Quarkonium studies in p+p andPb+Pb collisions with CMS

– Torsten Dahms (on behalf of CMS) –

CERN

ReteQuarkonii Thematic Day, IPN Orsay

February 9th, 2010

Quarkonia production at the LHCThe LHC will provide p+p and Pb+Pb collisions at high energies and luminosities (“latest plan” according to Chamonix 2010 meeting):

•In November/December 2009 proton beams were collided at√s = 900 GeV (2.36 TeV) during which CMS recorded an integrated luminosity of about 10 bμ −1 (400 mb−1)

•Starting February/March 2010 with p+p collisions at √s = 7 TeV with peak luminosities up to 1032 cm−2s−1 until 1 fb−1 of integrated luminosity has been collected (18 − 24 months)

•Corresponding Pb+Pb collision energy will be √sNN = 2.8 TeV

(beam time in Fall 2010)

•Running at higher energies up to 14 TeV only after a longer shutdown(≥ 2013?)

•The Pb+Pb runs will then occur at 5.5 TeV per NN collision, with4×1026 cm−2s−1 Pb+Pb instantaneous luminosity

Quarkonium states will be produced at very high rates2

A transverse slice of the CMS barrel

3

Si TrackerSilicon micro-stripsand pixels

CalorimetersECAL PbWO4

HCAL Plastic Sci/Steel sandwich

Muon BarrelDrift Tube Chambers (DT)Resistive Plate Chambers (RPC)

CMS phase-space coverage

• CMS: full and almost full acceptance at the LHCφ η• charged tracks and muons: | | < 2.4η• electrons and photons: | | < 3η• jets, energy flow: | | < 6.7 (plus | | > 8.3 for neutrals, with the ZDC)η η

4

• excellent granularityand resolution• very powerful

High-Level-Trigger

HF HF

CAST

OR

CAST

OR

ZDC

ZDC

J/ψ detection and dimuon mass resolutionin CMS

5

• CMS is ideal to measure quarkoniain the dimuon decay channel:– large rapidity coverage (| |<2.4)η– excellent dimuon mass resolution

• Good muon momentum resolution:– matching between the tracks in the muon

chambers and in the silicon tracker– strong solenoidal magnetic field (3.8 T)

• Because of the increasing material thickness traversed by the muons, the dimuon mass resolution changes with pseudo-rapidity, from ~15 MeV at ~0 to ~40 MeV at ~2.2

Inclusive differential J/ψcross sections in p+p collisions

• The observed J/ yield results from:ψ– direct production– decays from ’ and ψ χc states– decays from B hadrons (non-prompt)

• CMS will measure the inclusive, prompt, and non-prompt productioncross sections

• CMS should collect several 104 J/ eventsin a matter of days at 1031 cm−2s−1 (∫L dt = 1 pb−1 at 14 TeV)

• The J/ yield is extracted by fitting the dimuon mass distribution, ψseparating the signal peak from the underlying background continuum

6

prompt

p+p √s = 14 TeV(3 pb−1)

CMS simulation

CMS PAS BPH-07-002

Inclusive differential J/ψ cross sections

7

A : convolution between the detector acceptanceand the trigger and reconstruction efficiencies, which depend on the assumed polarization

corr : needed if MC description of trigger and offline efficiencies does not match “reality”

Competitive with Tevatron results after only 3 pb−1

A

CMS simulation

CMS PAS BPH-07-002 p+p at 14 TeV3 pb−1 ~ 75 000 J/ψ

€

dσ

dpTJ /ψ

⋅Br(J /ψ → μ +μ−) =NJ /ψfit

Ldt∫ ⋅ΔpTJ /ψ ⋅A ⋅λcorr

CMS simulation

CMS PAS BPH-07-002

Feed-down from B meson decays

8

An unbinned maximum likelihood fit is made, in pT bins, to determine the non-prompt fraction, fB, using the dimuon mass and the pseudo proper decay length

Systematic error dominated byluminosity and polarization uncertainties

B fraction fit

J/ pψ T : 9–10 GeV/c

CMS simulation

CMS PAS BPH-07-002

CMS simulation

CMS PAS BPH-07-002

€

l xy =Lxy

J/ ψM J/ ψ

PTJ/ ψ

Quarkonium production in Pb+Pb collisions

9

μΥ μ μμ−

dNch/d = η3500

CMS simulation

ϒ→μ+μ−: acceptance and mass resolution

• CMS has a very good acceptance for dimuons in the Upsilon mass region• The dimuon mass resolution allows us

to separate the three Upsilon states:~ 54 MeV within the barrel and~ 86 MeV when including the endcaps• 1 month Pb+Pb at 5.5 TeV with average

luminosity of 4×1026 cm−2s−1 0.5 nb−1

10

CMS simulation

CMS simulation

Pb+Pb √sNN = 5.5 TeV(0.5 nb−1)

σ = 54 MeV/c2

Barrel: both muons in || < 0.8

Barrel + endcaps: muons in || < 2.4

pT (GeV/c)

Acce

ptan

ce

’

’’

CMS simulation

CMS PTDR Addendum 1

J/ψ→μ+μ−: acceptance and mass resolution• The material between the silicon tracker and the muon chambers (ECAL, HCAL,

magnet’s iron) prevents hadrons from giving a muon tag but impose a minimum muon momentum of 3.5–4.0 GeV/c:– No acceptance problem for due to high massΥ

– but for J/ ’s this sets a relatively high threshold on the pψ T

• The low pT J/ acceptance is better at forward rapidityψ

11

barrel +endcaps

barrel

pT (GeV/c)

J/ψ

Acce

ptan

ce

p T (G

eV/c

)

barrel + endcaps(| |<2.4)ησJ/ψ = 35 MeV/c2

Pb+Pb √sNN = 5.5 TeV(0.5 nb−1)

CMS simulation

CMS PTDR Addendum 1

The High Level Trigger

• CMS High Level Trigger:12 000 CPUs of 1.8 GHz ~ 50 Tflops!

• Executes “offline-like” algorithms

• p+p design luminosity L1 trigger rate: 100kHz• Pb+Pb collision rate: 3 kHz (peak = 8 kHz)

p+p L1 trigger rate > Pb+Pb collision rate run HLT codes on all Pb+Pb events

• Pb+Pb event size: ~2.5 MB (up to ~9 MB)• Data storage bandwidth: 225 MB/s

10–100 Pb+Pb events / second• HLT reduction factor: 3000 Hz → 100 Hz• Average HLT time budget per event: ~4 s

• Using the HLT, the event samples of hard processes are statistically enhanced by considerable factors

12

ET reach ×2

×35

×35

jets

Pb+Pb at 5.5 TeVdesign luminosity

CMS PTDR Addendum 1

pT reach of quarkonia measurements

13

● produced in 0.5 nb−1

■ rec. if dN/d ~ 2500○ rec. if dN/d ~ 5000

J/ψ

Υ

0.5 nb−1 : 1 month at 4×1026 cm−2s−1

Expected rec. quarkonia yields:J/ : ~ 180 000 : ~ 26 000ψ Υ

Statistical accuracy (with HLT) of’ / ratio vs. pΥ Υ T should be good

enough to rule out some models

Pb+Pb √sNN = 5.5 TeV

CMS simulation

CMS PTDR Addendum 1

CMS simulation

CMS PTDR Addendum 1

CMS simulation

CMS PTDR Addendum 1

ϒ production in ultra-peripheralPb+Pb collisions

• CMS will also measure Upsilon photo-production, occurring in collisions with impact parameters larger than the Pb nuclear radii

• This will allow us to study the gluon distribution function in the Pb nucleus

• Around 500 events are expected after 0.5 nb−1, adding the e+e− and μ+μ− decay channels

14

using neutron tagging in the ZDCs

CMS simulation

CMS PTDR Addendum 1

CMS simulation

CMS PTDR Addendum 1

Summary (of the expectations)

• CMS has a high granularity silicon tracker, a state-of-the-art ECAL, large muon stations, powerful DAQ and HLT systems, etc. excellent capabilities to study quarkonium production, in p+p and Pb+Pb• Dimuon mass resolutions:

~30 MeV for the J/ ; ~90 MeV for the , over ||<2.4ψ Υ Good S/B and separation of (1S), (2S) and (3S)Υ Υ Υ• Expected p+p rates are high enough to collect J/ and dimuons up toψ Υ

pT ~ 40 GeV/c in a few days at 7 TeV

• Expected Pb+Pb rates at √sNN = 5.5 TeV:180 000 J/ and 26 000 (1S) per 0.5 nbψ Υ −1 (one month) Studies of Upsilon suppression as signal of QGP formation

• J/ and polarization, and ψ Υ χc → J/ + studies require larger samplesψ γ

15

A first look at “real” data

16

Start of the LHC: First CollisionsMonday 23rd November

CMS Experiment at the LHC, CERNDate Recorded: 2009-11-23 19:21 CETRun/Event: 122314/1514552 Candidate Collision Event

Events recorded: All CMS ON900GeV: ~400k2.36 TeV: ~20k

17

First Di-photon Distribution in CMS

• Data and MC comparison (uncorrected distributions)• Almost identical S/B, mass and width

compatible• M(π0) is low in both data and MC - Mostly

due to the readout threshold (100 MeV/Crystal) and conversions

Using “out of the box” corrections

18

Data: N( )/N(η π0) = 0.020 ± 0.003

MC: N( )/N(η π0) = 0.021 ± 0.003

Data: N( )/N(η π0) = 0.020 ± 0.003

MC: N( )/N(η π0) = 0.021 ± 0.003

Eta and Phi

ηη CMS 2009 Preliminary Uncorrected

φφ CMS 2009 Preliminary

19

Muons: A Dimuon Event at 2.36 TeV

20

pT(μ1) = 3.6 GeV, pT(μ2) = 2.6 GeV,m(μμ)= 3.03 GeV

pT(μ1) = 3.6 GeV, pT(μ2) = 2.6 GeV,m(μμ)= 3.03 GeV

• Expected one J/ → ψ μ+μ− event in 500k min. bias events at √s = 2.36 TeV• Got one J/ → ψ μ+μ− candidate in 20k events

• S/B ratio: 16/1 in [3.0,3.2] GeV/c2 region (background: ~ 0)

Dimuon event at 2.36 TeV

21

Backup

Some numbers

Bμμ × σPbPb ( b)μ

J/ψ ’ψ Υ ’Υ ”Υ

48 900 880 300 80 44

dNch/d |η =0η , Δη

S/B N(J/ )ψ S/B N( )Υ N( ’)Υ N( ”Υ)

2500, | |< 2.4η 1.2 184 000 0.12 26 000 7 300 4 400

2500, | |< 0.8η 4.5 11 600 0.97 6 400 2 00 1 200

5000, | |< 2.4η 0.6 146 000 0.07 20 300 5 900 3 500

5000, | |< 0.8η 2.8 12 600 0.52 6 000 1 800 1 100

23

Why to study Quarkonia at the LHC?

In Pb+Pb collisions:• Debye screening in deconfined phase leads

to melting of quarkonia when screening length exceeds binding radius

• Binding energy depends on quarkonium state and feed down from higher states lead to sequential suppression of J/ and ψ

with increasing temperatureΥ• It is important to measure quarkonium

yields in Pb+Pb collisions as function of pT and collision centrality

In p+p collisions:• Base line for heavy ion collisions• Cross section measurements• Polarization

24

ψ’

c

Hard Probes at the LHC

• Experimentally & theoretically controlled probes of the early phase in the collision

• Very large cross sections at the LHC

• CMS is ideally suited to measure them

• Pb+Pb instant. luminosity: 1027 cm-2s-1

• ∫ L dt = 0.5 nb-1 (1 month, 50% run eff.)

• Hard cross sections: Pb+Pb = A2 x p+p

p+p equivalent ∫ L dt = 20 pb-1

1 event limit at 0.05 pb (p+p equiv.)

25

Impact of the HLT on the pT reach of RAA

Nuclear modification factor = AA-yield / pp-yield = “QCD medium” / “QCD vacuum”

26

Pb-Pb (PYQUEN) 0.5 nb-1

HLT

Important measurement to compare with parton energy loss models and derivethe initial parton density and the medium transport coefficient

Jet ET reach and fragmentation functions

Jet spectra up to ET ~ 500 GeV (Pb+Pb, 0.5 nb-1, HLT-triggered)

Detailed studies of medium-modified (quenched) jet fragmentation functions

27

min. bias

HLT Gluon radiation:large angle (out-of-cone) vs. small angle emission

γ, γ*, and Z tagging of jet production

28

Unique possibility to calibrate jet energy loss (and FF) with back-to-back gauge bosons (large cross sections and excellent detection capabilities).

Heavy quark dimuon (dominant) background can be rejected by a secondary vertex cut.Resolutions: 50 mm in radius and 20 mm in φ

Z0+jet

dimuon trigger

associated hadrons

g *away side