1. HFT + TOF: Heavy Flavor Physics Yifei Zhang University of Science & Technology of China Lawrence Berkeley National Lab TOF Workshop, Hangzhou, 26-29 April, 2009 STAR D0D0 K  VpVp HFT: Heavy Flavor Tracker HFT = IST + PXL IST: Inner Silicon Tracker PXL: Pixel detector

2. Outline:  Motivation Why measure heavy flavor? Why TOF plays crucial role?  Simulation with HFT + TOF Open charm reconstruction D&B  e simulation  Summary

3. October 7, 2015 Yifei Zhang USTC/LBNL 3 October 7, 2015 3 Motivation - Heavy quark masses >> T c,  QCD, M uds M b  4.8 GeV M c  1.5 GeV ① Higgs mass: electro-weak symmetry breaking. (current quark mass) ② QCD mass: Chiral symmetry breaking. (constituent quark mass)  Heavy quark masses are not modified by QCD vacuum. Strong interactions do not affect them.  Important tool for studying properties of the hot-dense matter created at RHIC energy. X.Zhu, et al., PLB 647 (2007) 366

4. October 7, 2015 Yifei Zhang USTC/LBNL 4 October 7, 2015 4 Heavy quark e-loss and v 2 The R AA of single electron from heavy flavor decay has the similar suppression as that of light flavor hadrons. Spectra, R AA (D  e) & R AA (B  e)? => heavy quark energy loss mechanism, heavy quark interaction with medium. v 2 (D  e) & v 2 (B  e)? => light flavor thermalization, drag constants. Directly measure D is not a problem with HFT STAR PRL 98 (2007) 192301 E-loss: b < c < q

5. October 7, 2015 Yifei Zhang USTC/LBNL 5 Measure B &  c Important for understanding the bottom contribution in current NPE measurements. Large systematic errors for both theory (FONLL) and data (STAR e-h correlation). Need improve the measurement accuracy. Measure this ratio directly from spectra.  No B meson spectra measured.  Separately measure B  e spectrum will indirectly measure B meson spectrum from its decay kinematics.  B  e = NPE  D  e STAR HFT has the capability to measure D 0 decay vertex topologically via hadronic decay channel. Measured D 0 spectrum constrains D  e.   c yield,  c / D 0 enhancement, di-quark? Lee, et. al, PRL 100 (2008) 222301

6. October 7, 2015 Yifei Zhang USTC/LBNL 6 Why TOF plays crucial role? K,  PID for D meson reconstruction. TPC K,  PID up to ~ pt = 0.7 GeV/c. TOF extend pt up to ~ 1.5 GeV/c. TPC proton PID up to ~ pt = 1.1 GeV/c. Proton can be reject up to 3 GeV/c by TOF. e PID for NPE, B meson reconstruction. TOF 1/  cut rejects hadron significantly. TOF 1/  cut (a) (b) (c)  e K p

7. October 7, 2015 Yifei Zhang USTC/LBNL 7 STAR Detector TPC – p, tracking HFT – decay vertex TOF – PID Large acceptance Mid-rapidity |  | < 0.9 Full barrel coverage 0 <  < 2 

8. October 7, 2015 Yifei Zhang USTC/LBNL 8 Inner Tracking Detectors  SSD Silicon Strip Detector, existing single layer detector, double side trips.  IST 500  m x 1cm strips along beam direction, 24 liquid cooled ladders equipped with 6 silicon strip-pad sensors.  PIXEL double layers, 18.4x18.4  m pixel pitch, 2 cm x 20 cm each ladder. Deliver ultimate pointing resolution, hit density for 1st layer ~ 60 cm. -2 Detector Radius (cm) Hit Resolution R/  - Z (  m -  m) Radiation length SSD2330 / 8571% X 0 IST14170 / 17001.2 %X 0 PIXEL 88.6 / 8.6~0.3 %X 0 2.58.6 / 8.6~0.3% X 0

9. October 7, 2015 Yifei Zhang USTC/LBNL 9 Simulation with HFT + TOF Full barrel TOF Radius: ~ 200 cm 120 trays 32 modules/tray Provide good PID capability for open charm daughters. SSD + IST + PXL Radius: 23, 14, 8, 2.5 cm New HFT geometry, provide good secondary vertex reconstruction capability.

10. October 7, 2015 Yifei Zhang USTC/LBNL 10 TOF geometry in STAR GEANT

11. October 7, 2015 Yifei Zhang USTC/LBNL 11 TOF PID Simulation New TOF geometry Smeared timing resolution 100ps TOF acceptance is ~87%. Weak decays or scattered out of acceptance ~12%, some of them were compensated. m 2 cuts were used for , K, p PID. K,  PID upto 1.5 GeV/c Good proton rejection upto 3 GeV/c High p T cut rejects some of the proton contamination. Misidentification of tracks were taken into account for background. 0.2 < p T < 3 GeV/c

12. October 7, 2015 Yifei Zhang USTC/LBNL 12 Simulation Performance pointing resolution in r-  to primary vertex for single particles (of , K, p.) including all hits in HFT. Tracking efficiency of single  + for 3 pileup hits densities.

13. October 7, 2015 Yifei Zhang USTC/LBNL 13 Hadronic channels STAR HFT has the capability to reconstruct the displaced vertex of D 0  K  (B.R.=3.8%, c  =123  m) and  c   Kp (B.R.=5.0%, c  =59.9  m) D0D0 cc

14. October 7, 2015 Yifei Zhang USTC/LBNL 14 D 0 reconstruction (a)D 0 efficiency vs p T (b)D 0 significance vs p T (c)D 0 mass at 3.0 < p T < 3.5 GeV/c before background subtraction. (a) (c) (b)

15. October 7, 2015 Yifei Zhang USTC/LBNL 15 Error estimate of D 0 v 2 and R cp Assuming D 0 v 2 distribution from quark coalescence. 500M Au+Au m.b. events at 200 GeV. Charm collectivity => Medium properties, light flavor thermalization. Assuming D 0 R cp distribution as charged hadron. 500M Au+Au m.b. events at 200 GeV. Charm energy loss => Energy loss mechanisms, QCD in dense medium.

16. October 7, 2015 Yifei Zhang USTC/LBNL 16  c reconstruction TOF PID in progress Good D 0 p T distribution measurement as reference. The unique charmed baryon.  c / D 0 ratio => enhancement?

17. October 7, 2015 Yifei Zhang USTC/LBNL 17 Additional Capability -- Semi-leptonic Channels particle c  (  m) Mass (GeV) q c,b →x (F.R.) x →e (B.R.) D0D0 1231.8650.540.0671 D±D± 3121.8690.210.172 B0B0 4595.2790.400.104 BB 4915.2790.400.109 The distance of closest approach to primary vertex (dca): Due to larger c , B  e has broader distribution than D  e Dca of D +  e is more close to that of B  e B.R. = Branching Ratio F.R. = Fragmentation Ratio dca

18. October 7, 2015 Yifei Zhang USTC/LBNL 18 Electron efficiency TPC tracking efficiency is included. W/o PXL hits required, efficiency ~ 75% With PXL hits required, efficiency ~ 61% D 0  e p T weighted ~60%

19. October 7, 2015 Yifei Zhang USTC/LBNL 19 Dca distributions Electrons: PID by TPC+TOF, |  | < 0.9, 2 PXL hits required. Photonic background can be removed from its small invariant mass character combining a pair of electrons. Other background is small. Normalized by the F.R. and B.R., and total electron yield was normalized to STAR measured NPE spectrum. (B  e) / NPE ratio was normalized to fit STAR measured data (from e-h correlation).

20. October 7, 2015 Yifei Zhang USTC/LBNL 20 Errors estimate of spectra In real experimental data, we can use the different dca distributions to fit the total dca distribution to extract the raw yield of each source of electrons. From the dca distributions and the efficiency, the D  e, B  e and B  D  e spectra can be obtained, and the statistical errors were estimated for 100M Au+Au central 200 GeV events (non-special trigger). R AA can be measured directly from the spectra with D  e, B  e separated. Understanding the heavy quark energy loss mechanisms.

21. October 7, 2015 Yifei Zhang USTC/LBNL 21 Errors estimate of (B  e)/NPE (B  e)/NPE ratio can be directly measured from spectra. The statistical errors are estimated for 100M Au+Au central 200 GeV events. We will have high p T electron trigger (EMC HT) in the future, high p T statistics will not be a problem.

22. October 7, 2015 Yifei Zhang USTC/LBNL 22 Measure v 2 from dca B  e v 2 and D  e v 2 can be measured from different dca cuts. For example: CaseCut (cm)e(D) eff. (%)e(B) eff. (%)r = e(B)/NPE I< 0.00545.522.30.325 II> 0.0215.339.60.718 r  v 2 (B) + (1-r)  v 2 (D) = v 2 (NPE) v 2 (NPE) is the total non-photonic electron v 2 after dca selection. v 2 (B) is B  e v 2 v 2 (D) is D  e v 2

23. October 7, 2015 Yifei Zhang USTC/LBNL 23 Error estimate for v 2 Heavy quark collectivity Study charm and bottom separately to understand the mass effect of such heavy quarks. Probe medium properties. Assuming D meson v 2, using decay form factor to generate D  e v 2 distributions. r  v 2 (B) + (1-r)  v 2 (D) = v 2 (NPE) v 2 (D) is D  e v 2 v 2 (B) is B  e v 2

24. October 7, 2015 Yifei Zhang USTC/LBNL 24 October 7, 2015 24  Topologically measure D 0 is important to measure charm collectivity and energy loss directly via hadronic channel reconstruction, and to provide a good reference for B measurement from electron channel.  Topologically measure  c yield will provide us the information on charmed baryon / meson ratio.  Measure D  e and B  e is important for understanding heavy quark physics with charm and bottom separately at RHIC. Summary TOF provide excellent PID capability. HFT + TOF has great performance to do it!!