Jets and high-pT probes in ALICE
Marco van Leeuwen,
Nikhef & Utrecht University
ALICE junior meeting, Wuhan
17 October 2013
2
Introduction
Data/results in order of experimental difficulty level:
• Inclusive spectra
• Di-hadrons
• Jets
Goal of the talk: discuss understanding of energy loss
in terms of physical mechanisms and theory/modelling
Interpretation aspects:
- Mean energy loss
- Geometry
- Energy loss distrubution
For a list of ongoing analyses and new ideas in ALICE, see my talk from Tuesday
4
Singularities in pQCD
Divergence of perturbative calculation is absorbed in hadronisation effects
(massless case)
Soft divergence Collinear divergence
p
rad
1This is why jets are collimated:
5
Hard processes in QCD
• Hard process: scale Q >> LQCD
• Hard scattering High-pT parton(photon) Q ~ pT
• Heavy flavour production m >> LQCD
Cross section calculation can be split into
• Hard part: perturbative matrix element
• Soft part: parton density (PDF), fragmentation (FF)
Soft parts, PDF, FF are universal: independent of hard process
QM interference between hard and soft suppressed (by Q2/L2 ‘Higher Twist’)
Factorization
c
chbbaa
abcd
ba
T
h
pp
z
Dcdab
td
dQxfQxfdxdxK
pdyd
d
0
/22
2)(
ˆ),(),(
parton density matrix element FF
Probably not strictly true in Pb+Pb; take as working hypothesis for now
6
Fragmentation and parton showers
large Q2 Q ~ mH ~ LQCD mF
Analytical calculations: Fragmentation Function D(z, m) z=ph/Ejet Only longitudinal dynamics
High-energy
parton
(from hard scattering)
Ha
dro
ns
MC event generators
implement ‘parton showers’ Longitudinal and transverse dynamics
7
Soft QCD matter and hard probes
Hard-scatterings produce ‘quasi-free’ partons Initial-state production known from pQCD Probe medium through energy loss
Heavy-ion collisions produce QCD matter
Dominated by soft partons p ~ T ~ 100-300 MeV
‘Hard Probes’: sensitive to medium density, transport properties
8
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: Npart
- Nucleons interact with all nucleons they encounter
Ncoll = nA x nB (ex: 4 x 5 = 20 + …)
Relevant for hard processes; short timescales: Nbin
Transverse view
Eccentricity
Density profile r: rpart or rcoll
22
22
xy
xy
x
y
L
9
Testing volume (Ncoll) scaling in Au+Au
PHENIX
Direct g spectra
Scaled by Ncoll
PHENIX, PRL 94, 232301
ppTcoll
AuAuT
AAdpdNN
dpdNR
/
/
Direct g in A+A scales with Ncoll
Centrality
A+A initial state is incoherent superposition of p+p for hard probes
10
Nuclear modification factor RAA
p+p
Au+Au
pT
1/N
bin d
2N
/d2p
T
‘Energy loss’
Shifts 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
ppTcoll
PbPbT
AAdpdNN
dpdNR
/
/
11
Getting a sense for the numbers – RHIC
Oversimplified calculation:
- Fit pp with power law
- Apply energy shift or relative E loss
Not even a model !
Ball-park numbers: DE/E ≈ 0.2, or DE ≈ 3 GeV
for central collisions at RHIC
0 spectra Nuclear modification factor
PH
EN
IX, P
RD
76, 0
51106, a
rXiv
:0801.4
020
12
From RHIC to LHC RHIC: 200 GeV
LHC: 2.76 TeV
per nucleon pair
Energy ~24 x higher
LHC: spectrum less steep,
larger pT reach
n
T
TT
pdp
dN
p
2
1
RHIC: n ~ 8.2
LHC: n ~ 6.4
Fractional energy loss:
2
1
D
n
AAE
ER
RAA depends on n, steeper spectra, smaller RAA
13
From RHIC to LHC
RHIC LHC
RHIC: n ~ 8.2 LHC: n ~ 6.4
20.023.012.6
32.023.014.4
Remember: still ‘getting a sense for the numbers’; this is not a model!
With same medium density/energy loss, expect less suppression at LHC
Similar/slightly lower RAA indicates significant increase in medium density!
14
Towards a more complete picture
• Energy loss not single-valued, but a distribution
• Geometry: density profile; path length 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
Many model calculations exist:
some are more realiable/realistic than others!
15
RAA at LHC and JEWEL
Suppression factor 2-6
Significant pT-dependence
JEWEL energy loss model agrees with measurements
(tuned at RHIC, LHC ‘parameter-free’)
JEWEL: Monte Carlo event generator with radiative+collisional energy loss
- Modified showers with MC-LPM implementation
- Geometry: expanding Woods-Saxon density
16
Dihadron correlations
associated
D
trigger
8 < pTtrig < 15 GeV
pTassoc > 3 GeV
Use di-hadron correlations to probe the jet-structure in p+p, d+Au
Near side Away side
and Au+Au
Combinatorial background
17
Path length: ‘surface bias’
Near side trigger,
biases to small E-loss
Away-side large L
Away-side suppression IAA samples longer path-lengths
than inclusives RAA
NB: ‘surface bias’ is not a sharp effect; it’s a shifted distribution
18
Di-hadron modeling T. R
en
k, P
RC
, arX
iv:1
10
6.1
74
0
L2 (ASW) fits data
L3 (AdS) slightly below Modified shower
generates increase at low zT
L (YaJEM): too little suppresion
L2 (YaJEM-D) slightly above
Model ‘calibrated’ on single hadron RAA
19
Di-hadrons and single hadrons at LHC
Need simultaneous comparison to
several measurements
to constrain geometry and E-loss
Here: RAA and IAA
Three models:
ASW: radiative energy loss
YaJEM: medium-induced virtuality
YaJEM-D: YaJEM with L-dependent
virtuality cut-off (induces L2)
Note also: IAA > RAA
Compatible with expected
‘L2’ dependence
however, many details
20
Jets and parton energy loss
Motivation: understand parton energy loss by tracking the gluon radiation
Qualitatively two scenarios:
1) In-cone radiation: RAA = 1, change of fragmentation
2) Out-of-cone radiation: RAA < 1
21
Jets at LHC ALICE
h
Transverse energy map of 1 event
Clear peaks: jets of fragments
from high-energy quarks and gluons
And a lot of uncorrelated ‘soft’ background
22
Jet reconstruction algorithms
Two categories of jet algorithms:
• Sequential recombination kT, anti-kT, Durham
– Define distance measure, e.g. dij = min(pTi,pTj)*Rij
– Cluster closest
• Cone
– Draw Cone radius R around starting point
– Iterate until stable h,jet = <h,>particles
For a complete discussion, see: http://www.lpthe.jussieu.fr/~salam/teaching/PhD-courses.html
Sum particles inside jet Different prescriptions exist, most natural: E-scheme, sum 4-vectors
Jet is an object defined by jet algorithm
If parameters are right, may approximate parton
23
Collinear and infrared safety Illu
stra
tion b
y G
. Sa
lam
Jets should not be sensitive to soft effects
(hadronisation and E-loss)
- Collinear safe
- Infrared safe
24
Collinear safety
Note also: detector effects,
such as splitting clusters in calorimeter (0 decay)
Illustra
tion b
y G
. Sa
lam
25
Infrared safety
Infrared safety also implies robustness
against soft background in heavy ion collisions
Illustra
tion b
y G
. Sa
lam
26
Background jets Raw jet spectrum
Event-by-event background subtracted
Low pT: ‘combinatorial jets’ - Can be suppressed by requiring
leading track
- However: no strict distinction at
low pT possible
Next step: Correct for background
fluctuations and detector effects by
unfolding/deconvolution
27
Removing the combinatorial jets
Correct spectrum and remove combinatorial jets by unfolding
Results agree with biased jets: reliably recovers all jets and removed bkg
Raw jet spectrum Fully corrected jet spectrum
28
PbPb jet spectra Charged jets, R=0.3
Jet spectrum in Pb+Pb: charged particle jets
Two cone radii, 4 centralities
M. Verweij@HP, QM
RCP, charged jets, R=0.3
Jet reconstruction does not
‘recover’ much of the radiated energy
29
Pb+Pb jet RAA
Jet RAA measured by
ATLAS, ALICE, CMS
RAA < 1: not all produced jets are seen;
out-of-cone radiation and/or ‘absorption’
For jet energies up to ~250 GeV; energy loss is a very large effect
ATLAS+CMS: hadron+EM jets
ALICE: charged track jets
Good agreement
between experiments
Despite different methods:
30
Particles and jets model comparison
M. Verweij@HP, QM2012
JE
WE
L: K
. Zapp e
t al, E
ur P
hys J
C69, 6
17
U. Wiedemann@QM2012
Charged particle RAA
Jet RAA
Schukra
ft et a
l, arX
iv:1
202.3
233
At least one model calculation reproduces the observed suppression
Understand mechanism for out-of-cone radiation?
31
G. de Barros et al., arXiv:1208.1518
pT,jet< 20 GeV/c:
No change with trigger pT
Combinatorial background
Hadron-triggered recoil jet distributions
pT,jet> 20 GeV/c:
Evolves with trigger pT
Recoil jet spectrum
32
Remove background by
subtracting spectrum with
lower pTtrig:
Δrecoil =[(20-50)-(15-20)]
Reference spectrum (15-20) scaled by ~0.96 to account for
conservation of jet density
Background subtraction: Δrecoil
Unfolding correction for background fluctuations and detector response
Δrecoil measures the change of the recoil spectrum with pTtrig
33
Recoil jet yield ΔIAAPYTHIA ≈0.75, approx. constant with jet pT
R=0.4 Constituents: pT
const > 0.15 GeV/c no additional cuts (fragmentation bias) on recoil jets
pp reference: PYTHIA
(Perugia 2010)
Ratio of Recoil Jet Yield ΔIAAPYTHIA
34
Hadrons vs jets II: recoil
PR
L108 0
92301
Hadrons Jets
Hadron IAA = 0.5-0.6
In approx. agreement with models;
elastic E-loss would give larger IAA
Jet IAA = 0.7-0.8
Jet IAA > hadron IAA
Not unreasonable
NB/caveat: very different momentum scales !
35
Summary
• Jet quenching is a large effect
• Goal: understand energy loss mechanism and learn about medium density/geometry – Need close contact with theory
– Need good and controlled models
• Need multiple measurements to constrain models/theory – Inclusive and di-hadron measurements constrain path length
dependence (+geometry)
• Jets show large suppression (with limited cone size R=0.2, 0.3 in ALICE) – Indicates large out-of-cone radiation
Work in progress – plenty of room to get involved
36
Four formalisms
• Hard Thermal Loops (AMY)
– Dynamical (HTL) medium
– Single gluon spectrum: BDMPS-Z like path integral
– No vacuum radiation
• Multiple soft scattering (BDMPS-Z, ASW-MS)
– Static scattering centers
– Gaussian approximation for momentum kicks
– Full LPM interference and vacuum radiation
• Opacity expansion ((D)GLV, ASW-SH)
– Static scattering centers, Yukawa potential
– Expansion in opacity L/l (N=1, interference between two centers default)
– Interference with vacuum radiation
• Higher Twist (Guo, Wang, Majumder)
– Medium characterised by higher twist matrix elements
– Radiation kernel similar to GLV
– Vacuum radiation in DGLAP evolution
Multiple gluon emission
Fokker-Planck
rate equations
Poisson ansatz
(independent emission)
DGLAP
evolution
See also: arXiv:1106.1106
37
RAA at RHIC: ‘calibrating’ the models B
ass e
t al, P
RC
79
, 02
49
01
ASW:
HT:
AMY:
/fmGeV2010ˆ2q
/fmGeV5.43.2ˆ2q
/fmGeV4ˆ2q
Large density:
AMY: T ~ 400 MeV
Transverse kick: qL ~ 10-20 GeV
Large uncertainty in
absolute medium density
PH
EN
IX, a
rXiv
:12
08
.22
54
All calculations describe the data
Not very sensitive to
energy loss distribution
38
RAA at LHC & models
ALICE: arXiv:1208.2711
CMS: arXiv:1202.2554
Broad agreement
between models and
LHC RAA
Extrapolation from
RHIC
tends to give too much
suppression at LHC
Many model curves: need more constraints and/or selection of models
39
So what are we trying to do...
• First: understand (parton) energy process
– Magnitude, dominant mechanism(s)
– Probability distribution
• Lost energy
• Radiated gluons/fragments
– Path length dependence
– Flavour/mass dependence
– Large angle radiation?
• Goal: use this to learn about the medium
High-energy
parton
(from hard scattering)
Ha
dro
ns
40
pTa
sso
c > 3
GeV
p
Ta
sso
c > 6
GeV
d+Au Au+Au 20-40% Au+Au 0-5%
Suppression of away-side yield in Au+Au collisions: energy loss
High-pT hadron production in Au+Au dominated by (di-)jet fragmentation
Di-hadrons at high-pT: recoil suppression
41
Dihadron yield suppression
Away-side: Suppressed by factor 4-5
large energy loss
Near side Away side
STAR PRL 95, 152301
8 < pT,trig < 15 GeV
Yield of additional
particles in the jet
Yield in balancing
jet, after energy loss
Near side: No modification
Fragmentation outside medium?
Near side associated
trigger
Away side associated
trigger
42
R=0.4
Recoil Jet ΔIAAPYTHIA: R dependence
Similar ΔIAAPYTHIA for R=0.2 and R=0.4
R=0.2
No visible broadening within R=0.4
(within exp uncertainties)
43
Nuclear modification factor
ppTcoll
PbPbT
AAdpdNN
dpdNR
/
/
Suppression factor 2-6
Significant pT-dependence
Similar at RHIC and LHC?
So what does it mean?
PHENIX at RHIC ALICE at LHC
√sNN = 200 GeV √sNN = 2760 GeV
44
0 RAA – high-pT suppression
Hard partons lose energy in the hot matter
g: no interactions
Hadrons: energy loss
RAA = 1
RAA < 1
0: RAA ≈ 0.2
g: RAA = 1