Direct Photon Results from ALICE and the Direct Photon Puzzle Ab initio approaches in many-body QCD confront heavy-ion experiments December 15-17, 2014 Klaus Reygers Physikalisches Institut University of Heidelberg, Germany
he Role of Direct Photons in Heavy-Ion Physics τ = 1.2 fm/c τ = 4.1 fm/c Escape medium unscathed Produced during all stages test of the hydro paradigm Experimental access to initial QGP temperature (?) τ = 7.0 fm/c τ =.1 fm/c QGP photon rate (lowest order): E dn d 3 p / s 2 e E / log E k 2 c 2
A Complication for the emperature Measurement: Blueshift due to Radial Flow C. Shen, U. Heinz, J.-F. Paquet, C. Gale, PRC 89 (2014) 4, 0449 E d3 N d 3 p / e E / e s 1+ flow e = 1 flow {z } 2for flow=0.6 Large blueshift at late times when 150-200 MeV Extraction of initial temperature from data requires comparison to (hydro) model 3
Photon Conversion Method: Precise Knowledge of Material Budget is Essential Y (cm) 0 reconstructed photon conversion points SSD η < 0.9 5 50 SDD 4 X/X0 = 0.114 ± 0.005 ( η < 0.9, R < 180 cm) 0 3 pconv 8.5% rel. uncertainty: 4.5% -50 2 SPD PC drift gas -0 ALI PUB 72606 PC inner field cage vessel PC Rods PC inner containment vessel -0-50 0 50 0 X (cm) 4
How to Measure Direct Photons? direct := inclusive decay =(1 1 R ) inclusive systematic uncertainties partially cancel in this ratio (efficiency, energy/momentum scale, material budget...) with R = ( inclusive/ 0 ) meas ( decay / 0 ) calc Calculated decay photon cocktail (π 0, η, ω,...), π 0 measured, other hadrons from m scaling so far 5
No Significant Direct-Photon Excess in pp at 7 ev /π 0 ) 2.0 decay /π 0 )/(γ 1.8 pp, s = 7 ev inc (γ 1.6 1.4 Direct photon double ratio NLO prediction: 1 + (γ for µ = 0.5,1.0,2.0 p /γ ) direct,nlo decay 1.2 1.0 0.8 ALI PREL 27932 0 2 4 6 8 12 p (GeV/c) 6
No Significant Direct-Photon Excess in Peripheral Pb-Pb /π 0 ) decay 1.8 40-80% Pb-Pb, s NN = 2.76 ev /π 0 )/(γ inc 1.6 (γ 1.4 Direct photon double ratio NLO prediction: 1 + (N for µ = 0.5,1.0,2.0 p γ coll /γ ) direct,pp,nlo decay 1.2 1.0 0.8 ALI PREL 28000 0 1 2 3 4 5 6 7 8 p (GeV/c) 7
Direct-photon Excess in Central Pb-Pb Collisions /π 0 ) decay /π 0 )/(γ inc 3.0 2.5 0-40% Pb-Pb, s NN = 2.76 ev (γ 2.0 Direct photon double ratio NLO prediction: 1 + (N for µ = 0.5,1.0,2.0 p γ coll /γ ) direct,pp,nlo decay 1.5 1.0 0.5 0 2 4 6 8 12 14 ALI PREL 27956 15-20% direct-photon excess for 1 < p < 2 GeV/c where contribution from pqcd photons is small p (GeV/c) 8
Direct Photon Spectrum in Central Pb-Pb Collisions (γ direct := γ all - γ decay ) c 2 ) -2 (GeV dy d N dp 2 p N ev. 2 π 1 3 2 1-1 -2 0-40% Pb-Pb, s NN = 2.76 ev Direct photons Direct photon NLO for µ = 0.5,1.0,2.0 p Exponential fit: A exp(-p /), = 304 ± (scaled pp) 51 MeV -3-4 -5-6 -7 ALI PREL 27968 0 2 4 6 8 12 14 Low p part of the spectrum described by exponential with inverse slope = 304 ± 51 MeV (GeV/c) p 9
Comparison of the Direct Photon Excess in 0-40% most central Pb-Pb collisions with Hydro Models R γ 1.8 ALICE preliminary 0-40% Pb-Pb, s NN = 2.76 ev 1.6 1.4 Direct photon double ratio NLO prediction: 1 + (N γ /γ ) coll direct,pp,nlo decay thermal (Shen et al., arxiv:1308.2111) thermal (Holopainen et al., Phys.Rev. C84 (2011) 064903) 1.2 1.0 0.8 Systematic uncertainties largely correlated in p 0 1 2 3 4 5 ALI PREL 75704 Direct photon excess appears to be larger than expected in hydrodynamic model Are we missing an important photon source in these models? p (GeV/c)
How to Measure the Direct-Photon v 2? Reaction plane (RP) from charged particles in forward direction 2.8 < η < 5.1-3.7 < η < -1.7 Inclusive photons (mid-rapidity): 2 = hcos(2(' RP 2 ))i C v,incl, C = resolution correction Decay photon v2 from cocktail calculation based on measured pion v2 (+ higher mass hadrons) Inclusive photon v2 is weighted average of decay photon and direct photon v2. hus one can calculate the direct-photon v2 as v,direct 2 =,incl Rv 2 v,decay 2 R 1 with R = incl decay =1+ direct decay 11
Measured Inclusive Photon and Calculated Decay Photon v 2 γ 2 v 0.20 ALICE preliminary γ,decay v 2 0.18 0.16 0.14 0.12 0-40% Pb-Pb, V0 event plane s NN = 2.76 ev γ,incl v 2 v2(incl) < v2(decay) for p > 3 GeV/c expected from v2 = 0 for prompt photons v2(incl) v2(decay) for p < 3 GeV/c: 0. 0.08 0.06 decay + NLO 0.04 Phys.Rev. D50 (1994) 1901-1916 decay + NLO + thermal (Shen et al.) arxiv:1308.2111 0.02 decay + NLO + thermal (Holopainen et al.) Phys.Rev. C84 (2011) 064903 0.00 0 1 2 3 4 5 6 ALI PREL 75753 p (GeV/c) If there is a large direct photon component its v2 must be very similar to the decay photon v2 v2(incl) described by models with small R γ predicted by the same models 12
Recap: What to expect for the Direct Photon v 2? Large inverse slope parameter: slope 304 ± 51 stat+syst MeV (>> c = 150-160 MeV) Could indicate that direct photons mostly come from early hot QGP phase Expect then small elliptic flow signal (v2 3% or so at maximum) as collective flow needs time to build up 13
Direct Photon Elliptic Flow Appears to be Larger than Expected in Hydro Models γ,dir 2 v 0.2 ALICE preliminary 0-40% Pb-Pb, V0 event plane s NN = 2.76 ev Many direct photons from late stage with c 150-160 MeV? 0.1 hen large inverse slope parameter due to Doppler blueshift with typical hadronic flow velocity βflow 0.6 c? 0.0-0.1 v2(dir) smaller than v2(thermal) due to v2(prompt) 0 ALICE preliminary thermal(shen et al.) NLO (Vogelsang) + thermal(shen et al.) thermal(holopainen et al.) NLO (Vogelsang) + thermal(holopainen et al.) However, current systematic uncertainties are sizable so that there is no big puzzle looking at the ALICE data alone ALI PREL 75925 0 1 2 3 4 5 p (GeV/c) 14
he Direct Photon Puzzle he two parts of the puzzle Direct photon yields at low p (1 < p < 3 GeV/c) not described by models Large direct photon v2, similar in magnitude to pion v2, not described by hydro models Currently mostly in RHIC data, however, similar trend at the LHC Challenges standard (hydro) model of the space-time evolution and/or current photon emission rates for the QGP and the HG 15
Statistical Significance of the Puzzle: Photon Excess R is the Key Quantity Direct photon excess: ake one point (e.g. p = 2 GeV/c): /π 0 ) decay /π 0 )/(γ 3.0 2.5 0-40% Pb-Pb, s NN = 2.76 ev inc R = 1.18 ± 0.07stat ± 0.09sys R = 1.18 ± 0.114tot (γ 2.0 Direct photon double ratio NLO prediction: 1 + (N for µ = 0.5,1.0,2.0 p γ coll /γ ) direct,pp,nlo decay he excess for this point is a 1.6 sigma effect Systematic uncertainties largely correlated in p Direct photon v2: v,dec 2 =(1+")v,incl 2 v,dir 2 = v,incl 2 1 " R 1 1.5 1.0 0.5 0 2 4 6 8 12 14 (GeV/c) ALI PREL 27956 Rtrue < Rexp,today would solve both parts of the direct photon puzzle Small but finite chance that this will be the solution of the puzzle p 16
Experimental Prospects of Direct Photon Measurements in ALICE Independent measurement with e.m. calorimeters Conversions method Use absence of signal in pp to constrain material budget uncertainty Use γconv-γcalo pairs (like PHENIX) rade material budget uncertainty for calorimeter energy scale uncertainty Very useful as independent method Don t expect large reduction of systematic uncertainty in ALICE Virtual photon method 17
EMMI Rapid Reaction ask Force on the Direct Photon Flow Puzzle Feb. 2014, 25 participants (theory + experiment) Open Symposium: https://indico.gsi.de/conferencedisplay.py?confid=2662 Detailed discussions on Averaging of vn over large centrality bins, definition of vn in models Definition of decay photon cocktail in experiment and models, contribution from short-lived resonances Comparison of the space-time evolution (hydro models, PHSD, parameterized fireball evolution) pqcd contribution in various models Initial flow, near c enhancement of photon rates, bremsstrahlung photons in the hadrons gas, Glasma photons, role of fragmentation photons, Puzzle remains after checking various aspects of the data/theory comparison 18
Possible Solutions of the Direct Photon Puzzle Maybe many more photons from late stage close to c and hadron gas phase (need large increase in HG rates) [van Hees, He, Rapp, arxiv:1404.2846] heoretical justification? Maybe just bremsstrahlung from the HG? (m+m m+m+γ, m+b m+b+γ) [Linnyk, Cassing, Bratkovskaya, arxiv:1311.0279] Important source in PHSD transport model Exotic new photon source, e.g., related to large initial B field? [Basar, Kharzeev, Skokov., arxiv:1206.1334] seems unlikely to me (centrality dependence, s dependence, v3) Initial flow before hydro evolution starts, e.g., IPGlasma model? important, but does not address the missing photon yield Glasma photons, i.e., large photon production in very early gluon-rich phase? [McLerran, Schenke, arxiv:1403.7462], [Klein-Bösing, McLerran, arxiv:1403.1174] promising, but so far based on simplified models calculations from first principles needed 1 R 2 dn dyd 2 p = F (Q sat /p ) 19
Outlook Direct-photon production currently not understood Direct photons started out as probe of the early phase [Shuryak, PLB 78 (1978) 150] Large direct photon v2: Early stage contribution outshined by photons from late phase with c and below? Key question: Early or late production? Possible paradigm shift in interpretation of direct photons in A+A collisions New measurements needed, e.g., direct-photon HB 20
Extra slides 21
ALICE A Large Ion Collider Experiment EMCAL IS SSD IS SDD IS SPD FMD V0 and 0 (C side) ZDC PMD V0 and 0 (A side) HMPID IS PC dipole magnet OF RD ZDC PHOS L3 magnet muon tracking (2.5 < η < 4) Photons in ALICE Conversions (this talk) E.m. calorimeters (PHOS, EMCal) Workshop on Discovery Physics at the LHC December 5, 2014 Klaus Reygers 22
esting Exotic Photon Sources By Measuring the Photon v 3 B v γ 3 0. ALICE preliminary 0-40% Pb-Pb, s NN = 2.76 ev V0 event plane γ,decay v 3 γ,incl v 3 µ µ e ~ B 0.08 Basar, Kharzeev, Skokov., arxiv:1206.1334 0.06 0.04 decay + NLO 0.02 Phys.Rev. D50 (1994) 1901-1916 decay + NLO + thermal (Shen et al.) arxiv:1308.2111 decay + NLO + thermal (Chatterjee et al.) arxiv:1401.7464 0.00 0 1 2 3 4 5 6 ALI PREL 75777 p (GeV/c) Photon production resulting from large initial B field? Could explain v2 Expect small v3 in these models Inclusive photon v3 measured, not yet conclusive 23
Photon Conversion Method: Precise Knowledge of Material Budget Essential ) -1 (cm dn γ dr ch N 1-3 Data MC conversion candidates MC true primary conversion MC true secondary conversion MC true π 0 Dalitz MC true η Dalitz MC true combinatorics MC true hadronic bck -4-5 Beam Pipe & SPD Layer st SDD 1 Layer + Support Structures nd SDD 2 Layer st SSD 1 Layer nd SSD 2 PC Inner Containment Vessel PC Inner Field Cage Vessel PC Gas -6 ALI PUB 726 0 20 40 60 80 0 120 140 160 180 R (cm) 24
A Large Ion Collider Experiment Decay Photon Cocktail from Measured π 0 Spectrum / π 0 decay γ 3 2 1 30/07/2012 all decay γ + - π 0 γ γ (e e γ) + - - η γ γ (π π γ,e + e γ,π 0 γ γ ) ω π 0 γ (η γ ) η ρ γ (ω γ, γ γ ) 0 φ η γ (π γ, ω γ ) - ρ π + π γ 0 (π γ, η γ ) -1-2 -3-4 -5 ALI PERF 27980 0-40% Pb-Pb, s NN = 2.76 ev 0 2 4 6 8 12 14 (GeV/c) π 0 spectrum measured, heavier mesons from m scaling Only π 0, η, ω relevant, the rest is negligible p Workshop on Discovery Physics at the LHC December 5, 2014 Klaus Reygers 25
Direct Photon at RHIC: Comparison of PHENIX data with Preliminary SAR data r= direct γ/inclusive γ 0.2 0.15 0.1 (b) Au+Au (Min. Bias) SAR, arxiv:1405.3940 PHENIX, arxiv:0912.0244 0.05 0 1 1.5 2 2.5 3 3.5 4 4.5 (GeV/c) p 26
Direct Photon Spectra at RHIC and the LHC 1 d 2 N dp dy [(GeV / c) 2 ] 1 2π p 0 1 2 3 4 5 6 7 ALICE PbPb 0-40% preliminary PHENIX pp PHENIX AuAu min. bias 8 9 0 2 4 6 8 12 14 p [GeV / c] 27