Shingo Sakai for PHENIX collaboration Univ. of Tsukuba

1. Azimuthal anisotropy of electrons in Au+Au collisions at √SNN=200GeV/cmeasured with PHENIX at RHIC Shingo Sakai for PHENIX collaboration Univ. of Tsukuba

2. gconversion p0 gee h gee, 3p0 w ee, p0ee f ee, hee r ee h’  gee page1 Abstract The measurement of electrons and positrons at transverse momenta above 1.0 GeV/c allows to study the production of heavy flavor quark-anti quark pairs via the semileptonic decays of charmed particles [1]. The azimuthal anisotropy of high pt electrons can carry information about the anisotropy of the parent charmed mesons. The observation of charm flow would indicate that collective motion develops already in the partonic phase of the collision. [1] (PHENIX: PRL 88(2002)192303) *Detail of single electron studies and new results in 200 GeV Au+Au Takashi Hachiya’s Poster

3. c b direct g (J. Alam et al. PRC 63(2001)021901) The PHENIX experiment has the unique capability to measure electrons and positrons at RHIC. In this analysis we study the elliptic flow pattern of inclusive electrons from Au+Au collisions at √SNN= 200 GeV as function of transverse momentum and estimate “non-photonic” - charm & beauty - electron v2 subtracted all other electron sources, such as Dalitz decays and photon conversions. PYTHIA

4. page2 Overview of PHENIX • Reaction plane is determined • by BBC • The tracking is done with DC+PC • In the PHENIX experiment electrons are identified as Cherenkov light by RICH. • RICH • -CO2 • -0.2 <pt <5.0 • (this analysis 0.4<pt<4.0) • - |y|  0.35

5. page3 dn/d of e distribution In this analysis electron v2 is estimated by using reaction plane method. (measure azimuthal angle of each electrons with respect to the reaction plane ) Fig. shows the azimuthal distribution of electrons which is measured with respect to the reaction plane (upper red). The blue histogram is miss ID electron. The bottom distribution is result which is subtracted miss ID electron and fitted by dn/d = N(1+2v2cos(2* )) e+(e-) candidate Miss ID e+(e-)

6. page4 Pt dependence of e v2 Fig. shows pt dependence electron v2. The error bars reflect statistical errors only. The data points are plotted at the center of gravity of the bins as determined from the pt distribution and the horizontal error bar is RMS. The yellow line is systematic error. The electrons v2 are including “photonic” & “non-photonic” electrons v2. Dalitz decays Di-electron decays Photon conversions Kaon decays Thermal dileptons charm decay beauty decay “photonic” “non-photonic”

7. page5 Comparison with v2 of hadrons Fig. shows the comparison electron v2 and hadron v2 (pion,proton). @Low pt (pt<1.0GeV/c) v2(e) is larger than v2(pion)& v2(proton) @high pt region This region particular interest because of the contributions from heavy-quark(c/b) decays is large.! v2

8. page6 Non-photonic electron v2 In this analysis we compare with electron v2 with simple model which is assuming charmed electron v2 is zero. And we also estimate non-photonic electron v2. The azimuthal distribution of electron gives as; dne/d = dnpho./d+ dnnon-pho./d From the Eq. electron v2 is given as; v2(e) = rv2(pho.)+(1-r)v2(non-pho.). Here r is ratio of Ne/Npho.(shown right plot) and v2(pho.) is photonic e v2. The value is estimated by simulation (next page ). ratio of Ne/Npho @√sNN=200GeV (T.Hachiya)

9. page7 (1) photonic e v2 --- electron v2 from π0 and K The dominant sources at low pt are photon conversions and Dalitz decays of pi0 [1]. Electrons from Kaon decays contribute a few % at Low pt. These electrons are also taken into consideration. Electron v2 from pi0 and K is estimated by simulation. electron v2 (simulation) from pi0 (PHENIX preliminary ) electron v2 (simulation) from K ( nucl-ex/0305013)

10. (2) e v2 compare with simple model page8 Fig. shows the comparison electron v2 and model v2 (electron v2 assuming non-photonic e v2=0). From the previous Eq. the v2 is estimated as; v2= rv2(pho.) The dashed line means systematic error of the model.

11. page9 (3) non-photonic e v2 Fig. shows non-photonic electron v2 which is obtained by subtracting photonic v2 (pi0 & K decay electron) from electron v2. v2(non-pho.) = {v2(e)-rv2(pho.)}/(1-r) The result is compared with two prediction of charmed electron & D meson v2 which is based quark coalescence model . v2(D)≈v2(light) + v2（charm） charm quark light quark

12. page10 (4) non-photonic e v2 compare with model(1) Charmed e v2 has been predicted (nucl-th0312100) with two scenario for charm quark distribution (1)thermalization with transverse flow (2) No reinteraction (PYTHIA spectra) Here non-photonic e v2 is compared with the scenario.

13. page11 (5) non-photonic e v2 compare with model(2) • D meson v2 has been • predicted (nucl-th0304045) • D v2 (charm v2>0) • D v2 (charm v2=0) • From the result we estimated • charmed electron v2 (triangles) . • Here non-photonic e v2 is • compared with D v2 & charm e v2

14. page12 Summary In this analysis we have measured the electron v2 by using the reaction plane method. With increasing pt the contribution from charm decays to the inclusive electron sample grows. We have compared the inclusive electron v2 with a simple model assuming that the charm v2 is zero. The model is consistent with the data within error bars. We have also determined the "charmed electron" v2 by subtracting the "photonic electron" v2 from the inclusive data. The observed v2 of electrons from charm decays is consistent within errors with various model calculations which assume extremely different scenarios, no reinteraction of the initially produced charm quarks or complete thermalization with the bulk matter.