syst MeV Comparison to hydrodynamical models: p < 3 GeV/c thermal contribution dominates over pqcd. Assume formation of a hot QGP with: 300 MeV < init < 600 MeV 0.6 fm/c < τ0 < 0.15 fm/c Models reproduce the data within a factor of two PHENIX, PRC 81 ( eff from fit ~ 0 MeV c from lattice QCD ~ 170 MeV 3

58 emperature of the early QGP Photons emitted during all stages of the medium evolution measurement is always time integrated Low mass virtual photons: Emission dominated by the hadronic phase How to measure early temperature from the partonic phase? Intermediate mass (1 3GeV/c provides another window to measure thermal dileptons NA60 at the SPS measured prompt excess above open charm continuum of possibly thermal origin semileptonic charm decays = non-prompt At the LHC charm cross section 3 times higher than at the SPS! Such measurement requires good vertexing resolution for lepton pairs Rapp, APP B4 ( NA60, EPJ C 59 (

59 Hadronic Cocktail Parameterization of PHENIX π ±, π 0 data (π 0 = (π + + π /!! E d3 dp 3 = A exp( ap bp +p n /p 0 Other mesons: fit with m scaling of π 0 p (p + mmeson mπ fit the normalization constant All mesons m scale! Hadronic cocktail was well tuned to individually measured yield of mesons in PHENIX for both p+p and Au+Au collisions Mass distributions from hadron decays are simulated by Monte Carlo: π 0, η, η, ω, φ, ρ, J/ψ, ψ Effects on real data are implemented 34

60 Hadronic Cocktail Parameterization of PHENIX π ±, π 0 data (π 0 = (π + + π /!! E d3 dp 3 = A exp( ap bp +p n /p 0 Other mesons: fit with m scaling of π 0 p (p + mmeson mπ fit the normalization constant All mesons m scale! Hadronic cocktail was well tuned to individually measured yield of mesons in PHENIX for both p+p and Au+Au collisions Mass distributions from hadron decays are simulated by Monte Carlo: π 0, η, η, ω, φ, ρ, J/ψ, ψ Effects on real data are implemented 34