Non-photonic electron-hadron azimuthal correlation for AuAu, CuCu and pp collisions at GeV
We present preliminary STAR results of azimuthal correlations between non-photonic electrons and hadrons in AuAu, CuCu and pp at GeV. Comparison of the e-h correlations from these colliding systems allows one to study the system-size dependence of heavy quark jet-medium interactions. We also report on the relative charm and bottom contributions to non-photonic electrons extracted from correlations measured in pp collisions. Our results, when combined with measurements of non-photonic electrons, constrain the charm and bottom energy loss in the dense medium.
Data for di-hadron correlations in AuAu and dAu using the STAR detector show a suppression of high- hadron yields and modifications in the azimuthal correlation in central AuAu collisions PRL98 (); PRL96 (). On the away-side of the trigger particles, the observed broadening in the correlations reflects the presence of strong jet-medium interactions, possibly a Mach cone effect ZYAM (); Starter (). Non-photonic electron triggered particle correlations probe heavy quark jet-medium interactions. Electron-hadron (e-h) correlations on the near side, triggered by non-photonic electrons from charm and bottom decays, have different patterns due to the large mass difference between and mesons. Using e-h correlations from pp collisions in comparison with PYTHIA calculations, we can estimate the relative and contributions to the non-photonic electrons.
2 Analysis Technique
The data sets used were collected by STAR at RHIC in 2007 for AuAu, 2006 for CuCu and 2005 ( GeV/c) and 2006 ( GeV/c) for pp. The Time Projection Chamber (TPC) TPC (), the heart of STAR, was used to reconstruct the tracks of charged particles. Additionally the Barrel Electromagnetic Calorimeter (BEMC) and Shower Maximum Detector (SMD) Barrel (), the electromagnetic calorimeters which surround the TPC in full azimuth, were required to have a least one track projected from the TPC onto it and also have an energy deposition greater than a predetermined minimum (i.e. high-tower threshold). This is to increase the number and purity of high electrons. For CuCu the threshold was at 3.75 GeV, 5.5 GeV for AuAu and 5.4 GeV for pp. For pp and AuAu the pseudorapidity used was , while for CuCu it was due to the partial installation of the BEMC/SMD. For the AA collisions we used a 0-20% centrality cut, the definition of which is explained in Centrality ().
To identify electron candidates one used a combination of measurements: ionization energy loss (dE/dx) in the TPC, ratio of particle momentum to energy deposited in the BEMC and lastly the electromagnetic shower size in the SMD—see Method (); Dong () for more detail.
The primary background are photonic electrons either coming from photon conversions inside STAR or the Dalitz decays of and . In both cases the electron pairs have a small invariant mass. This background is removed by pairing electron candidates with a track having passed a loose dE/dx cut around the electron ionization band. The distribution of 2D invariant masses ( is ignored to minimize tracking resolution effects Dong ()) for opposite sign (OppSign) pairs is obtained upon which a cut of GeV/ is applied to reject most photonic pairs. The combinatorial background is estimated using the 2D invariant mass distribution of same sign (SameSign) pairs.
The azimuthal correlation of non-photonic electrons and hadrons begins with a semi-inclusive (semi-inc.) electron sample, which is the inclusive electron sample minus the OppSign background (after having applied the mass cut).
The correlation is done via the following method:
is the same sign combinatorial background, while are the photonic electrons which weren’t reconstructed due to inefficiencies , which is the photonic electron reconstruction efficiency estimated via simulations and found to be for AuAu, for CuCu and for pp. There is only a minor transverse momentum dependence of the efficiency in the momentum range being looking at. More details on this method using a semi-inclusive electron sample are in Electrons (); Electrons2 ().
Figures 1 and 2 plot the e-h azimuthal correlation for AuAu and CuCu at 200 GeV. The left panels show the raw correlations along with dashed curves for elliptic flow () v2 () and zero yield at minimum (ZYAM) ZYAM (). Systematic uncertainty of is determined by using a lower limit of zero and upper limit of 80% for AuAu or 60% for CuCu of charged hadron . The momentum ranges used are 3 GeV/c GeV/c for trigger electrons and 0.15 GeV/c GeV/c for hadrons in AA collisions. To decrease the error bars, the correlation is folded into [0, ] and the data points beyond are reflections. One clearly already sees a modification of the away-side. The right panels plot the correlation after the subtraction and ZYAM application. On the away-side (around ) instead of a single peak there is a broadening (AuAu) or possible double-peak (CuCu) structure. Using a PYTHIA calculation for pp to fit this one sees the /ndf is rather poor. This modification of the away-side is similar to the di-hadron case in AuAu Starter () and probably indicates heavy quark interaction with the dense medium.
Fitting the near-side of the e-h correlation over various trigger ranges with a PYTHIA simulation (fitting method is described in Electrons (); Electrons2 ()) one sees equal charm and bottom contribution above 5 GeV/c, Figure 3 (a). Combining this with the previously measured for non-photonic electrons (Method (), Figure 3 (c)) one has
Figure 3 (b) shows the most probable values for and with the dashed line being the 90% confidence limit (taking and uncertainties into consideration). Even if D decay electrons are fully suppressed . This indicates B meson yields are suppressed at high in heavy ion collisions presumably due to quark energy loss in the dense matter Model1 (), modification of the fragmentation process due to the dense medium and/or dissociation/absorption of heavy flavor hadrons in the dense medium during the evolution Model2 (). From it one can see that electrons coming from B decays are as suppressed as those from D decays. Models I, II and III are described in Model1 (), Model2 () and Model3 () respectively.
- (1) B. I. Abelev et al., Phys. Rev. Lett. 98 (2007) 192301.
- (2) S. S. Adler et al., Phys. Rev. Lett. 96 (2006) 032301.
- (3) J. Adams et al., Phys. Rev. Lett. 95 (2005) 152301.
- (4) B.I. Abelev et al., STAR Collaboration, Phys. Rev. Lett. 102 (2009) 52302.
- (5) M. Anderson et al., Nucl. Instrum. Methods A 499 (2003) 659.
- (6) M. Beddo et al., Nucl. Instrum. Methods A 499 (2003) 725.
- (7) J. Adams et al., Phys. Rev. C 72 (2005) 014904.
- (8) B. I. Abelev et al., Phys. Rev. Lett. 98 (2007) 192301.
- (9) W. J. Dong Ph. D. Thesis (2006) UCLA.
- (10) X. Lin et al., J. Phys. G: Nucl. Part. Phys. 34 (2007) S821.
- (11) X. Lin Ph. D. Thesis (2007) IOPP, China.
- (12) A. M. Poskanzer and S. A. Voloshin, Phys. Rev. C 58 (1998) 1671.
- (13) M. Djordjevic, Phys. Rev. B 632 (2006) 81.
- (14) A. Adil and I. Vitev, Phys. Rev. B 649 (2007) 139.
- (15) H. Hees, V. Greco and R. Rapp, Phys. Rev. C 73 (2006) 034913.
- (16) M. Cacciari, P. Nason and R. Vogt Phys. Rev. Lett. 95, (2005) 122001.
- (17) A. Mischke et al. (STAR Collaboration), J. Phys. G: Nucl. Part. Phys. 35 (2008) 104117.
- (18) A. Mischke, Phys. Lett. B 671 (2009) 361.