US participation in Heavy Ion Physics with Compact Muon Solenoid at LHC M. Ballintijn,K Barish,R Betts, BE Bonner, W.Busza, D.Cebra, G.Eppley, E.Garcia,

US participation in Heavy Ion Physics with Compact Muon Solenoid at LHC

M. Ballintijn,K Barish,R Betts, BE Bonner, W.Busza, D.Cebra, G.Eppley, E.Garcia, F.Geurts, C.Halliwell, D.Hofman, P.Kulinich, W.Llope, M.Murray, G. van Nieuwenhuizen, E.Norbeck, R.Nouicer, Y.Onel, C.Roland, G.Roland,

R.Seto, G.S.F.Stephans, B. Wyslouch, P.Yepes

MIT, Rice, TAMU, UC Davis, UC Riverside, UI Chicago, U Iowa

CMS HI workshopwas held at MITFebruary 2002

US Heavy Ion physics today

• RHIC has completed first long run of Au-Au at (sNN) =200 GeV/c (Its highest energy)– Expect to run for many years

• It has reached design luminosity– Expect improvements

• The four detectors: BRAHMS, PHENIX, PHOBOS and STAR– BRAHMS and PHOBOS have short lifetime (<4 y)

• Much better detectors and running conditions than ever before in the field of relativistic heavy ions

• Practically no involvement at LHC

Main RHIC physics and tools

• Equation of state of QCD: connection to lattice results predicting phase transitions

• Chiral symmetry restoration• Quark Gluon Plasma, How does a high-temperature

quark-gluon state emerge from the collision of nuclei?• Photon and di-lepton probes ( J/), studying effects of

the plasma on particle production and survival• Production of high pt particles is affected by the energy

loss in nuclear matter

Early results are accumulating. No large, obvious effects seen, but some puzzles start to appear

Integrated Au-Au luminosity

FY2000(66 GeV/amu)

FY2001 – 02100 GeV/amu

PHENIX during last 10 days:24 (b)-1/week

Lave(week) = 0.4 1026 cm-2 s-1

Lave(week)/Lave(store) = 27 %

Multiplicity per nucleon-nucleon interaction as a function of collision energy for heavy ions and for proton-antiproton collisions

Preliminary

Comparison of momentum spectra of particles AuAu vs ppToo few high pt particles ?

ppN

AuAuR

binaryAA

STAR

Production of high pT particles, e.g. 0

Heavy Ion physics at the LHCIncrease energy (sNN) 200->5500 GeV

• Saturated gluon distributions in colliding nuclei, initial state of the system calculable in perturbative QCD

• Plasma has higher temperature and lives longer in partonic state

• Spectrum of quarkonia produced with high statistics

• Jets clearly visible and identifiable

• Z0 produced in large numbers

HI group in CMS has produced many excellent physics studiesResults need some update and extensions to take into account RHIC results

The experiments

ATLAS: pp experiment, recent HI interest

ALICE: dedicated HI experiment

CMS: pp experiment with HI program

Design of LHC experiments is affected by the multiplicity of charged particles

Peter Steinberg, 2002

LHC?

Extrapolated to LHC:dN/d~1400

LHC detectors are being designed for

dN/dy~8000

Quarkonia, CMS is very good here

J/ family

’/ ratioaffected by the initial conditions: a plasma thermometer(Ramona Vogt)

Detailed studies using full simulation,reconstruction, background subtractiondN/dy studied from 2500 to 8000

Very large event rate

Uses muon detector, outer tracker, pixels

Pb+Pb Sn+Sn Kr+Kr Ar+ArJ/psi 28.7 210 470 2200psi' 0.8 5.5 12 57Upsilon 22.6 150 320 1400Upsilon' 12.4 80 180 770Upsilon'' 7 45 100 440

Yield/month (kevents, 50% eff)

Can we use tracker ? CMS Tracker Occupancy

• Calculated for PbPb dNch/dy=5000

• For reference: STAR TPC occupancy reaches 22%

Jet fragmentation

• Find jets using calorimetry• Study charged particle momenta inside of a jet using the tracker• For this study use 4-5 outer layers of the tracker (use

conservative resolution obtained in pp studies: AA plausible with low occupancy in outer layers)

Particles in jet

Background

Field OFFField OFF Field ONField ON

Field OFFField OFFField ONField ON

Global observables e.g. Et

Our proposal to DoE/Nuclear

• Extend the physics reach of the US heavy ion community beyond RHIC’s energy scale

• Concentrate US physics effort on the study of phenomena most likely to be affected by the energy increase, the “hard probes”:– Quarkonia and heavy quark production– Jet production, jet-jet, jet-gamma and jet-Z0

correlations

• Provide US/RHIC expertise and tools to study high pt processes in AA collisions at the LHC

• Use detector designed for high pt physics: CMS

• LHC starts in mid-2007 with pp, AA to follow in 2008(?): Be ready with strong group in 2008

Our proposal to DoE/NuclearSpecific plans of the existing groups

• Physics studies, software development• High Level Trigger code development +

request to fund 2/8 slices of Event Filter Farm (Rice, MIT, UC Davis, UCR)

• Zero degree calorimeter (U Iowa, UIC, TAMU)

• Total ~ 5 M$ from DoE/Nuclear• review took place 2&3 of April @ DoE, no final

report yet. Closeout conclusion: “continue studies, come back later”

• competing with ALICE and ATLAS(!)

Centrality: Participants vs. Spectators

“Spectators”

Zero-degreeCalorimeter

“Spectators”

Many things scale with Npart:• Transverse Energy• Particle Multiplicity• Particle Spectra

“Participants”

Only ZDCs measure Npart

specpart NAN

Detectors at 90o

The collision geometry (i.e. the impact parameter) determines the number of nucleons that participate in the collision

Beam pipe splits 140m from IR.

INNER WALL

Z-AXIS

ZDC LOCATION

BEAM

BEAMPAIR OF PANTS(POP)

ALIGNMENT PINS

Zero Degree Calorimetry for CMS

RHIC ZDCs work very well

High Level Trigger (HLT)

• All event data available:– Fine data for

Calorimetry and Muon Detectors

– Tracker

• Refine triggered object

• Allows to go lower in pT

• Processing time O(s)• Filtering Farms of

commodity processors (Linux)

• L1 in AA has larger backgrounds than in pp due to underlying event.

• Efficiency trigger requires more careful analysis. HLT can do a better job than L1.

• HLT to play a greater role in AA

AA Event Size & Data Flow

0

500

1000

1500

2000

2500

3000

pp OO ArAr SnSn PbPb

Eve

nt S

ize

(Kby

tes)

Pixel Si TrackerECAL HCALMuon

Event Size Detector # Channels(1000)

Pixel 45,000

Si Tracker 12,000

ECAL 230HCAL 14

Muon Det. 400

Total 57,644

Data Flow and Rates

L1

HLT

HLT better

trigger job

CPU Estimate of HI Online Tracking

June 16, 2000 First STAR event

tracked on line

STAR CMS ScalingdN/dy 600 3000 5.0Acceptance, ||< 1.5 2.5 1.7Max. Number Layers 45 13 0.3Clustering 1 2 2.0Clustering Type 1 2 2.0All Factors 9.6

STAR CMSMIP s/event 386 3706MIP s/ CPU 460 13800Event/CPU/s 1.2 3.7Max Input L3/HLT (Hz) 100 7400Minimum Number CPUs 84 1987

Moore’s Law: double/18 months

for ~8 years

• Use STAR (Rice) experience

• Scale linearly with # clusters from STAR AuAu to CMS PbPb

• Assume Moore’s Law

Near term plan

• Work together with heavy ion group in CMS on expanding the physics scope of the heavy ion program

• Develop good physics case for tracking in CMS, at HLT level and in offline analysis

• Develop and possibly prototype Zero degree calorimeter.

• Fight for funding from DoE, expand collaboration

Summary

• CMS has a potential to be an excellent detector for heavy ion physics

• RHIC results will really determine the scope of interesting physics, the US community will have direct access to the RHIC experience

• HI physicists could provide useful expertise, manpower and source of funding

Z0 production

• Z0- can be reconstructed with high efficiency

• A probe to study nuclear shadowing and parton energy loss

• Z0 also proposed as reference to production.– Nuclear effects may

depend on mass MZ>M

– Different production mechanisms:

• Z0: antiquark-quark, quark-gluon and antiquark-gluon.

: gluon-gluon.

Open b: high mass +- : medium modification, energy loss for b-quarks

Muon tracking (muon+tracker)

Displaced vertices(pixels)