J/? and displaced vertex muon production

1
J/? and displaced vertex muon production

• Raison dêtre
• Detector concept
• Results on the J/?
• Open charm signal

2
Observations from previous experiments

• Since 1986, many experiments studied high-energy
  nuclear collisions at the CERN SPS to search for
  the QCD phase transition
• Some of the theory-driven signatures required
  measuring lepton pairs and motivated NA38, CERES,
  HELIOS-3 and NA50
• changes in the ? spectral function (mass shifts,
  broadening, disappearance)when chiral symmetry
  restoration is approached
• the production of thermal dimuonsdirectly
  emitted from the new phase, if in thermal
  equilibrium
• the suppression of charmonium states (J/?, ?,
  ?c)dissolved when certain critical thresholds
  are exceeded
• Some of the measurements done by those
  experiments were consistent with the expectations
  derived from the theoretical predictions if a QGP
  phase is formed

Talk of Rob Veenhof
This talk
3
Excess of intermediate mass dimuon production

• The yield of intermediate mass dimuons seen in
  heavy-ion collisions (S-U, Pb-Pb) exceeds the
  sum of Drell-Yan and D meson decays, which
  describes the proton data

NA38/NA50
proton-nucleus data
Pb-Pb data
4
Thermal dimuons or charm enhancement?
The intermediate mass dimuon yields in heavy-ion
collisions can be reproduced either
by scaling up the open charm contribution by up
to a factor of 3 (!)
or by adding thermal radiation from a
quark-gluon-plasma phase
NA38/NA50
? direct evidence of a thermalized
pre-hadronization phase
But the data collected by NA38/NA50 cannot
distinguish among these
5
J/? production from p-A to S-U collisions
The study of the J/? absolute production
cross-sections in p-A collisions at 200, 400 and
450 GeV, by NA3, NA38, NA50 and NA51, gives a
J/? absorption cross-section in normal nuclear
matter of 4.11 0.43 mb
No anomalous suppression is seen in light-ion
collisionsO-Cu, O-U and S-U, while the Pb-Pb
value is significantly lower, even when
integrated over all centralities
NA38/NA50
sabs 4.11 0.43 mb
6
J/? suppression in S-U and Pb-Pb collisions vs.
centrality
NA38/NA50

• No anomalous suppression is seen in S-U
  collisions,
• But the Pb-Pb values depart from the normal
  nuclear absorption curve for energy densities
  just above the most central S-U data? evidence
  of the QCD phase transition ?

7
Questions arising from these measurements

• Is the observed intermediate mass excess due to
  thermal dimuons from a QGP?
• Or is the open charm yield enhanced in
  nucleus-nucleus collisions?
• ? We need to measure secondary vertices with 50
  µm precision to separate prompt dimuons
  from D meson decays

• Is the J/? suppressed in the QGP phase?
• What is the physics variable driving the J/?
  suppression? L, Npart, energy density?
• ? Measure the J/? suppression pattern in
  Indium-Indium and compare it with Lead-Lead

• What is the impact of the ?c feed-down on the
  observed J/? suppression pattern?
• ? Study the nuclear dependence of ?c production
  in p-A collisions

New and accurate measurements needed, hence
8
Measuring dimuons
magnetic field
Energy loss Multiple scattering

• Cannot distinguish vertex muons from decay muons
• Degraded dimuon mass resolution

9
Measuring dimuons the NA60 way
2.5 T dipole magnet
beam tracker
vertex tracker
targets
Matching in coordinate and momentum space

• Origin of muons can be accurately determined
• Improved dimuon mass resolution

10
NA60 detector overview
Hadron absorber
Muon spectrometer
Vertexdipole magnet
Ironwall
Zero degree calorimeter (inside)
Beam
Toroidalmagnet
Tracking chambers
Target box
Vertex telescope
Beam tracker
11
Experimental requirements
High-energy (nuclear) collisions
Dimuonproduction
High statistics
High precision
As a function of centrality
Smallcross-section
High multiplicityenvironment
High luminosity
Fixed-target
Selectivedimuon trigger
State of the art silicon detectors
Radiation-hardzero degree calorimeter
Radiation tolerance
Accurate time tagging
Granularity
Segmented target
12
The silicon pixel vertex telescope

• 8 small 4-chip planes, plus
• 8 big 8-chip planes
• 2 X0 per plane

• 12 tracking points with good acceptance
• 9 X (bending plane, B2.5 T), and
• 3 Y

• 800000 channels, 50x425 µm2 pixel size

targets
beam tracker
13
The 2003 Indium run
5-week long in Oct.Nov. 2003 41012 ions
incident on target 230 million dimuon triggers
on tape
Centrality coveragemeasured by the ZDC and by
the number of tracksin the vertex telescope
Opposite-sign dimuon mass distributions before
event selection and muon track matching
14
J/? production in Indium-Indium collisions
Background
withmuon matching
J/?
Charm
?
DY
A multi-step fit is performeda) M gt 4.2 GeV
normalize the DY
b) 2.2 lt M lt 2.5 GeV normalize the charm
(DY
fixed)
c) 2.9 lt M lt 4.2 GeV get the J/? yield
(DY charm fixed)
15
J/? / Drell-Yan in Indium-Indium collisions
B s(J/?) / s(DY) 18.15 0.81 integrated over
all centralities
? 0.79 0.03 with respect to normal
nuclear absorption
After acceptance corrections Acc(J/?) 12.4
Acc(DY) 13.4
Preliminary
region within reach for centrality dependency
studies for Indium-Indium collisions
16
J/? transverse momentum and polarization

• Measuring the J/? polarization is important to
  understand quarkonium production
• CEM - no polarization
• NRQCD - transverse polarization at high pT
• So far (E866, CDF) no polarization seen
• Measured from the angular distribution of the µ
  in J/??µµ- decays, in the J/? rest frame

17
J/? polarization in Indium-Indium
In case of QGP formation (hep-ph/0306176) the
expected polarization is ? 0.6 (for pT
0) When taking into account the initial
transverse momentum of gluons, ? should still be
higher than zero ? 0.35 0.40
Our values seem closer to zero
as a function of pT
Preliminary
as a function of xF
as a function of number of participants
18
Down the matching and vertexing slope
Open charm
19
Fake muon matches
Fake matches use the closest non-muon track,
deteriorating kinematics and offset resolution.
Monte Carlo
Work in progress to subtract them by matching
muons with tracks from different events In the
present study fakes are not subtracted
Data
20
Combinatorial background from p,K?µ decays
Subtracted by event mixing muon pairs where each
muon comes from different like-sign events
controlled by comparingthe mixed event
like-sign dimuon spectra with the corresponding
measured data
21
Vertexing beam direction

• Robust algorithm resolves multiple vertices (if
  they are on different targets)
• Good target identification even for the most
  peripheral collisions (gt 3 tracks)

22
Vertexing transverse coordinates
Vertexing resolution and systematicscontrolled
using theBeam Tracker measurement (20 µm
resolution extrapolating to the target)
23
Measuring the muon offset
Offsets dX, dY between the vertex and the track
impact point in the transverse plane at
Zvertex Resolution depends on track momentum use
offset weighted by the covariance matrices of the
vertex and of the muon track
Fake matches tend to have large offsets, thus
degrading the charm selection capability. Problem
will be solved once fake subtraction is under
control
24
Discriminating prompt from offset muons
Additional cut on weighted distance, ?, between
muons at Zv to reduce influence of bad vertices
Cut on the weighted offset of the muon closest to
the vertex
25
Discriminating prompt from offset muons

• Only 20 of the total statistics has been
  explored so far
• Selection cuts still to be optimized(once
  subtraction of fake matches is done)

Ratio betweenoffset and prompt dimuons

• Reduced offset yield at the dimuon masses
  dominated by the ?, ? and J/?
• Remaining shape of the offset sample resembles
  expectations from open charm decays

26
Summary 2003 Indium run

• The Indium-Indium data collected in 2003, at 158
  A GeV, allows NA60 to
• Identify the physics variable driving the
  suppression of the J/? meson
• We presented the cross section ratio J/? / DY in
  Indium-Indium collisions, integrated over all
  centralities, and a preliminary study of the J/?
  pT and polarization angle
• Understand the origin of the intermediate mass
  dimuon excess
• A charm signal is already visible. Work is in
  progress to subtract fake matches and include
  individual pixel efficiencies in the Monte Carlo
  simulations
• Furthermore, we will also be able to
• Measure the absolute cross-section of D meson
  production in Indium collisions
• Study the onset of the ? suppression, in a
  single collision system

27
Outlook 2004 proton-nucleus run
7 nuclear targets at two proton beam energies
400 and 158 GeV Collected 300000 J/? events
in p-A collisions at 400 GeV At 158 GeV we
should be able to determine a(J/?) with an error
better than 1 It should also allow us to
measure the fraction of J/? mesons resulting from
decays of the ?c states and its nuclear dependence
p-A _at_ 158 GeV
Beam of 4 109 p/burst
30000 J/? events 3 days data
p beam158 GeV
Al
U
W
Cu
In
3 x Be
Pb
(before muontrack matching)
28
Anyquestions heavier than1.0 GeV ?