Christof Roland MIT

1

Heavy Ion Physics with the CMS Experiment at the
LHC
Christof Roland / MIT For the CMS Heavy Ion
Group Rencontres de Moriond 2004 La Thuile
2
CMS as a Detector for Heavy Ion Physics

• DAQ and Trigger
• High rate capability for AA, pA, pp
• High-Level Trigger capable of inspecting/selecting
  HI events in real-time

• Silicon Tracker
• Good efficiency and low fake rate
• for pTgt1 GeV
• Excellent momentum resolution
• Dp/p1

• Fine Grained High Resolution Calorimeter
  (E-calH-cal)
• Hermetic coverage up to hlt5
• (hlt7 proposed using CASTOR)
• Zero Degree Calorimeter (proposed)

• Muon Reconstruction
• Tracking m from Z0, J/?, ?
• Wide rapidity range hlt2.4
• sm 50 MeV at ?

Fully functional at highest expected
multiplicities Detailed studies at
dN/dy3000-5000 and cross-checks at 7000-8000
3
Heavy Ion Physics at ?sNN 5.5TeV

• Large Cross section for Hard Probes
• Copious production of high pT particles
• Nuclear modification factors RAA at very high pT
• Large cross section for J/? and ? family
  production
• Different melting for members of ? family
• Large jet cross section
• Jets directly identifiable
• Study in medium modifications

4
Global Event Characterization

• Charged Particle Multiplicities
• Predictions vary by a factor of 4!
• dN/dy 2000 8000
• (RHIC extrapolation vs. HIJING)

5
Azimuthal Asymmetry

• Measure azimuthal anisotropy of
• Charged Particle Production (vs. pT)
• Energy flow in the Calorimeters

6
Energy dependence of high pT suppression

• Inclusive pT spectra vs. collision centrality
• Determine nuclear modification factors RAA
• Hard spectrum and high data rate will allow to
  perform this measurement very early in the LHC
  program and out to high pT

1000 Pb Pb Events _at_ LHC
I. Vitev and M. Gyulassy, Phys.Rev.Lett. 89
(2002)
7
Quarkonia

• Excellent mass resolution (sm 50 MeV )and high
  statistics
• for the J/y and U family
• full simulation of di-muon channel.
• Level 1 trigger simulations in the barrel

8
High Mass Dimuon, Z0 Production

• Z0-gtmm can be reconstructed with high efficiency
• A probe to study nuclear shadowing
• Z0 also proposed as reference for ? production.
• High statistics (1 month)

9
Jet Reconstruction in CMS using Calorimeters
100GeV Jet in a PbPb event (after background
subtraction)
10
Balancing g or Z0 vs. Jets Calibrated jet
energy !

• ETjet, ggt120 GeV in the barrel

1 month at 1027 cm-2s-1 PbPb
Direct Measurement of parton energy Loss! Trigger
capabilities large acceptance are essential
11
Jet Shapes and Fragmentation
Longitudinal momentum fraction z along the
thrust axis of a jet
pT relative to thrust axis

• Jet Reconstruction in Calorimeters plus
  high-precision tracking allow for detailed jet
  shape analysis to study the energy loss mechanism
• Use tagged b Jets to measure charged-particle
  fragmentation function and jet-shapes of heavy
  quark jets and compare properties of light quark
  jets.

12
Conclusions

• LHC will extend energy range and in particular
  high pT reach of heavy ion physics
• CMS is preparing to take advantage of its
  capabilities
• Excellent coverage and resolution
• Quarkonia
• Jets
• Centrality, Multiplicity, Energy Flow reaching
  very low pT
• Essentially no modification to the detector
  hardware
• New High-Level Trigger algorithms
• Zero Degree Calorimeter, CASTOR and TOTEM
  proposed to extend forward coverage
• Heavy Ion program is well integrated into overall
  CMS Physics Program
• The knowledge gained at RHIC will be extended to
  the new energy domain

13
To be continued
14
The Algorithm

• Adapted from default pp reconstruction.
• Based on Kalman Filter
• Modifications to the pp Algorithm
• Trajectory Seed Generation
• Three pixel hit combinations compatible with
  primary event vertex
• Trajectory Building
• Special error assignment to merged hits
• Trajectory cleaning
• Allow only one track per trajectory seed
• Trajectory Smoothing
• Final fit with split stereo layers
• Running and tested in ORCA_6_3_0 and ORCA_7_2_4

15
Occupancy

• Occupancy in central PbPb Event
• 1-3 in Pixel Layers
• Up to 70 in Strip Layers _at_ dNdy 7000

16
Pixel Triplet Seeds

• Pixel triplets provide precise initial estimate
  of track parameters.

See M.Konecki CPT Week 5/03

• Generate only seeds by consistent with primary
  event vertex (constraint dr 100mm dz 350mm)
• Minimal number of seeds is crucial for runtime
  performance and low number of fake tracks
• HitTriplets are copied to TrajectorySeeds
• gt CombinatorialSeedGeneratorFromPixelTriplets.cc

17
Trajectory Building

• Study pattern recognition with standard tools
  (e.g AnalysingTrajectoryBuilder)
• Number of candidates per layer drops much faster
  than occupancy.

18
Geometrical Acceptance

• Require the track to cross more than 8 (12 hits)
• detector layers and hits in three pixel layers.
• Geometrical acceptance 80
• Defines cutoff at low pT (1GeV)

19
Algorithmic Efficiency and Fake Rate
h lt 0.7

• Require more than 12 hits on Track and
• Fit Probability gt 0.01 to reject Fake Tracks
• High efficiency and low fake rate even at very
  high track density

20
Performance of the Track Reconstruction

• Match Reconstructed tracks to MC input on a hit
  by hit basis.
• (Event sample dn/dy 3000 one 100GeV
  Jet/Event)

h lt 0.7
dpT/pT lt 1
21
Tracking in a Jet Cone

• The increased local track density in a jet-cone
  leads to a decrease in reconstruction efficiency
  of 5-10
• Can be corrected for since jets will be
  reconstructed by the calorimetry

22
Background Subtraction Algorithm

• Thanks to I. Vardanyan, A. Oulianov, O. Kodolova

• Event-by-event background subtraction
• Calculate ltETTower(h)gt and DTower (h) for each h
  ring
• Recalculate all ETTower tower energies
• ETTower ETTower Etpile-up
• Etpile-up ltETTower(h)gt DTower (h)
• Negative tower energies are replaced by zero

• Find Jets with ETjet gt Etcut using standard
  iterative cone algorithm using new tower energies
• Recalculate pile-up energy with towers outside of
  the jet cone
• Recalculate tower energy with new pile up energy
• Final jets are found with the same iterative
• cone algorithm ETJet ETcone Etpile-up new

23
Transverse Energy Flow
central Pb-Pb background dNch/dy 5000
(HIJING)
Transverse Tower energy
Dispersion of Tower energy
ECAL HCAL
BARREL ltETTowergt 1.7- 0.9 GeV ENDCAP
ltETTowergt 4.8- 5.0 GeV
Most of the energy is deposited in the ECAL
24
Jet reco. in pure HIJING Events
central Pb-Pb background dNch/dy 5000
(HIJING)
FAKE JETS
With pile-up subtraction
With pile-up subtraction
After background subtraction max. ET of fake
jets is less than 30 GeV
25
Jet Reconstruction

• QCD dijet events with initial parton energy of
  50-300 GeV
• Reconstruct
• without background
• with Pb-Pb background
• dNch/dy 5000
• Barrel and Endcap
• Only one jet with maximal jet energy was used for
  further analysis
• Threshold on reconstructed jet energy is 30 GeV
• Observe linear response of the reconstructed
  energy to the MC input Jet Energy

26
Jet Purity

• Find jet without background
• Find same jet with background and
• calculate the number of common cells
• Define the true reconstructed jet as a
• jet with more than 60 overlapping
• cells. This corresponds to dRmin lt 0.25

Find the fraction of true jets in all
reconstructed jets
27
Efficiency, Purity vs. Jet Energy
Reconstructing 50-300 GeV Jets in Pb-Pb background

• EFFICIENCY
• Number of events with true reco. Jets/Number of
  all generated events
• PURITY
• Number of events with true reco. QCD Jets/ Number
  of all reco. Jet events (truefake).
• Threshold of jet reco. ET gt30 GeV.
• Above 75(100) GeV we achieve
• 100 efficiency and purity
• in the barrel (endcap)

28
Jet Spatial Resolution
Dh
Df
Endcap h,f resolution of jets slightly better

• The h,f resolution is degraded in Pb Pb
  collisions
• Still better than h,f size of calorimeter tower
  (0.087x0.087)

29
Jet Reconstruction Summary

• The event by event pile-up subtraction method
  will allow for jet reconstruction in heavy ion
  events with high efficiency and purity
• Jet direction can be reconstructed with good
  accuracy. The resolution in h and f is smaller
  than the size of one calorimeter tower
  (0.087x0.087)