QCD and the QGP at the LHCMarco van Leeuwen, 


Nikhef and Utrecht University

IoP colloquium UvA, Amsterdam 19 March 2015

2

Structure of matter

Quarks and gluons are the building blocks of nuclear matter Main interaction: Strong interaction (QCD)

r0 1 2 3 4 5 6

(r)QED

V

10−

8−

6−

4−

2−

0

3

The QCD potential

S. Deldar, hep-lat/9909077

QCDstrong interaction

QEDelectromagnetic interaction

PotentialField lines in dipole system

QCD is very different from EM, gravity; common intuition may fail

4

QCD strings

G.S. Bali, hep-ph/0411206

A simple picture of the strong interaction

QCD potential for 2-quark system
rises indefinitely

For larger separation: generating a qqbar pair is energetically favoured

Thought experiment: 
separating charges

Color charges (quarks and gluons) cannot be freedConfinement important at length scale 1/ΛQCD ~ 1 fm

5

The extremes of QCD

This is the basic theory, but what is the phenomenology?

Small coupling 
Quarks and gluons 


are quasi-free

Calculable with pQCD

Two basic regimes in which QCD theory gives quantitative results: Hard scattering and bulk matter

QCD Lagrangian

Nuclear matter Quark Gluon Plasma

High density Quarks and gluons 


are quasi-free

Bulk QCD matter

Calculable with Lattice QCD

Hard scattering

6

The Quark Gluon Plasma

Bernard et al. hep-lat/0610017

Tc ~ 170 -190 MeV

Energy density from Lattice QCD

Deconfinement transition: sharp rise of energy density at Tc Increase in degrees of freedom: hadrons (3 pions) -> quarks+gluons (37)

εc ~ 1 GeV/fm3

4gT∝εg: deg of freedom

Nuclear matterQuark Gluon Plasma

7

RHIC and LHC

STARSTAR

RHIC, Brookhaven LHC, GenevaAu+Au √sNN= 200 GeV Pb+Pb √sNN= 2760 GeV

First run: 2000 First run: 2009/2010

STAR, PHENIX,PHOBOS, BRAHMS

ALICE, ATLAS, 
CMS, (LHCb)

Currently under maintenanceRestart 2015 with higher energy: 


pp √s = 13 TeV, PbPb √sNN = 5.02 TeV

A nucleus-nucleus collision

8

Colored spheres: quarksWhite spheres: hadrons, i.e. bound quarks

In a nuclear collision, a Quark-Gluon Plasma (liquid) is formed⇒ Study this new state of matter

9

Size: 16 x 26 meters Weight: 10,000 tons

Technologies:18 Tracking: 7 PID: 6 Calo. : 5 Trigger, Nch:11

ALICE

Optimised for low momentum, high-multiplicity 
tracking and Particle Identification

10

Heavy ion collisions

‘Hard probes’ Hard-scatterings produce ‘quasi-free’ partons

⇒ Probe medium through energy loss pT > 5 GeV

Heavy-ion collisions produce
‘quasi-thermal’ QCD matter

Dominated by soft partons 
p ~ T ~ 100-300 MeV

‘Bulk observables’ Study hadrons produced by the QGP

Typically pT < 1-2 GeV

Two basic approaches to learn about the QGP 1) Bulk observables 2) Hard probes

11

Part I: the bulk; QGP fragments

12

pT-spectra – radial flow

Mass dependence: same Lorentz boost (β) gives larger momentum for heavier particles (mp > mK > mπ)

Large increase in mean pT from RHIC (√sNN=200 GeV) to LHC

First indication of collective behaviour; pressure

PRL109, 252301, PRC, arXiv:1303.0737

13

Hydrodynamics

Kolb, S

ollfrank, Heinz, P

RC

62, 054909

0=∂ µνµTEnergy-momentum conservation

Continuity equations for
conserved currents

0=∂ µµ ijEquation of state

Measured spectra shapes agree
with hydrodynamical calculation ⇒ It really looks like a fluid

Heinz&Song

PLB, arXiv:1401.1250

14

Time evolution

All observables integrate over evolution

Radial flow integrates over entire ‘push’

15

Elliptic flow

Hydrodynamical calculation

Reac

tion p

lane

Anisotropy reduces during evolution 
v2 more sensitive to early times

Elliptic flow: Yield modulation in-out reaction plane

reaction plane

ϕ

b

( )ϕϕ

2cos21 2vNddN

+=

16

Elliptic flow

Mass-dependence of v2 measures flow velocityGood agreement between data and hydro

( )ϕϕ

2cos21 2vNddN

+=

17

Higher harmonicsAlver and Roland, PRC81, 054905

3rd harmonic ‘triangularity’ v3 is large (in central events)

Dominant effect in azimuthal correlations 
at pT = 1-3 GeV

Mass ordering also seen for v3 indicates collective flow

18

Higher harmonicsSchenke and Jeon, Phys.Rev.Lett.106:042301

In general: initial state may be ‘lumpy’ (not a smooth ellipse)

η/s = 0

η/s = 0.16

How much of this is visible in the final state, depends on shear viscosity η

19

ViscosityViscous liquids dissipate energy

For a dilute gas:

η increases with T

Liquid, densely packed, so:TEvace /−∝η

Evac: activation energy for jumps of vacancies

η decreases with T

Liquid

gasTEvace /−∝η

21

T∝η

Viscosity minimal at liquid-gas transition

QGP viscosity lower than any atomic matter

Probing the Quark-Gluon Plasma

20

Detector

Probe beam

particles

Not feasible:Short life time

Small size (~10 fm)Use self-generated probe:

quarks, gluons from hard scattering 
 large transverse momentum

21

Nuclear modification factor at RHIC

Oversimplified calculation: -Fit pp with power law -Apply energy shift or relative E loss

Not even a model !

Ball-park numbers: ΔE/E ≈ 0.2, or ΔE ≈ 3 GeV 
for central collisions at RHIC

π0 spectra Nuclear modification factorPH

EN

IX, P

RD

76, 051106, arXiv:0801.4020

RHIC √sNN = 200 GeV

22

From RHIC to LHC

RHIC: 200 GeV LHC: 2.76 TeV
 per nucleon pair

LHC: spectrum less steep, 
larger pT reach

RHIC: n ~ 8.2 LHC: n ~ 6.4

Fractional energy loss:

RAA depends on n, steeper spectra, smaller RAA

23

From RHIC to LHC

RHIC LHC

RHIC: n ~ 8.2 LHC: n ~ 6.4

( ) 20.023.01 2.6 =− ( ) 32.023.01 4.4 =−Energy loss at LHC is larger than at RHIC RAA is similar due to flatter pT dependence

24

Towards a more complete picture

• Geometry: couple energy loss model to model of evolution of the density (hydrodynamics)

• Energy loss not single-valued, but a distribution • Energy loss is partonic, not hadronic

– Full modeling: medium modified shower – Simple ansatz for leading hadrons: energy loss followed by

fragmentation – Quark/gluon differences

25

Medium-induced radiation

2ˆ~ LqE Smed αΔ

propagating parton

radiated gluon

Key parameter:
Transport coefficient

Mean transverse kick per unit path length

Depends on density ρ through mean free path λ

Fitting the model to the data

26

Burke et al, JET Collaboration, arXiv:1312.5003

10 20 30 40 50 60 70 80 90 1000

0.2

0.4

0.6

0.8

1

AA

R

(GeV)T

p

/fm2=1.4, 1.8, 2.2, 2.6, 3.0 GeV0

q

CMS (0-5%)Alice (0-5%)

6 8 10 12 14 16 18 200

0.2

0.4

0.6

0.8

1

AA

R

(GeV)T

p

/fm2=0.8, 1.0, 1.2, 1.4, 1.7 GeV0

q

PHENIX 2008 (0-5%)

PHENIX 2012 (0-5%) RHIC

LHC

1 1.5 2 2.5 3 3.50

1

2

3

4

/d.o

.f2

χ/fm (LHC)2 GeV

0q

PHENIX 08+12

CMS+ALICE

𝜒2 of data wrt model

Clear minimum: found best value 
for transport coefficient

Factor ~2 larger at LHC than RHIC

Comparing several models

27

0

1

2

3

4

5

6

7

0 0.1 0.2 0.3 0.4 0.5

McGill-AMY

HT-M

HT-BW GLV-CUJET

MARTINI

Au+Au at RHIC

Pb+Pb at LHCqN/T

3 (DIS)eff

ˆ

T (GeV)

q/T

3ˆ

values from different models agree

larger at RHIC than LHC

RHIC:

LHC:

Burke et al, JET Collaboration, arXiv:1312.5003Expect factor 2.2 from 


multiplicity + nuclear size

(T=370 MeV)

(T=470 MeV)

Transport coefficient and viscosity

28

Transport coefficient:momentum transfer per unit path length

Viscosity: General relation:

Expect for a QCD medium

Majumder, Muller and Wang, PRL99, 192301

Relation transport coefficient and viscosity

29

H. Song et al, PRC

83, 054912

0 20 40 60 80centrality

0

0.02

0.04

0.06

0.08

0.1

v2

RHIC: η/s=0.16LHC: η/s=0.16LHC: η/s=0.20LHC: η/s=0.24

STAR

ALICEv

2{4}

MC-KLNReaction Plane

0

1

2

3

4

5

6

7

0.2 0.25 0.3 0.35 0.4 0.45 0.5

Au+Au at 0.2 TeV

Pb+Pb at 2.76 TeVqN/T3 (DIS)effˆ

T (GeV)

0

q/T

3ˆ

≈≈

??

????

NL

O S

YM

Increase of η/s and decrease of q/T3 with collision energyare probably due to a common origin, e.g. running 𝛼S

Elliptic flow

(Scaled) viscosity slightly larger at LHC

Transport coefficient from RAA

Scaled transport coefficient
 slightly smaller at LHC

Results agree reasonably well with expectation:

Summary

• Heavy ion collisions study hot and dense nuclear matter• Two main experimental approaches:

• Study QGP fragments, e.g. elliptic/triangular flow 
Indicates very low value of η/s


• Probe QGP with self-generated probes, e.g. high-pT particle production 
Large density, energy loss


• Both observations consistent with very dense system, small mean free path

30

Run 2 of the LHC: larger data samples to explore 
flow, parton energy loss mechanisms

Extra slides

32

The Inner Tracking System

1698 double sided strip sensors 73 * 40 mm2 300 um thick

768 strips on each side

Strip detector, cross section

• Charged particle generates 
free electrons+holes

• Drift (E-field) to p, n-doped strips • Detect image charge in Al strips

+ 2 layers of Silicon Drift detectors + 2 layers of Silicon Pixel detectors

Dutch contribution to ALICE

33

Geometry

Density profile

Profile at τ ~ τform known

Density along parton path

Longitudinal expansion 
dilutes medium ⇒ Important effect

Space-time evolution is taken into account in modeling

34

)/()( , jethadrTjetshadrT

EpDEPdEdN

dpdN

⊗Δ⊗=

`known’ from e+e-known pQCDxPDF

extract

Parton spectrum Fragmentation (function)Energy loss distribution

This is where the information about the medium isP(ΔE) combines geometry 
with the intrinsic process


– Unavoidable for many observables

Notes: • This is the simplest ansatz – most calculation to date use it (except some

MCs) • Jet, γ-jet measurements ‘fix’ E, removing one of the convolutions

A simplified approach

0

0.2

0.4

0.6

0.8

20 40 60 80 100

RAA

(pT)

pT (GeV/c)

CUJET2.0(αmax,fE,fM)

CMS 0-5%ALICE 0-5%

(0.20,1,0)(0.21,1,0)(0.22,1,0)(0.23,1,0)(0.24,1,0)(0.25,1,0)(0.26,1,0)(0.27,1,0)(0.28,1,0)(0.29,1,0)(0.30,1,0)(0.31,1,0)(0.32,1,0)

RHIC and LHC

35

Burke et al, JET Collaboration, arXiv:1312.5003

0

0.2

0.4

0.6

0.8

6 8 10 12 14 16 18 20

RAA

(pT)

pT (GeV/c)

CUJET2.0(αmax,fE,fM)

PHENIX 0-5% 2012PHENIX 0-5% 2008(0.20,1,0)(0.21,1,0)(0.22,1,0)(0.23,1,0)(0.24,1,0)(0.25,1,0)(0.26,1,0)(0.27,1,0)(0.28,1,0)(0.29,1,0)(0.30,1,0)(0.31,1,0)(0.32,1,0)

10 20 30 40 50 60 70 80 90 1000

0.2

0.4

0.6

0.8

1

AA

R

(GeV)T

p

/fm2=1.4, 1.8, 2.2, 2.6, 3.0 GeV0

q

CMS (0-5%)Alice (0-5%)

6 8 10 12 14 16 18 200

0.2

0.4

0.6

0.8

1

AA

R

(GeV)T

p

/fm2=0.8, 1.0, 1.2, 1.4, 1.7 GeV0

q

PHENIX 2008 (0-5%)

PHENIX 2012 (0-5%)

RHICRHIC

LHC LHC

Systematic comparison of energy loss models with dataMedium modeled by Hydro (2+1D, 3+1D)

pT dependence matches reasonably well

RHIC and LHC

36

1 1.5 2 2.5 3 3.50

1

2

3

4

/d.o

.f2

χ/fm (LHC)2 GeV

0q

PHENIX 08+12

CMS+ALICE

0

2

4

6

8

10

0.2 0.22 0.24 0.26 0.28 0.3 0.32

χ2/d

.o.f.

(pT>

8)

αmax

PHENIX 08+12CMS+ALICE

CUJET 2.0 HT-BW

CUJET: 𝛼s is medium parameterLower at LHC

HT: transport coeff is parameterHigher at LHC

Burke et al, JET Collaboration, arXiv:1312.5003

37

Nuclear geometry: Npart, Ncoll

b

Two limiting possibilities: - Each nucleon only interacts once, ‘wounded nucleons’


Npart = nA + nB (ex: 4 + 5 = 9 + …) 
Relevant for soft production; long timescales: σ ∝ Npart

- Nucleons interact with all nucleons they encounter
Ncoll = nA x nB (ex: 4 x 5 = 20 + …) 
Relevant for hard processes; short timescales: σ ∝ Nbin

38

Nuclear modification factor RAA

p+p

A+A

pT

1/N

bin d

2 N/d

2 pT

‘Energy loss’

Shift spectrum to left

‘Absorption’

Downward shift

Measured RAA is a ratio of yields at a given pT The physical mechanism is energy loss; shift of yield to lower pT

The full range of physical pictures can be 
captured with an energy loss distribution P(ΔE)

39

Nuclear modification factor

ppTcoll

PbPbTAA dpdNN

dpdNR

+

+=/

/

Suppression factor 2-6 Significant pT-dependence Similar at RHIC and LHC?

So what does it mean?

Quarks and the strong interaction

40

Atom Electron 
elementary, point-particle

Protons, neutrons Composite particle ⇒ quarks

up charm top down strange bottom

Quarks: Electrical charge Strong charge (color)

electron Muon Tau νε νµ ντ

Leptons: Electrical charge

Force carriers:photon EM force gluon strong force W,Z-boson weak force

Standard Model: elementary particles

+anti-particles

EM force binds electrons
to nucleus in atom

Strong force binds nucleons
in nucleus and quarks in nucleons

41

QCD and quark parton model

At low energies, quarks 
are confined in hadrons

Goal of Heavy Ion Physics: Study dynamics of QCD and confinement

in many-body systems

gqqee →−+At high energies, quarks and

gluons are manifest

protons, neutrons, pions, kaons

+ many othersExperimental signature: jets of hadrons

42

ALICE in real life

43

Collision centrality

Central collisionPeripheral collision

top/side 
view:

front view:

b~0 fm

b

Nuclei are large compared to the range of strong force

b finite

This talk: concentrate on central collisions

44

Centrality continuedcentralperipheral

Multiplicity distribution

Experimental measure of centrality: multiplicityNeed to take into account volume of collision zone for production rates

45

Testing volume (Ncoll) scaling in Au+Au

PHENIX

Direct γ spectra

Scaled by Ncoll

PHENIX, PRL 94, 232301

Direct γ in A+A scales with Ncoll

Centrality

A+A initial production is incoherent superposition of p+p for hard probes

46

π0 RAA – high-pT suppression

Hard partons lose energy in the hot matter

γ: no interactions

Hadrons: energy loss

RAA = 1

RAA < 1

π0: RAA ≈ 0.2

γ: RAA = 1

PHENIX@RHIC√sNN = 200 GeV