From High-Energy Heavy-Ion Collisions to Quark Matter
Episode III : Back to the future
SPS
From the SPS and RHIC to the LHC
RHIC
LHC
Carlos Lourenço, CERN
CERN, July, 2010
1
QGP physics from SPS to RHIC
• SPS : 1986 – 2003 : Pb-Pb and In-In at s = 20 GeV
 J/y and y’ (and cc ?) suppression  deconfinement
 thermal dimuon production  thermal QCD medium
 compelling evidence for a “new state of matter” with “QGP-like properties”
• RHIC : 2000 – ?? : Au-Au at s = 200 GeV
 jet quenching: parton energy loss  very dense QCD medium
 baryon/meson elliptical flow scaling  partonic degrees of freedom
 compelling evidence for a strongly-coupled QGP (the “perfect fluid”)
2
“Quark soup” and “perfect fluids”
RHIC’s Quark Soup is made of “liquid stuff”
Black holes and perfect
fluids in physics today
3
The “perfect fluid” found at RHIC
M. Roirdan and W. Zajc,
Scientific American, May 2006
4
The “perfect fluid” in the press
When physicists talk about a perfect liquid, they don’t
mean the best glass of champagne they ever tasted.
The word “perfect” refers to the liquid’s viscosity…
A perfect liquid has no viscosity at all, which is impossible in reality but useful for theoretical
discussions… material swallowed by black holes might also have extremely low viscosity…
5
QGP physics from SPS / RHIC to the LHC
• LHC : 2010 – ?? : Pb-Pb at s = 2750 GeV (and 5500 GeV later on)
 confirm interpretation of SPS & RHIC results by testing new predictions
 explore & understand high-density QCD properties with original measurements
heavy quarks (charm, beauty), jets, upsilons
 is the initial state at the LHC yet another state of matter ?
colour glass condensate ? (QCD in the classical field theory limit)
 transition from a strongly coupled QGP to an ideal QGP ?
 surprises ? more puzzles ?
what will little Alice find behind the curtain ?
6
One small step for a man, a giant leap for mankind…
SPS
RHIC
LHC
√sNN (GeV)
17
200
dNch/dy
500
850
1500-4000
t0QGP (fm/c)
1
0.2
0.1
T/Tc
1.1
1.9
3-4
Hotter
e (GeV/fm3)
3
5
15-60
Denser
tQGP (fm/c)
≤2
2-4
≥10
5500 [ 2750 in 2010 ]
Longer
tf (fm/c)
~10
20-30
30-40
Vf (fm3)
few 103
few 104
few 105
Bigger
The LHC is a giant leap forward in QGP physics, well beyond previous facilities
7
LHC “nominal” running parameters
Collision system
√sNN (TeV)
L0 (cm-2s-1)
pp
14
1034
Pb-Pb
5.5
1027
p-Pb
8.8
1029
Ar-Ar
6.3
1029
<L>/L0 (%)
Run time (s/year)
107
50
106
106
65
106
Expected integrated luminosity in a typical Pb-Pb run : Lint (Pb-Pb) ~ 0.5 nb-1/year
The LHC is expected to run “heavy-ions” for around 1 month each year
8
Hard Probes of QCD matter at LHC energies
• Very large cross sections at the LHC
pp s = 5.5 TeV
• Pb-Pb instant. luminosity: 1027 cm-2s-1
• ∫ L dt = 0.5
1 mb
nb-1
(1 month, 50% running efficiency)
J/y
• Hard cross sections: Pb-Pb = A2 x pp
 pp-equivalent ∫ L dt = 20 pb-1
1 nb

h+/h-
 1 event limit at 0.05 pb (pp equiv.)
g*+jet
Note: one Pb-Pb collision is roughly
equivalent to 40 000 pp collisions…
jet
Z0+jet
1 pb
gprompt
1 event
9
Solenoid magnet 0.5 T
ALICE
Forward detectors
Specialized detectors:
• HMPID
• PHOS
Central tracking system:
• ITS
• TPC
• TRD
• TOF
Muon spectrometer:
• absorbers
• tracking stations
• trigger chambers
• dipole
10
A global view of the ALICE experiment
• Covers very low-pT (~ 100 MeV/c) and high-pT (> 100 GeV/c)
• Has particle identification over a large momentum range
• Is able to handle large charged particle multiplicities
• Will measure open charm, beauty, direct photons, J/y, etc
11
The ALICE TPC
Readout chambers
Largest TPC ever
88 m3, 570k channels
90% Ne – 10% CO2
Field cage
End plate
HV membrane
2/19/2007
Raimond.Snellings@nikhef.nl
12
Other ALICE detectors
Silicon drift
TRD
Silicon pixels
TOF
Muon slats
ZDC
PHOS PbWO4 crystals
13
ALICE had its first detector upgrade already
Lead-scintillator sampling calorimeter
Shashlik fiber geometry
Avalanche photodiode readout
Coverage: | | < 0.7,
= 110°
~13k towers (
x
~ 0.014 x 0.014)
Design resolution: E/E ~ 1% + 0.08/√E
The first EMCal modules were installed in March
2009 between the magnet and the “space frame”
that holds the TPC and other central detectors
14
ATLAS will also study QGP physics with Pb-Pb collisions
• ATLAS is fully operational
• Extensive preparations for the Pb-Pb program show a promising performance
15
ATLAS will measure jets in Pb-Pb collisions
Fragmentation function: D(z)
ATLAS simulation
ATLAS simulation
Reliable reconstruction of D(z):
Reconstructed tracks with pT > 2 GeV
matching calorimeter jets
Comparing PYTHIA to PYQUEN
gives the scale of possible
modifications of the fragmentation
function in Pb-Pb
ATLAS can measure jet quenching of
the size simulated by PYQUEN
16
Inclusive jet production in pp collisions
 Observed inclusive jet multiplicity
distribution (top left)
 Inclusive pT distribution for jets with
pT > 30 GeV and |y|< 2.8 (top right)
 Jet‐jet mass distributions (left)
Data compared to PYTHIA Monte Carlo
calculations (yellow histogram)
17
A global view of the CMS experiment
18
Phase space coverage of the CMS detector
CMS + TOTEM: full φ and almost full η acceptance at the LHC
 charged tracks and muons: |η| < 2.5
 electrons and photons: |η| < 3
 jets, energy flow: |η| < 6.7 (plus η > 8.3 for neutrals, with the ZDC)
 excellent granularity
HF
h = -8
-6 -4 -2
and resolution
 very powerful
High-Level-Trigger
HF
0
2
4
6
8
19
h±, e±, g, m± measurement in the CMS barrel (|h| < 2.5)
Si Tracker
+
ECAL
+
muon-chambers
Si Tracker
Calorimeters
Muon Barrel
Silicon micro-strips
and pixels
ECAL
HCAL
Drift Tube Chambers (DT)
Resistive Plate Chambers (RPC)
PbWO4
Plastic Sci/Steel sandwich
20
Charm and beauty yields vs. energy and collision system
Charm cross section at the LHC is higher
Including EKS98 shadowing
by a factor ~ 10 w.r.t. RHIC energies and
SPS
RHIC LHC LHC
LHC
by a factor ~ 1000 w.r.t. SPS energies:
central central pp
p-Pb central
pp
• s = 20 GeV  scc ~ 5 mb
Pb-Pb Au-Au
Pb-Pb
N(cc)/event
0.2
10
0.16 0.8
115
• s = 200 GeV  sccpp ~ 600 mb
-0.05 0.006 0.03
4.6
• s = 5.5 TeV  sccpp ~ 6600 mb N(bb)/event
Abundance of charm production at the LHC
will enable detailed studies of several topics,
including charm thermalisation
The detection of D and B mesons requires
an accurate determination of the collision
vertex and of the distance between the
extrapolated charged tracks and the vertex
Typical impact parameters: a few 100 mm
for D decays and ~500 mm for B mesons
21
Heavy flavour production at LHC energies
Initial state effects:
Heavy Quark energy loss:
Nuclear shadowing suppresses low-pT heavy
flavoured particles in p-A and A-A collisions:
~ 20% reduction of beauty production and
~ 40% reduction of charm (EKS98)
Parton energy loss is expected to occur by:
• medium-induced gluon radiation
• collisions in the medium
It is expected to depend on the properties of
the medium (length, energy density, etc.)
DE (L, eQGP)
Pb-Pb / pp
and also on the quark mass
s = 5.5 TeV
beauty
We will probe heavy quark energy loss
through ratios of pT distributions, between
Pb-Pb and pp, between B and D mesons, etc
charm
We will also do these studies using jets
tagged by the presence of D or B mesons
22
Quarkonia studies in ALICE with dimuons
Pb-Pb simulation
J/y
Y
Y’, Y’’
After background
subtraction
Rapidity window:
2.4–4.0
Mass resolution:
70 MeV at the J/y
100 MeV at the Y
Mmm (GeV/c2)
pp data !
ALICE has been collecting pp collisions
at 7 TeV and has already seen a J/y
peak in the dimuon channel
dimuon mass resolution : 80 – 90 MeV
Figure shown at the LHCC meeting,
at CERN, on Wednesday, July 7, 2010
23
Quarkonia studies in ALICE with electron-positron pairs
Combining the ITS, TPC and TRD data,
available for |h| < 0.9, ALICE can access
vertexing information for the electrons
Pb-Pb simulation
Good mass resolution, ~50 MeV, due to
the low material budget of ALICE, which
was measured with photon conversions
pp data
Figures shown at the LHCC meeting,
at CERN, on Wednesday, July 7, 2010
24
Tomography of the CMS and ATLAS inner detectors
Figures shown at the LHCC meeting,
at CERN, on Wednesday, July 7, 2010
25
Quarkonia studies have also started in CMS and ATLAS
J/y + continuum
pp data
events / 20 MeV
full rapidity
all muons
In CMS, it is ~40 MeV integrating the
full rapidity coverage and ~20 MeV at
mid-rapidity
mass resolution
= 43 MeV
J/y + continuum
continuum
central rapidity
golden muons
pp data
events / 80 MeV
continuum
The good mass resolution results from
the matching of the muon tracks to the
tracks in the silicon tracker
pp data
mass resolution
= 21 MeV
Figures shown at the LHCC meeting,
at CERN, on Wednesday, July 7, 2010
26
Measuring beauty yields from displaced J/y production
entries / event
prompt
J/y
J/y from B
prompt
J/y
J/y from B
simulation
J/y  m+m-
simulation
J/y  e+e-
vtr (mm)
Many of the J/y mesons observed at the LHC
come from decays of B mesons
They can be separated from the “prompt” J/y mesons
because they are produced away from the collision vertex
27
A decay of a B meson to J/y + K seen in CMS
CMS experiment at LHC, CERN
Run 136100 / Event 256858438
2010-25-5 03:43:48 CEDT
secondary
vertex
B- → J/y K- candidate
primary
vertex
Figures shown at the LHCC meeting,
at CERN, on Wednesday, July 7, 2010
28
Wake up! we are almost finished…
Oh my God…
Dear Dalai
Lama…
29
 studies in CMS
J/y
Upsilons have been simulated
in Pb-Pb collisions… and already
measured in pp collisions !
 (1S)
y’
Pb-Pb @ 5.5 TeV
dNch/dh = 3500
  m+mFigure shown at the LHCC meeting,
at CERN, on Wednesday, July 7, 2010
30
Expected pT reach of quarkonia measurements in Pb-Pb
● produced in 0.5 nb-1
■ rec. if dN/dh ~ 2500
○ rec. if dN/dh ~ 5000
J/y
0.5 nb-1 : 1 month at 4x1026 cm2s-1
Expected rec. quarkonia yields:
J/y : ~ 180 000  : ~ 26 000
Pb-Pb
Pb-Pb at 5.5 TeV
design luminosity

CMS simulation
Statistical accuracy (with HLT) of
’ /  ratio vs. pT should be good
enough to rule out some models
Similar low pT yields
for J/y and 
with HLT
31
The CMS High Level Trigger
• CMS High Level Trigger:
12 000 CPUs of 1.8 GHz ~ 50 Tflops !
• Executes “offline-like” algorithms
• pp design luminosity L1 trigger rate: 100 kHz
• Pb-Pb collision rate: 3 kHz (peak = 8 kHz)
 pp 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
Pb-Pb at 5.5 TeV
design luminosity
ET reach x2
jets
x30
x30
• Using the HLT, we can keep all the interesting
events (hard processes) and reduce by a factor
around 30 the more common events
32
Impact of the HLT on the pT reach of RAA
Nuclear modification factor =
“QCD medium” / “QCD vacuum”
Pb-Pb 0.5 nb-1
Pb-Pb 0.5 nb-1
HLT
CMS simulation
Important measurement to compare with parton energy loss models and derive
the initial parton density, dNg/dy, and the medium “transport coefficient”
33
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
Gluon radiation:
large angle (out-of-cone)
vs. small angle emission
CMS simulation
HLT
34
g, g* and Z tagging of jet production
photon or Z
The dense QCD medium redistributes the initial parton
energy, Ejet, in the hadron jet
This redistribution is measured in the Fragmentation
Function… if we know Ejet
But it is very difficult to access Ejet in HI collisions,
because of the medium modifications…
g
away
side
g*
jet
(hadrons)
Sometimes, the parton that fragments to a jet is
produced back-to-back with a photon: Eg = Ejet
Measuring the photon, unaffected by the medium,
gives an ideal way to calibrate the jet energy loss
Z0+jet
The Z0 can also be used: large production cross
sections at LHC energies and easy to detect
CMS simulation
35
Lessons from the SPS and RHIC to the LHC
Before the measurements are made,
theorists often think that the interpretation
of the data will be easy
However, theorists are often wrong…
especially before the measurements are made
This is a data-driven field; the SPS and RHIC
“learning curves” gave us clear directions
concerning the path to follow at the LHC...
We will find the way out...
36
“Take some more tea”, the March Hare said to Alice, very earnestly.
“I've had nothing yet”, Alice replied in an offended tone, “so I can't take more”.
“You mean you can't take LESS”, said the Hatter:
“it's very easy to take MORE than nothing”.
Lewis Carroll
Alice in Wonderland
We are looking forward to
“take some more”
LHC Pb-Pb collisions…
37