2005 05 topwg bernardo

W and top polarization with Rome TopReX ttbar samples: 

W and top polarization with Rome TopReX ttbar samples Goal: redo ATLfast analyses with semileptonic ttbar events on: W polarization (note ATL-COM-PHYS-2005-015) ttbar spin correlation (note ATL-PHYS-PUB-2005-001) using Rome full simulation F. Hubaut, E. Monnier, P. Pralavorio, B. Resende (CPPM)

Samples: 

Samples Rome samples : 004520 and 004521 200k semileptonic ttbar events (l = e, mu) Parameters Mtop=175 GeV CTEQ5L ISR/FSR on Multiple Interaction = DC2 Q-scale = default Toprex (sqrt(Mtop**2+PTttop**2) Generation : 9.0.4 TopReX 4.09 and Pythia 6.226 Simulation, digitization : 9.0.4 Initial detector geometry Reconstruction : 10.0.1 ~190k done without pile-up Use CBNT

Lepton reconstruction (1): 

Lepton reconstruction (1) Electrons IsEM=0 (TRT mask) Efficiency ~75% Muons MuonBoy reconstruction Efficiency ~91% Selection Only 1 lepton with Pt>20 GeV, eta<2.5 Isolation criteria: DR (l, jets)>0.2 Exclude jet corresponding to electron

Lepton reconstruction(2): 

Lepton reconstruction(2) Matter effects, brem. Electron PT Electron η Muon PT Muon η

Jet/PTmiss reconstruction(1): 

Jet/PTmiss reconstruction(1) Jet Cone algorithm, size DR=0.7 No jet recalibration for top events B-tagging: SV2 weight>3. Efficiency~60%, rejections light/c~80/8 Etmiss Ex/yMissFinal : includes muon contributions Selection At least 4 jets with Pt>30 GeV, eta<2.5 At least 2 b-tagged Ptmiss>20 GeV

Jet/PTmiss reconstruction(2): 

Jet/PTmiss reconstruction(2) Missing PT Number of jets Jets PT Jets η Calibration and cone size are different (0.7 Rome/ 0.4 Atlfast)

Selection efficiency (%): 

Selection efficiency (%) Number of events after all cuts (0.8fb-1): ~11500 Main differences arise from reconstructed electron and jet spectra εl=90%

Event reconstruction(1): 

Event reconstruction(1) Select 2 non-b jets with Mjj closest to MW Select the b jet with Mjjb closest to Mt PTmissPTν ; Pzν by constraining Mlν to Mw Select the b jet closest to the lepton Select the n with Mlνb closest to Mt Quality cuts : D(MWhad) < 20 GeV D(Mtop) < 35 GeV

Event reconstruction(2): 

Event reconstruction(2) 80.4 GeV σ=5.3 GeV 80.6 ± 0.9 GeV σ=7.6 ± 1.6 GeV 174.1 GeV σ=8.9 GeV 176.6 ± 0.8 GeV σ=12.3 ± 1.2 GeV 173.1 GeV σ=7.9 GeV 173.0 ± 0.6 GeV σ=6.4 ± 0.9 GeV

W polarization: 

W polarization is measured through angular distribution of charged lepton: W polarization Longitudinal W+ Fraction F0 Left-handed W+ Fraction FL Right-handed W+ Fraction FR

Measurement method (1): 

Measurement method (1) Reconstruction and cuts parton level Selection cuts and reconstruction distort the parton level distribution: Differences Full Sim. / Atlfast < 20% Analysis limited by Rome statistics Correction function

Measurement method (2): 

Measurement method (2) F0=0.699 ± 0.005 FL=0.299 ± 0.003 FR=0.002 ± 0.003 Fast sim (Atlfast) 10 fb-1 F0=0.691 ± 0.028 FL=0.304 ± 0.014 FR=0.005 ± 0.014 TopReX, selection+ recons. + correction Fit results with 2 parameters (constraint F0+FL+FR=1) Full sim (Rome) 0.8 fb-1

Conclusions / Perspectives: 

Conclusions / Perspectives Rome Toprex events reconstructed W polarization analysis performed To do for Rome Perform ttbar spin correlation analysis Improve jets: recalibration, cone size Pile-up (if available?) MC@NLO sample? Further studies (more stat.?) Cone vs Kt Trigger/selection effects