1. Pavel Nevski ATLAS detector performance in Heavy Ion Collisions at LHC ATLAS detector performance in Heavy Ion Collisions at LHC Pavel Nevski BNL Motivations Motivations Event Characteristics Subsystems Performance

2. Pavel Nevski Heavy Ions at the LHC  Initial energy density about 5 times higher than at RHIC:  Lifetime of a quark-gluon plasma much longer : 10-15 fm/c at LHC as compared to 1.5-4 fm/c at RHIC  Access to truly hard probes with sufficiently high rates : p T > 100 GeV/c (at RHIC p T  20 GeV/c) copious production of b and c quarks  deconfinement  restoration of the chiral symmetry,  physics of parton densities close to saturation Study of QCD matter at extremely high energy densities and ~vanishing baryon chemical potential: RHIC LHC 200 5500 GeV

3. Pavel Nevski ATLAS as a Heavy Ion Detector 1.High Resolution E.M. and Hadronic Calorimeters —Hermetic coverage up to |  | < 4.9 —Fine granularity (with longitudinal segmentation) 2.Large Acceptance Muon Spectrometer —Coverage up to |  | < 2.7 3.Si Tracker —Large coverage up to |  | < 2.5 —Finely segmented pixel and strip detectors —Good momentum resolution High p T probes Muons from , J/ , Z 0 decays Tracking particles with p T  1.0 GeV/c 2.+ 3. Heavy quarks(b), quarkonium suppression( ,  ’) 1.& 3.Global event characterization

4. Pavel Nevski ATLAS Detector ATLAS is an excellent detector for high p T physics and jet studies

5. Pavel Nevski Simulation DataFlow

6. Pavel Nevski Simulation Tools: Generators HIJING Event Generator: Based on PYTHIA and Lund fragmentation scheme (Soft string dynamics + hard pQCD interactions) with nuclear effects: nuclear shadowing, jet quenching However, HIJING jet quenching model does not fit the RHIC measurements quenching, no shadowing quenching, shadowing no quenching, shadowing Pb+Pb b=0 fm  s NN =5.5 TeV

7. Pavel Nevski Stable Particles after HIJING Per 10 Events All decays faster then pi0 are now done by hijing, But you can switch some of them off

8. Pavel Nevski Central Pb+Pb Collision in ATLAS  About 75,000 stable particles  ~ 40,000 particles in |  |  3.2  CPU – 6 h per central event (800MHz)  Event size 50MB (without TRT) N ch (|y|  0.5)

9. Pavel Nevski Simulated Event Samples HIJING + full GEANT3 ATLAS detector simulations Only particles within |y| < 3.2 for the moment  High Geant thresholds 1 MeV tracking/10 MeV production — 5,000 events in each of 5 impact parameter bins: b = 0-1, 1-3, 3-6, 6-10, 10-15 fm  Standard ATLAS thresholds 100 keV tracking/1 MeV production — 1,000 central events, b = 0-1fm  Initial layout – 2 pixel barrel layers — 1,000 central events, b = 0-1fm

10. Pavel Nevski Global Measurements Day-one measurements: N ch, dN ch /d ,  E T, dE T /d , b  Constrain model prediction  Indispensable for all physics analyses Predictions for Pb+Pb central collisions at LHC ( dN ch /d  )  0 Model/data ~12500 HIJING:with quenching, no shadowing ~ 6500 HIJING:with quenching, with shadowing ~ 3200 HIJING:no quenching, no shadowing ~ 2300 Saturation Model (Kharzeev & Nardi) ~ 1500 Extrapolation from lower energy data

11. Pavel Nevski Measurements of N ch (|  | < 3) Based on the correlation between measurable quantity Q and the true number of charged primary particles: Q = f(N ch ) Q: N sig - all Si detectors,except PixB  E tot EM,  E tot HAD  E T EM,  E T HAD Caution: Consistency between the measured signals and the simulated ones Monte Carlo dependency

12. Pavel Nevski Estimate of the Collision Centrality Monotonic relation between measurable quantities Q and centrality parameter b (Npart,Ncoll) allows for assigning to a certain fraction of events, selected by cuts on Q, a well defined average impact parameter. Correlation improves with a larger rapidity coverage. N sig E T - EME T - HAD

13. Pavel Nevski Event Reconstruction n Most of the standard ATLAS reconstruction packages developed for PP physics are working on HI events after minimal parameter tuning: –We have successfully exercised all calorimeter reconstruction - photons, jets, missing energy. –Silicon Pixel and Strip detectors have reasonable occupancy and can provide track reconstruction already with existing PP codes. –Muon reconstruction is even simpler in HI events - provided the muon energy is above 6 GeV - provided the muon energy is above 6 GeV n Dedicated HI reconstruction packages will be developed in due time: –Jet reconstruction is a tricky issue -work is ongoing to develop an appropriate code

14. Pavel Nevski Inner Detector Occupancy Pixel DetectorSilicon Tracker Impact parameter b=0-1 fm, HIJING event generator. TRT is excluded from analysis

15. Pavel Nevski Track Reconstruction Track reconstruction performed with ATLAS pp tracking code using the Pixel and SCT detectors (xKalman++). —p T threshold for reconstructed tracks is set to 1 GeV. —Tracking cuts are optimized to get a decent efficiency and low rate of fake tracks. — Further high p T fake rejection can be achieved using calorimeter For p T 1 to 15 GeV/c: efficiency ~ 70 % fake rate ~5 % Much better in |y|<1 May very with cuts: Eff. ~80%, fake rate 15-20% Eff. ~65%, fake rate ~2% - Subject to further optimization

16. Pavel Nevski Track Reconstruction Momentum resolution Efficiency versus rapidity Flat dependency for |y| < 2~3% for p T up to 20 GeV/c ~2% for |y|<1

17. Pavel NevskiCalorimetry Energy Per Cell: n 0.10 x 0.10 cell in e.m. calorimeter n 0.10 x 0.10 cell in hadron calorimeter

18. Pavel Nevski Jets and Clusters n Reconstructed e.m. clusters – exotic processes can be observed with cluster energy more than ~15 GeV (?) n Reconstructed hadronic jets – jet signature can be used with Pt above 50 GeV (?)

19. Pavel Nevski Modified Jet Reconstruction - Pythia jets embedded in Hijing events - Local energy level is evaluated and subtracted - Reconstructed Jet parameters are compared to MC truth for embedded jets

20. Pavel Nevski Heavy Quark Production Heavy quarks live through the thermalization of QGP  can be affected by the presence of QGP Their radiative energy loss is different than for light quarks. Preliminary study: —Standard ATLAS algorithm for pp —Higgs events embedded into pp or Pb-Pb event —Cuts on the vertex impact parameter in the Pixel and SCT Promising, should be improved when combined with muon tagging! Rejection factors against light quarks versus b-tagging efficiency p-p Pb-Pb

21. Pavel NevskiMuons n On average muons loose 5 GeV in calorimeter and have strong multiple scattering angles - Use combined info from ID+muon spectrometer to increase accuracy - Use combined info from ID+muon spectrometer to increase accuracy n Association based on geometrical cuts:  φ x  η after back extrapolation at vertex  φ x  η after back extrapolation at vertex + global fit of all possible combinations, ordered in decreasing χ2 + χ2 cut + global fit of all possible combinations, ordered in decreasing χ2 + χ2 cut n Loose cuts at the beginning: 96.2% of μ from  are kept 96.2% of μ from  are kept n Compare 2 samples: pure  and Hijing events: 5000   μ+μ- generated with T=240 MeV 5000   μ+μ- generated with T=240 MeV 5000 Pb-Pb Hijing events with b=0 5000 Pb-Pb Hijing events with b=0 ( after full Geant 3 + reconstruction) ( after full Geant 3 + reconstruction) n n Invariant mass is calculated using the overall fit

22. Pavel Nevski Quarkonium Suppression Upsilon family  (1s)  (2s)  (3s) Binding energies (GeV) 1.1 0.54 0.2 Dissociation at the temperature ~2.5T c ~0.9T c ~0.7T c  μ+ μ- Signal – 5000 generated  μ+ μ- 0.18 μ, 0.008 μ+ μ- pairs reconstructed per event Hijing(b=0): 0.18 μ, 0.008 μ+ μ- pairs reconstructed per event 0.0009 expected μ+ μ-/ev, 387,640 mixed pairs For min.bias -> 0.0009 expected μ+ μ-/ev, 387,640 mixed pairs 8b/410μb  >  8b/410μb  >-> 18 background pairs per one   +  – Background estimate (HIJING+G3)  S/B ~ 0.6:

23. Pavel Nevski Trigger DAQ For Pb+Pb collisions the interaction rate is 8kHz, a factor of 10 smaller than LVL1 bandwidth. We expect further reduction to 1kHz by requiring central collisions and pre-scaled minimum bias events (or high p T jets or muons). The event size for a central collision is ~ 5 Mbytes. Similar bandwidth to storage as pp at design L implies that we can afford ~ 50 Hz data recording.

24. Pavel NevskiConclusion n n ATLAS detector will be capable of measuring many aspects of High pT Heavy Ion physics n n Simulation, Reconstruction and Analysis tools exist to evaluate the detector performance n n Work is in progress to understand the detector performance for studying the truly high pT phenomena