Outline: Executive summary ;-)

1/35

Outline:

• Executive summary ;-)

• A few topics related to charmonia production from recently obtained results (by CDF, HERA-B and NA50) more or less relevant to CMS

Towards quarkonia studies in CMS

Carlos Lourenço, LIP Workshop “Preparing for Physics in CMS”, Lisbon, Nov. 3, 2006

2/35

Executive summary

Some measurements of heavy quarkonia properties will be crucial to understand the production mechanisms (e.g., colour-singlet vs. colour-octet) and the validity of present theoretical approaches (e.g., factorisation in NRQCD)

Topics to be studied and prepared in advance of first data:• Production cross sections of the direct J/ (a first step in “B physics”),

of the ’ and of the c (using the decay c → J/ + )

• Polarisation measurements, as a function of pT

• Differential cross sections of the Upsilon family

It may also be possible to probe the gluon distribution function in the proton

The CMS experiment can (must) play an important role in exploring the properties of heavy quarkonia in pp collisions: an interesting physics topic in itself and important to control “backgrounds” or “baselines” of other physics topics

In the context of high-density and high-temperature QCD, heavy quarkonia studies are the best signature known up to now of the formation of a deconfined state of quarks and gluons, where the bound cc and bb states are “dissolved” above successive thresholds in the energy density of the QCD medium

3/35

Some numbers...

The LHC will be a heavy quarkonia factory: high statistics up to large pT

Design centre of mass energy: 14 TeV ; Design luminosity: 1034 cm-2s-1

The first years are well suited for dedicated studies on heavy quarkonia at the LHC, in view of affordable trigger rates, modest pile-up, event reconstruction, etc.The production rates for heavy flavours will be huge (mostly D and B mesons).

Expected total cross section: ~100 bProduction cross sections: charm: 7.8 mb; beauty: 0.5 mb; top: 0.8 nb

Production yields for an integrated luminosity of 1 fb-1 (1 week at 2 x 1033 cm-2s-1):• 7.8 x 1012 charm events• 5 x 1011 beauty events Nev = L . 1 fb-1 = 1015 b-1

• 8 x 105 top events

The number of events used for data analysis will be reduced by the acceptances and efficiencies...

From “Heavy Quarkonium Physics”, N. Brambilla et al. (QWG)CERN yellow report CERN-2005-005 (hep-ph/0412158)

4/35

Direct J/ production: from CDF to CMS

Singlet and octet contributions to direct J/ production at CDF and LHC.The extrapolation of the Tevatron fits to LHC energies seems rather insensitive to the details of the underlying theoretical description, and different approaches yield similar predictions as long as the appropriate NRQCD matrix elements are used.

J/ cross section times branching ratio into dimuons, in the “barrel acceptance”, for pT = 12 GeV/c: at CDF = 0.09 nb/GeV; at the LHC = 3 nb/GeV: 30 times larger

CDF LHC

From “Bottom Production”, P. Nason et al., hep-ph/0003142“1999 CERN Workshop on SM physics (and more) at the LHC”

5/35

Charmonia polarisation: NRQCD vs. CDF data

A crucial test of the NRQCD approach to charmonium production is the analysis of the J/ and ’ polarisation at large transverse momentum.At large pT NRQCD predicts transversely polarised J/ and ’. The polarisation

can be measured through the angular decay distribution: 1 + cos2 , where denotes the angle between the lepton three-momentum in the rest frame and the three-momentum in the lab frame. Pure transverse polarization implies = 1.

The angular distribution is predicted to change drastically as pT increases.

The CDF ’ data do not support this prediction, but the experimental errors are too large to draw final conclusions.

The analysis of the J/ measurements is more complicated because the data sample includes J/’s from decays of higher excited states.

From “Bottom Production”, P. Nason et al., hep-ph/0003142“1999 CERN Workshop on SM physics (and more) at the LHC”

Helicity frame

6/35

Quarkonia polarisation at the LHC

The absence of a substantial fraction of transverse polarization in production at large pT would be a serious problem for the application of the NRQCD factorization

approach to the charmonium system. To clarify this issue, a high-statistics measurement extending out to large pT will be necessary. Such a measurement

can only be carried out at the LHC.

The application of NRQCD should be on safer grounds for the bottomonium system where higher-order terms in the velocity expansion (in particular colour-octet contributions) are expected to be less relevant than in the case of charmonium.

If the charmonium mass is not large enough for a non-relativistic expansion to be reliable, the onset of transverse polarization at pT M≫ may become the

single most crucial test of the NRQCD factorization approach.

From “Bottom Production”, P. Nason et al., hep-ph/0003142“1999 CERN Workshop on SM physics (and more) at the LHC”

7/35

Some results from the HERA-B experiment

920 GeV p

s = 41.6 GeV

p-beam

z [cm]

12C

48Ti

184W

8/35

Three reference systems used to study the J/ polarisation

Beam frame: beam direction in the J/ rest frameHelicity (HCM) frame: J/ direction in the hadron (p-n) CM frameCollins-Soper (CS) frame: bisector between beam and (–)target directions in the J/ rest frame

z (beam frame)

z (Collins-Soper)

z (hadron CM; helicity frame)

tan = pT / MJ/ for pT = 0 the three reference systems are identical

direction of e+(or μ+) as seen in the J/ rest frame

θpolarisation axis

dN/d(cos) 1 + cos2

9/35

The J/ polarisation depends on the reference system

dNd(cosθ)

cosθ

1N

cosθ

BEAMBEAMBEAMBEAM

HCMHCMHCMHCM

CSCSCSCS

HERA-B

preliminarydN/d(cosθ) 1 + λθ cos2θdN/d(cosθ) 1 + λθ cos2θ

ee

|θ|

|θ|

|θ|

>

>

xF pT [GeV/c]

HCM HCM BEAM BEAM CSCS

HCM HCM BEAM BEAM CSCS

θ

ee ee

10/35

The hierarchy of frames: interpretation

pT [GeV/c]

θ

HCM HCM CSCS

HCM HCM CSCS

pT [GeV/c]

θ

The measurements can be reasonably well described if the polarisation is generated in the CS frame and translated into the HCM frame... ... but not the other way around.

The Collins-Soper frame is the one closest to the natural polarisation frame.

The choice of the reference frame appears to be crucial: the strong polarization seen in the CS frame is much attenuated in the HCM frame, with the BEAM frame in-between.

11/35

The J/ acquires its polarisation with respect to an axis which is, a priori, unknown. When seen from other reference frames, the angular distribution undergoes a sort of smearing, because the angle between the chosen polarisation axis and the “true” polarisation axis changes from event to event, depending on the momentum of the produced J/. Hence, in all other frames the angular distribution is more uniform, attenuating the extracted polarisation.

The measured polarisation is maximal in the CS frame, where the polarisation axis is parallel to the relative velocity of the interacting partons, and is almost completely washed out in the HCM frame, where the J/ momentum is assumed as the reference direction. This indicates that the polarization is a consequence of the production process rather than an intrinsic property of the J/ mesons.

The HERA-B data indicate a negative value longitudinal J/ polarisation.

And at low pT ...

The CDF collaboration should reanalyse the data using the CS reference frame.

The J/ polarisation: summary

12/35

’ over J/ production ratio in HERA-B

muon sample electron sample

'

'''

ψ

J/ψ

J/ψ

ψ

J/ψ

ψ

ε

ε

N

N

σ)llBR(J/ψ

σ)llBR(ψ

New value of ’ → branching ratio:0.75 ± 0.07 ± 0.03 %[PDG value: 0.73 ± 0.08 %]

13/35

CEM

NRQCD

’ / J/ versus xF and pT and feed-down

Which fraction of the J/ yield is due to feed-down from ’ decays?

7% of the J/ mesons observed by HERA-B are due to ’ decays

(7.0 ± 0.2 ± 0.4BRs) %

14/35

c production in HERA-B

From the 2000 data sample:370 ± 74 c’s ( + e+e-):

R(c) = 0.32 ± 0.06 ± 0.04[Phys. Lett. B 561 (2003) 61]

ll

Jc

c

cci

c

J

JJ

ici

N

NJBR

R /

//

2

1

)/(

entr

ies/

(10

MeV

/c2 )

m(μ+μ-γ)-m(μ+μ-) [GeV/c2]

preliminary2002/2003 data

(di-muon sample)

background:mixed events

afterbackgroundsubtraction

Fraction of the J/ yieldresulting from c decays:

2002/2003 data: 40 times larger χc statistics

c

15/35

Feed-down from the c

21 ± 5 % of the J/ mesons observed by HERA-B are due to c decays

Lower than the 30–40% values coming from earlier data (inc. 2000 HERA-B data)Based on 1300 c’s reconstructed in the dimuon channel; <10% of the full statistics!

Of the observed J/ mesons: 7% are from ’ decays, ~20% from c decays

more than 70% are directly produced

16/35

Studies of high-energy nucleus-nucleus collisions: why?

/T4 ~ number of degrees of freedom

hadronicmatter:

few d.o.f.

deconfinedQCD matter:many d.o.f.

QCD lattice calculations indicate that, above a critical temperature, Tc, or energy

density, c, strongly interacting matter undergoes a phase transition to a new state

where the quarks and gluons are no longer confined in hadrons

17/35

The QCD phase diagram

Tem

pera

ture

net baryon density

Ear

ly u

nive

rse

nucleinucleon gas

hadron gas colour superconductor

quark-gluon plasma (QGP)Tc

0

neutronstars

We should reach the QGP phase by compressing or heating hadronic matter to very high densities or temperatures...We do that by colliding heavy nuclei at very high energies, using the LHIC

How do we know that we produced a state of deconfined QCD matter?

18/35

QGP ?

We probe the produced matter by studying how it affects well understood probes,as a function of the temperature of the system (centrality of the nuclear collisions)

Calibrated“probe source”

Matter under study

Calibrated“probe meter”

CalibratedHeat Source

Probe

Probing the QCD matter produced in heavy-ion collisions

19/35

Quarkonia studies in nucleus-nucleus collisions: why?

In a deconfined phase the QCD binding potential is Debye screened and the heavy quarkonia states are “dissolved”. In other words, the free hard gluons are energetic enough to break the bound QQ states into open charm and beauty mesons.Different heavy quarkonium states have different binding energies and are, hence, dissolved at successive thresholds in energy density of temperature of the medium. Their suppression pattern is a thermometer of the produced QCD matter.

Latti

ce Q

Qba

r fr

ee e

nerg

y

T

H. Satz, hep-ph/0512217

20/35

A “smoking gun” signature of QGP formation

The feed-down from higher states leads to step-wise J/ and suppression patterns. Will there be any quarkonium states produced in Pb-Pb collisions at the LHC energies? It is very important to do these studies as a function of pT and of

collision centrality

state J/ c

' (1s) b

(2s) b' (3s)

Mass [GeV} 3.096 3.415 3.686 9.46 9.859 10.023 10.232 10.355B.E. [GeV] 0.64 0.2 0.05 1.1 0.67 0.54 0.31 0.2

Td/Tc --- 0.74 0.15 --- --- 0.93 0.83 0.74 1.10 0.74 0.15 2.31 1.13 0.93 0.83 0.74

21/35

Challenge: find the good probes of the produced QCD matter

vacuum

QGP

hadronicmatter

The good probes should be:

Well understood in “pp collisions”

Only slightly affected by the hadronic matter, in a very well understood way, which can be “accounted for”

Strongly affected by the deconfined QCD medium...

Heavy quarkonia (J/, ’, c, , ’, etc) are particularly good probes!

22/35

“Tomography” of QCD matter

Tomography:Uses a calibrated probe, and a well understood interaction, to derive the 3-D density profile of the medium from the absorption profile of the probe

In heavy-ion collisions:The suppression of the quarkonium states tells us if the matter they cross is confined or deconfined

23/35

Some results from the NA50 experiment

J/ normal nuclear

absorption curve

exp(-L abs)

In S-U and peripheral Pb-Pb collisions, the data points seem to follow the normal nuclear absorption, defined by the proton-nucleus data, which scales with the length of nuclear matter crossed by the J/.

In central Pb-Pb collisions the scaling is broken and an “anomalous suppression” sets in.

24/35

The J/ production yield measured in Pb-Pb collisions, relative to the DY yield,is suppressed from peripheral to central collisions... at low transverse momentum

The transverse momentum is an important observable

Ri =(NJ/ / NDY) (ETi)

(NJ/ / NDY) (ET1)

⟨pT

2 ⟩(G

eV

/c)2

J/

p-A 400 G

eV

S-U 200 G

eV

Pb-Pb 158 G

eV

L (fm)

Only the low pT J/ mesons get suppressed !

The J/ “central over peripheral ratio” strongly depends on pT (at the SPS)...

How much of this behaviour is due to “the Cronin effect”, and how much is dueto melting of low momentum ccbar states in the QCD plasma?

25/35

The ’ suppression pattern in S-U and in Pb-Pb shows a significantly stronger drop than expected from the “Glauber extrapolation” of the p-A data

abs ~ 20 mb

’

The “change of slope” is very significant and looks very abrupt...If the extra ’ suppression is due to the QGP, Lattice QCD says that this indicates that Tc is reached in the most peripheral S-U or Pb-Pb collisions at SPS energies...

The ’ is also suppressed from p-nucleus to nucleus-nucleus

’

26/35

Quarkonia studies in CMS

So far, only the dimuon decay channel has been considered.The physics performance has been evaluated with the 4 T field (2 T in return yoke) and requiring a full track in the muon chambers. A good momentum resolution results from the matching of the muon tracks to the tracks from the silicon tracker.

27/35

J/ studies in CMS: acceptances

The material between the silicon tracker and the muon chambers (ECAL, HCAL and the magnet’s iron) prevents the hadrons from giving a muon tag but imply that the muons must have a momentum above 3.5 GeV/c in the barrel and 4.0 GeV/c in the endcaps in order to be reconstructed.

This is not a problem for muons from the decay of the Upsilon states, given their high mass, but sets a relatively high threshold on the pT of the detected J/’s.

pT (

GeV

/c)

pT (GeV/c)

both muons|| < 2.4

both muons|| < 0.8

Betterlow pT J/

acceptanceat forwardrapidities

28/35

Upsilon studies in CMS: acceptances and mass resolutions

There is a good acceptance for dimuons in the Upsilon mass region.

Barrel: both muons in || < 0.8

Barrel + endcaps: muons in || < 2.4

The dimuon mass resolution is good enough to separate the three Upsilon states:~ 54 MeV only using the barrel and~ 86 MeV when including the endcaps

For the J/ mass region, the dimuon mass resolution is 35 MeV, in the full region

barrel + endcaps

29/35

Quarkonia studies in CMS: some open questions

Dimuon decays are well suited for the studies of Upsilon production while the charmonia states should be better studied through the ee decay channel.What is the J/ acceptance curve, vs. pT, for the J/ → ee decay channel?

Can we measure c production through the decay channel c → J/ + ?

Is the CMS ECAL sensitive to photons of around 1 GeV or even lower?Are such studies feasible at low luminosity but not any longer at design luminosity?

Can we use the ECAL and HCAL information to separate hadrons from muons?Can we accept as muon candidates tracks that only make it up to the first (of four) set of muon chambers?Would that result in a significant improvement of the low pT acceptance of the J/?

At which cost in terms of mass resolution and signal/background?

So far, the dimuon signals have been extracted as N – 2 × (N × N)Would the results significantly change if we would use a more correct approach?

Should we operate (the heavy-ion runs) at a lower magnetic field?Would that improve significantly the low pT acceptance of the J/?

At which cost in terms of mass resolution?

30/35

What is ATLAS doing in this field?

Quarkonia Physics in Heavy-Ion collisionswith the ATLAS Detector

Laurent RosseletECT-heavy flavor workshop, September 8th 2006

Quarkonia Physics in Heavy-Ion collisionswith the ATLAS Detector

31/35

Cut on the decay μ’s

+-

B/2

Full field

A compromise has to be found between acceptance and resolutionto clearly separate states with maximum statistics (e.g. |η| < 2)

32/35

global fit pT >3 GeV

global+tag || <1 || <2 || <2.5

Acceptance 2.6% 8.1% 12.0% +efficiency 4.7% 12.5% 17.5%

Resolution 123 MeV 145 MeV 159 MeV

S/B 0.4 0.3 0.3 0 0 0.3 0.2 0.2

S/√ S+B 31 45 55 u 37 46 55 Rate/month 10000 0 15000

reconstruction

|| <2

For |η| < 2 (12.5% acc+eff) we expect 15K /month of 106s at L=41026 cm-2 s-1

No improvement with the B/2 mode:acceptance/resolution ~ cte…

33/35

global fit B/2global+tag pT

>3 pT >1.5 pT

>1.5

Acceptance 0.039% 0.151% 0.529% +efficiency 0.055% 0.530% 1.100%

Resolution 68 MeV 68 MeV 76 MeV

S/B 0.5 0.2 0.25 0 0.4 0.15 0.15

S/√ S+B 52 72 140 u 56 113 164Rate/month 8000 30000 104000 0 11000 104000 216000

J/ reconstruction

|| <2.5pT

>1.5 GeV

We expect 8K to 216K J/+- per month of 106s at L=41026 cm-2 s-1

Resolution is 15% worse, but acceptance is 2-3 times better with B/2 Significance is also much better

34/35

ee, J/ ee

as a tracker:

simplest strategy for central Pb+Pb: keep the 2 first time steps (out of 13) ! of the drift tubes ! => occupancy of 30% as in pp

=> 4 to 6 additional hits for track reconstruction

=> improves mass resolution, reduces fake tracks

as a transition radiation detector:

defines a road where to look for transition radiation to identify electrons

=> the ATLAS e+e- trigger with pT> 2 GeV could be used to get and ! J/

e+e-

The Transition Radiation Tracker can be used fully if Nch is low enough ! partially in central Pb+Pb

Scenario under evaluation

35/35

A good road for the LIP people: quarkonia physics in CMS

The instructions are clear: we just need to follow the yellow brick road...