Elliptic flow of electrons from heavy flavor decays

Elliptic flow of electrons from heavy flavor decays Shingo Sakai for PHENIX Collaborations (Univ. of Tsukuba JSPS) 2006/6/4

Elliptic flow (v 2) ｙ n Elliptic flow ｘ d. N/dφ ∝ N 0(1+2 v 2 cos(2φ)) Initial spatial anisotropy n A powerful probe of the initial state of the high energy heavy ion collision n transfer initial spatial anisotropy to momentum space anisotropy p macroscopic ; hydro model => pressure gradient p microscopic => scattering in the medium 2006/6/4 pｙ pｘ Momentum space anisotropy of particle emission 2

v 2 already developed in partonic phase ? n identified hadrons v 2 after scaling p. T and v 2 number of quarks n v 2 after scaling fall on same curve n v 2 already formed in the partonic phase for hadrons made of light quarks (u, d, s) => partonic level v 2 n charm quark also flow @ RHIC ? 2006/6/4 3

Charm quark n Charm is believed to be produced in initial collisions via gluon fusion => total cross-section ; binary scaling (Ncoll) n Charm propagates through medium created in the collisions => good probe of medium n initial charm v 2 might be 0 Phys. Rev. Lett. 94: 082301, 2005 => charm quark v 2 due to scattering (PHENIX) in medium => non-zero charm v 2 indicate very high dense medium created in the collision and quark level thermalization 2006/6/4 4

Charm quark study @ PHENIX n Electron sources charm decay non-photonic beauty decay Dalitz decays Di-electron decays Photon conversions Kaon decays Thermal dileptons photonic n　Subtract photonic electrons following methods p“Cocktail subtraction” – calculation of “photonic” electron background from all known sources p“Converter subtraction”– extraction of “photonic” electron background by special run with additional converter 　(brass, X = 1. 7%) 2006/6/4 5

Charm quarks interact with medium ? n non-photonic electron RAA n clear suppression @ high p. T in more central collisoion (RAA < 1. 0) => one of the evidence charm quarks interact with the medium n low p. T and peripheral collision (centrality > 40 %) is consistent with binary scale n v 2 measurement (p. T, cent) gives us additional info. of the interaction between charm and the medium 2006/6/4 6

Electron v 2 measurement @ PHENIX n Electron v 2 is measured by R. P. method d. N/d( - ) = N (1 + 2 v 2 obscos(2( - ))) n R. P. --- determined with BBC n Tracking (p. T, φ)　--- DC + PC n electron ID --- RICH & EMCal e- e. ID @ RICH After subtract B. G. Fig : Energy (EMcal) & momentum matching of electrons identified by RICH. Clear electron signals around E/p-1 = 0 2006/6/4 (E-p/p/sigma) distribution 7

Non-photonic electron v 2 measurement n Non photonic electron v 2 is given as; d. Ne/d = d. Npho. e /d + d. Nnon-pho. e /d => v 2 enon-γ = {(1+RNP) v 2 - v 2 eγ} } / RNP p v 2 --- inclusive electron v 2 p RNP --- e (non-γ)/ e（γ） p v 2 eγ --- photonic electron v 2 e (non-γ) / e (γ) Run 04: X=0. 4% n photonic electron v 2 determination Run 02: X=1. 3% pconverter method (experiment) Measure inclusive electron v 2 with/without converter. (QM 05 F. Kajihara) Then separate non-photonic & photonic e v 2 pcocktail method (simulation) Determined photonic electron v 2 with simulation 2006/6/4 8

Inclusive electron v 2 (in/out converter) n inclusive electron v 2 in / out converter n difference between converter in / out => photonic & non-photonic e v 2 is different 2006/6/4 9

Inclusive electron & photonic electron v 2 (converter method) photonic electron v 2 (simulation ; pi 0 -> e) 　 inclusive electron v 2 (converter out) ○ n compare with inclusive & photonic electron v 2 n photonic e v 2 ; p. T < 1. 0 (conv. ) & p. T > 1. 0 (cock. ) n inclusive electron v 2 is smaller than photonic electron v 2 2006/6/4 10

Non-photonic electron v 2 n Non-photonic electron v 2 : p. T < 1. 0 (conv. ) & p. T>1. 0 (cock. ) n Non-photonic electron v 2 has non-zero v 2 2006/6/4 11

Non-photonic electron v 2 (centrality dep. ) n non-photonic electron v 2 seems like getting larger with centrality 2006/6/4 12

Compare with RAA n RAA ~ 1. 0 @ peripheral collision but v 2 still non-zero n indicate charm quarks interact with medium not only central but also peripheral 2006/6/4 13

D meson flow ? simulation Phys. Lett. B 597: 328 -332, 2004 n electron v 2 from D meson decay well reflect parent v 2 n non-zero non-photonic electron v 2 suggest D meson also flow @ RHIC 2006/6/4 14

expected D v 2 from non-photonic e v 2 n expected D v 2 from non-γ v 2 n process (1) D v 2 = a*f(p. T) a ; free parameter f(p. T) ; pi, K, p (2) D -> e v 2 (3) Calculate χ2 (4) Find χ2 minimum for “a” π, deuteron v 2 --- PHENIX preliminary n　expected D meson v 2 from non-photonic electron v 2 p smaller than pi v 2 p larger than Deuteron v 2 “if D v 2 shape same as pi, k, p” Mass D meson~ Mass Deuteron 2006/6/4 15

Charm quark flow ? n Compared with quark coalescence model prediction. with/without charm quark flow (Greco, Ko, Rapp: PLB 595 (2004) 202) - No Bottom contribution - c v 2 small u v 2 @ low p. T - quark v 2 flat @ high p. T n Below 2. 0 Ge. V/c ; consistent with charm quark flow model. => favor charm quark flow model p. T[Ge. V/c] n assume c v 2 < u v 2 @ low p. T => mass effect in partonic level ? 0 2006/6/4 1 2 3 4 p. T[Ge. V/c] 16

v 2 B meson contribution ? D & B v 2 - assume D & B v 2 same Hendrik, Greco, Rapp nucl-th/0508055 w. o. B meson (c flow) D -> e w. B meson (c, b flow) n based on quark coalescence model n c, b reso ; c, b get v 2 by elastic scat. in QGP B -> e p. T n B meson contribution is model dependent now but the contribution is getting larger at high p. T. n electron v 2 from B meson is smeared due to large mass difference => high p. T single electron v 2 is getting smaller ? 2006/6/4 17

Near future n single μ v 2 n new reaction plane detector p good resolution => reduce error from R. P. p J/ψ v 2 & high p. T non-photonic electron v 2 n silicon vertex detector p direct measurement D meson v 2 PHENIX muon arm m a be [Silicon vertex detector] 2006/6/4 [Reaction plane detector] 18

Summary n Non-photonic electron v 2 from heavy flavor decays has been measured with RHIC-PHENIX n Non-photonic electron v 2 has non-zero v 2 => indicate non-zero D meson v 2 n Clear centrality dependence of non-photonic e v 2 n Compare with model calculation with charm flow (c v 2 < u v 2) => consistent with charm flow model below 2. 0 Ge. V/c => mass effect in partonic level ? n Estimate D meson v 2 from non-photonic electron v 2 assuming D meson v 2 shape same as pion, Kaon, proton. 2006/6/4 19

n Backup slides 2006/6/4 20

Converter method Separate non-photonic & photonic e v 2 by using Non-converter run & converter run Non-converter ; Nnc = Nγ+Nnon-γ Converter ; Nc = R *Nγ+Nnon-γ (1+RNP)v 2 nc = v 2γ + RNPv 2 non-γ (R +RNP) v 2 c = R v 2γ + RNPv 2 non-γ R --- ratio of electrons with & without converter (measured) RNP --- non-photonic/photonic ratio (measured) v 2 nc --- inclusive e v 2 measured with non-converter run (measured) 　v 2 c --- inclusive e v 2 measured with converter run (measured) v 2 non-γ(non-photonic) & v 2γ(photonic) is “experimentally” determined ! 2006/6/4 21

Cocktail method Determined photonic electron v 2 with simulation Then subtract it from electron v 2 measured with non-converter run d. Ne/d = d. Npho. e /d + d. Nnon-pho. e /d v 2 non-γ = {(1+RNP) v 2 - v 2γ} } / RNP measured calculate RNP --- non-photonic/photonic ratio experimentally determined v 2 --- inclusive electron v 2 (without converter) v 2γ --- photonic electron v 2 calculated from pi 0 (pion) v 2 2006/6/4 22

D-> e v 2 D meson v 2 electron v 2 Decay 2006/6/4 23

D v 2 estimate from non-γ e v 2 (2) Minimum ; 70 % Minimum ; 75 % Minimum ; 65 % n χ２/ndf v. s. scale factor (a) n current result can’t decide what shape is better. 2006/6/4 24

New reaction plane detector PHENIX muon arm 2006/6/4 n. A new reaction plane detector n position ; 1~|h|~2. 5, n resolution ; <cos 2 D > ~ 0. 7 (BBC ; <cos 2ΔΦ> ~ 0. 3) n materials ; Pb converter + scintillator 25