Measurements of direct photons in Au + Au collisions with PHENIX - PowerPoint PPT Presentation

Measurements of direct photons in Au + Au collisions with PHENIX Hard Probes 2013 Benjamin Bannier for the PHENIX collaboration Stony Brook University November 5, 2013 1 / 19

Outline Low momentum direct photons: 0 . 4 GeV / c < p T < 5 . 0 GeV / c How are low momentum real photons measured in PHENIX? Spectra and centrality dependence of the low momentum real photons from RHIC 2 / 19

Low momentum direct photons ◮ long mean free path, escape heavy ion collision with almost no final state interaction ◮ produced at all stages of the collision in scatterings of constituents of each other or the medium ◮ probe complete temperature and flow evolution of the collision ◮ experimentally characterized by momentum-dependent yields and angular correlations with event planes 3 / 19

Low momentum direct photons (experiment) 0 BBC 0.25 π v ( Φ ) (a) 0.25 (b) 2 2 dir. BBC γ v ( Φ ) inc. BBC 2 2 γ v ( Φ ) 0.2 2 2 0.2 0.15 0.15 0.1 0.1 0.05 0.05 0 0 Min. Bias -0.05 -0.05 0 2 4 6 8 10 12 0 2 4 6 8 10 12 (c) (d) 0.25 0.25 2 0.6 fm/c v 0.2 0.2 0.4 fm/c dir. 0.15 0.15 γ , inc. 0.1 0.1 γ 0.05 0.05 , 0 π 0 0 0~20 [%] -0.05 -0.05 0 2 4 6 8 10 12 0 2 4 6 8 10 12 (e) (f) 0.25 0.25 0.2 0.2 0.15 0.15 0.1 0.1 0.05 0.05 0 0 20~40 [%] -0.05 -0.05 0 2 4 6 8 10 12 0 2 4 6 8 10 12 p [GeV/c] T (a) PRL 104, 132301 (2010) (b) PRL 109, 122302 (2012) 4 / 19

Low p T photons via external conversions in PHENIX � ε f � N γ incl. Y γ N γ π 0 incl. R γ = = Y γ Y γ hadron hadron Y γ π 0 Requirements: ◮ clean photon sample ◮ high π 0 -tagging efficiency � ε f � Photon sample ◮ measurement of low momentum photons in electromagnetic calorimeters is difficult due to e.g. MIPs ◮ PHENIX has good electron reconstruction capability down to p T = 200 MeV/c ◮ reconstruct real photons down to 400 MeV/c from e + e − pairs → no hadron contamination 5 / 19

+ - FG e e Pairs from Data 0.03 [GeV] 600 0.025 atm 500 M 0.02 400 0.015 300 0.01 200 0.005 100 0 0 0 0.005 0.01 0.015 0.02 0.025 0.03 M [GeV] cgl ◮ momentum can be reconstructed assuming production at the nominal event vertex or a defined radius ◮ conversion pairs can be selected through their invariant mass under hypotheses for production radius N γ incl. = Y γ incl. p conv a e + e − ε e + e − 6 / 19

+ - FG e e Pairs from Data 0.03 [GeV] 600 0.025 atm 500 M 0.02 400 0.015 300 0.01 200 0.005 100 0 0 0 0.005 0.01 0.015 0.02 0.025 0.03 M [GeV] cgl ◮ momentum can be reconstructed assuming production at the nominal event vertex or a defined radius ◮ conversion pairs can be selected through their invariant mass under hypotheses for production radius N γ incl. = Y γ incl. p conv a e + e − ε e + e − 7 / 19

π 0 -decay photon tagging ◮ a second photon measured with very loose cuts in the calorimeter is paired with converted photons ◮ the combinatorial background is modelled with a mixed-event sample of uncorrelated converted and calorimeter photons N γ π 0 = Y γ π 0 p conv a e + e − ε e + e − × � ε f � (a) p T ,γ = 0 . 8 − 1 . 0 GeV/c (b) p T ,γ = 2 . 0 − 2 . 5 GeV/c 8 / 19

Tagging efficiency correction � ε f � ◮ 2nd photon in acceptance → ε ◮ 2nd photon lost → f , The tagging efficiency � ε f � is calculated in a Monte Carlo simulation. ◮ f can be calculated accurately, ε ≈ 90% N γ Y γ Y γ incl. p conv a e + e − ε e + e − incl. incl. = π 0 p conv a e + e − ε e + e − × � ε f � = N γ Y γ Y γ π 0 � ε f � π 0 9 / 19

R γ in Au + Au at √ s NN = 200 GeV (a) (b) 1.6 1.4 R γ 1.2 1.0 0-20% 20-40% (c) (d) PRL 104, 132301 PH ENIX 1.6 preliminary 2007 1.4 2010 1.2 1.0 Au+Au √ s NN = 200GeV 40-60% 60-92% 1 2 3 4 1 2 3 4 p T [ GeV / c ] Figure: R γ from virtual and real photons (red, blue) in 0-20%, 20-40%, 40-60% and 60-92% more central collisions. 10 / 19

Direct photon p T spectrum d p T d y [( GeV / c ) − 2 ] (a) (b) 10 1 10 0 Au+Au √ s NN = 200GeV 10 − 1 10 − 2 10 − 3 d 2 N 10 − 4 10 − 5 2 π p T 0-20% 20-40% 1 10 − 6 (c) (d) 10 1 T AA -scaled pp fit PH ENIX 10 0 preliminary 2007, 2010 10 − 1 PRL 104, 132301 10 − 2 10 − 3 10 − 4 10 − 5 40-60% 60-92% 10 − 6 0 1 2 3 4 1 2 3 4 5 p T [ GeV / c ] Figure: Direct photon p T spectra Y γ = ( R γ − 1) Y hadron in 0-20%, γ 20-40%, 40-60% and 60-92% more central collisions. A N coll -scaled fit � − c to RHIC pp data is shown in green. � 1 + p 2 a T / b 11 / 19

Excess photon p T spectrum d p T d y [( GeV / c ) − 2 ] (a) (b) T eff = ( 237 ± 25 ± 29 ) MeV / c T eff = ( 260 ± 33 ± 31 ) MeV / c 10 1 10 0 Au+Au √ s NN = 200GeV 10 − 1 10 − 2 10 − 3 d 2 N 10 − 4 10 − 5 2 π p T 0-20% 20-40% 1 10 − 6 (c) (d) T eff = ( 228 ± 28 ± 27 ) MeV / c T eff = ( 254 ± 53 ± 25 ) MeV / c 10 1 Ae − p T / T eff PH ENIX 10 0 data - scaled p + p preliminary 10 − 1 10 − 2 10 − 3 10 − 4 10 − 5 40-60% 60-92% 10 − 6 0 1 2 3 4 1 2 3 4 5 p T [ GeV / c ] Figure: Excess photon p T spectra after subtraction of hard-scattering component in 0-20%, 20-40%, 40-60% and 60-92% more central collisions. Red lines are fits of Ae − p T / T eff in p T = 0 . 6 − 2 . 0 GeV/c. 12 / 19

Centrality dependence of excess photon yield 10 1 PH ENIX preliminary 10 − 3 10 0 d y / N 1 . 48 part d N d y 10 − 1 d N 10 − 4 10 − 2 p T > 0 . 4GeV / c p T > 1 . 0GeV / c p T > 0 . 6GeV / c p T > 1 . 2GeV / c PH ENIX Au+Au √ s NN = 200GeV p T > 0 . 8GeV / c p T > 1 . 4GeV / c preliminary 10 − 3 10 − 5 10 1 10 2 10 1 10 2 N part N part Figure: Left : Integrated excess photon yield as a function of Glauber N part . Right: Residuals of fits to power laws AN x part with x = 1 . 48 ± 0 . 08(stat) ± 0 . 04(sys). 5 GeV / c d 2 N � � d N � 1 2 π p ( i ) T ∆ p ( i ) � � d y ( p T ) = � T 2 π p T d p T d y p ( i ) � p ( i ) T T = p T 13 / 19

Summary We have measured R γ and p T spectra for real photons. Real and virtual photons show similar R γ . An excess yield of photons is seen across all centralities. No change in the shape of the photon p T spectra is seen between centralities outside uncertainties. The excess photon yield grows stronger than N part in the p T window 0 . 6 − 2 . 0GeV / c and is described by a power law with x = 1 . 48 ± 0 . 08(stat) ± 0 . 04(sys). 14 / 19

Backup 15 / 19

Characterization of excess photon p T spectra Excess photon spectra are roughly exponential in low p T range. The shape of the spectra doesn’t change outside uncertainties across centralities. 0-20% 20-40% 40-60% 60-92% [ T eff ] 237 ± 25 ± 29 260 ± 33 ± 31 228 ± 28 ± 27 254 ± 53 ± 25 MeV/c Integrated yields To quantify the centrality-dependence of the yield we can calculate 5 GeV / c d 2 N d N � 1 � � 2 π p ( i ) T ∆ p ( i ) � � d y ( p T ) = � T 2 π p T d p T d y p ( i ) � p ( i ) T T = p T 16 / 19

1 . 5 4 1 . 0 2 (spectrum - fit)/fit (spectrum - fit)/fit 0 . 5 0 0 . 0 − 0 . 5 − 2 0-20% 0-20% 20-40% 20-40% − 1 . 0 40-60% 40-60% PH ENIX PH ENIX − 4 60-92% preliminary preliminary 60-92% − 1 . 5 0 . 0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0 0 . 0 0 . 5 1 . 0 1 . 5 2 . 0 2 . 5 3 . 0 3 . 5 4 . 0 p T [ GeV / c ] p T [ GeV / c ] Figure: Normalized fit residuals for fits of the excess photon spectra to an exponential in p T = 0 . 6 . . . 2 . 0 GeV/c ( left ) and zoomed ( right ). 17 / 19

Hadron decay photon simulation To calculate R γ the efficiency-corrected ratio needs to be scaled by the expected ratio of photons yields from hadron and π 0 decays Y γ hadron / Y γ π 0 . We implement a cocktail including ◮ π 0 → γγ ◮ η → γγ , π + π − γ ◮ η ′ → γγ , π + π − γ , ωγ ◮ ω → π 0 γ using experimental π p T spectra and m T scaling for other mesons with experimental meson/ π 0 ratios. 18 / 19

(a) (b) 1.6 1.4 R γ 1.2 1.0 0-20% 20-40% (c) (d) PRL 104, 132301 PH ENIX 1.6 preliminary 2007+2010 1.4 1.2 1.0 Au+Au √ s NN = 200GeV 40-60% 60-92% 1 2 3 4 1 2 3 4 p T [ GeV / c ] Figure: R γ from virtual and real photons in 0-20%, 20-40%, 40-60% and 60-92% more central collisions. 19 / 19