Xiaoyan Lin 林晓燕 (for the STAR Collaboration) Central China Normal University Wuhan, P.R. China

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20061

Study B and D Contributions to Non-photonic Electrons via Azimuthal Correlations between Non-Photonic Electrons and

Charged Hadrons Xiaoyan Lin

林晓燕(for the STAR Collaboration)

Central China Normal University

Wuhan, P.R. China

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20062

Motivation Data analysis

---- Electron identification

---- Photonic electron background

---- Electron-hadron correlations

Preliminary results of B/(B+D) Summary

Outline

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20063

Heavy quark RAA has the similar magnitude as light quark RAA.

The high pT region non-photonic electron RAA is surprising !

Where is the bottom contribution?

Features in Heavy Quark Measurements at RHIC----Non-Photonic Electron RAA

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20064

The decay kinematics of D and B mesons are different!

The same D and B v2 can lead to very different non-photonic electron v2 !

Features in Heavy Quark Measurements at RHIC----Non-Photonic Electron v2

Y. Zhang, hep-ph/0611182PYTHIA

Reduction of v2 at pT > 2 GeV/c.

Bottom contribution??

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20065

Quantitative understanding of features in heavy quark measurements requires experimental measurement of B and D contributions to non-photonic electrons !

Such information should be best obtained from direct measurement of hadronic decays of charm and bottom mesons.This motivates the STAR vertex detector upgrade! See Talk by Andrew Rose (1.4)

B and D Contributions to Electrons

Poor (wo)man’s approach to measure B/D contributions to non-photonic electrons

---- e-h correlationsX.Y. Lin, hep-ph/0602067

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20066

PYTHIA Simulation of e-h Correlations

B

D

Associated pT > 0.3 GeV/c. Significant difference in the near-side correlations. Width of near-side correlations largely due to decay kinematics.

X.Y. Lin, hep-ph/0602067

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20067

Major Detectors Used

Time Projection Chamber (TPC) Electro-Magnetic Calorimeter (EMC) Shower Maximum Detector (SMD)

Data Sample:

p+p collisions at sNN = 200 GeV in year 5 run.

2.37 million EMC HT1 triggered events with threshold 2.6 GeV; 1.68 million EMC HT2 triggered events with threshold 3.5 GeV.

Signal: Non-photonic electron

Background: HadronPhotonic electron

Charm decay

Bottom decay

Photon conversionπ0 Dalitz decayη Dalitz decaykaon decayvector meson decays

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20068

Electron ID Using TPC and EMC

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 20069

The purity of electron sample is above 98% up to pT ~ 6.5 GeV/c.

Electron ID Using TPC and EMC

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200610

The combinatorial background is small in p+p collisions. Reconstructed photonic = Opposite sign – Same sign. Photonic electron = reconstructed-photonic/ ε. ε is the background reconstruction efficiency calculated from simulations.

m<100 MeV/c2

Photonic Background

Electron candidates are combined with tracks passing a loose cut on dE/dx around the electron band. The invariant mass for a pair of photonic electrons is small.

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200611

All Tracks

Inclusive electron

Pass EID cuts

Non-photonic electron Photonic electron

Reco-photonic electron=OppSign - combinatorics

Not-reco-photonic electron=(1/eff-1)*(reco-photonic)

Procedure to Extract the Signal of e-h Correlations

Semi-inclusive electron

Signal:non-photonic = semi-inclusive +combinatorics-(1/eff-1)*reco-photonic Each item has its own corresponding Δφ histogram.

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200612

e-h Azimuthal Correlations after Bkgd. Subtraction

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200613

Use PYTHIA Curves to Fit Data Points

Fit function: R*PYTHIA_B+(1-R)*PYTHIA_D R is B contribution, i.e. B/(B+D), as a parameter in fit function.

B

D

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200614

B/(B+D) consistent varying fit range.

Use PYTHIA Curves to Fit Data Points

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200615

Preliminary Results: B Contribution .VS. pT

Error bars are statistical only! Data uncertainty includes statistic errors and systematic uncertainties from:

---- photonic background reconstruction efficiency (dominant).

---- difference introduced by different fit functions. Preliminary data is within the range that FONLL calculation predicts. Non-zero B contribution is observed.

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200616

Non-photonic electron and charged hadron correlations are sensitive to D and B contributions to non-photonic electrons.

We have measured e-h correlations in 200 GeV p+p collisions.

The preliminary data indicates at pT ~ 4-6 GeV/c the measured B contribution to non-photonic electrons is comparable to D contribution based on PYTHIA model.

Our measurement of B/(B+D) provides a constraint to the FONLL prediction.

Summary

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200617

Backup slides

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200618

Δφnon-pho = Δφsemi-inc + Δφcombinatorics - Δφnot-reco-pho

= Δφsemi-inc + Δφcombinatorics - (1/eff -1) *Δφreco-pho-no-partner

Method to Extract the Signal of e-h Correlations

non-pho. e = semi-incl. e +combinatorics - not-reco-pho. = semi-incl. e +combinatorics - (1/eff-1)*reco-pho.

Note Δφnot-reco-pho = (1/eff -1) *Δφreco-pho-no-partner! Δφreco-pho-no-partner is the reco-pho after removing the conversion partner. The photonic background has two parts: reco-pho and not-reco-pho. In electron yield or v2 analysis, the not-reco-pho part can just be calculated by reco-photonic part after an efficiency correction, i.e. not-reco-photonic = (1/eff-1)*reco-pho. However, in e-h correlation analysis, that is different. The reco-pho electron means we find the conversion partner, while the not-reco-pho electron means we miss the conversion partner. The resulting e-h correlations for these two parts are different. If we use reco-pho part to calculate the not-reco-pho part, we have to remove the conversion partner of reco-pho part.

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200619

The distributions of ChiSquare .VS. ratio_B

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200620

The distributions of ChiSquare .VS. ratio_B

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200621

Preliminary Results: B Contribution .VS. pT

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200622

-3σ < z distance < 3σ and -3σ < φdistance < 3σ were set to remove lots of random associations between TPC tracks and BEMC points.

Electron Identification: Projection Distance

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200623

PYTHIA Simulation: e pT .VS. parent pT

C-quark needs to have larger momentum than b-quark to boost the decayed electron to high pT.

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200624

PYTHIA Simulation: e pT .VS. hadron pT

The efficiency of associated pT cut is different between D decay and B decay. Therefore, it is better to use lower pT cut on the associated particles in order to avoid analysis bias!

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200625

PYTHIA Simulation: e pT .VS. hadron pT

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200626

PYTHIA parameters used in this analysis

PYTHIA version: v6.22

δ fragmentation function used for both charm and bottom.

Parameters for charm: PARP(67) = 4 (factor multiplied to Q2)

<kt> = 1.5 GeV/c

mc = 1.25 GeV/c2

Kfactor = 3.5MSTP(33) =1 (inclusion of K factor)MSTP(32) = 4 (Q2 scale)CTEQ5L PDF

Parameters for bottom are the same as for charm except

mb = 4.8 GeV/c2.X.Y. Lin, hep-ph/0602067

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200627

Near-side width due to decay kinematics

All hadrons Hadrons from D

Backgroundwith δ fragmentation function

Xiaoyan Lin Quark Matter 2006, Shanghai, Nov. 14-20, 200628

Near-side width does not strongly depend on FF

2.5-3.5 GeV/c 3.5-4.5 GeV/c

4.5-5.5 GeV/c 5.5-6.5 GeV/c

Will be included in the systematic uncertainties in the future.