GAMMA-HADRON CORRELATION
MEASUREMENT WITH ALICE AT LHC
Yaxian Mao, Daicui Zhou, Yves Schutz
In ALICE Physics Workgroup: High pT and photons
( for ALICE collaboration -- Wuhan)
MYX@Nanjing
28 April 2008
0
Why -jet/hadronp correlation?
0
Jet
Fragmentation 
Prompt 
 The photon 4-momentum remains unchanged by the medium and
sets the reference of the hard process
 Balancing the jet and the photon provides a measurement of the
medium modification experienced by the jet
 Allows to measure jets in an energy domain (E < 50 GeV) where

The jet looses a large fraction of its energy (DE ≈ 20 GeV)
 The jet cannot be reconstructed in the AA environment
MYX@Nanjing
28 April 2008
1
Photon sources
 Direct photons (the signal)
 Prompt pQCD photons (E > 10 GeV)
 g Compton scattering
 qq annihilation
 Fragmentation
q
g
q
g
g
q
q
γ
g
q
q
q
g
γ
γ
 Photons produced by the medium (Eγ <
q
10 GeV)
γ  Bremsstrahlung
g
q
γ  Jet conversion
 Thermal
q
MYX@Nanjing
28 April 2008
2
 Decay photons
(the background)
Rate
Photon sources
Hadron Decay
QGP Thermal
 Hadrons, mainly π0
 But suppressed by
the medium
Jet conversion
Bremsstrahlung
pQCD prompt
E
10GeV
MYX@Nanjing
28 April 2008
3
High pT hadrons suppression
1
R AA 
N AA
d 2
d 2
AA
dpT dy
pp
dpT dy
• Hard scattered partons interact with the color dense
medium
• The energy loss is imprinted in the fragmentation hadrons
• The medium is transparent to photons
MYX@Nanjing
28 April 2008
4
From RHIC to LHC
 The medium is formed
with energy densities
larger by a factor 2-3: a
different QGP ?
 The lifetime of the QGP
is increased by a factor
2-3: more favorable for
observation
MYX@Nanjing
28 April 2008
5
From RHIC to LHC
 Cross section of hard
probes increased by
large factors
 105 for very high pT jets
 Differential
measurements become
possible
 Jet fragmentation
function
 Photon tagging
MYX@Nanjing
28 April 2008
6
Measurement with ALICE@LHC
h
Charged hadrons:
TPC
D =360º || < 0.9
Dp/p < 5% at E < 100GeV
Photons:
PHOS

D =100º || < 0.12
DE/E = 3%/E
MYX@Nanjing
28 April 2008
7
Strategy for a feasibility study
 Identify prompt photons with ALICE PHOS detector (PID
+ Isolation Cut)
 Construct -charged hadrons correlation from detected
events (detector response)
 Compare the imbalance distribution (CF) and
fragmentation function (FF)
 Do the same study in -jet events (signal) and jet-jet
events (background).
 Estimate the contribution of hadrons from underlying
events.
 Start with pp, base line measurement for AA
measurement
MYX@Nanjing
28 April 2008
8
Monte Carlo Data production
 +jet in final state
  – jet @ √s = 14 TeV
 Prompt  is the signal under study: 6×105 events (5 GeV <
E  < 100 GeV)
 2 jets in final state
 jet – jet @ √s = 14TeV
 These events constitute the background: high-pT p0 [O(S)]
and fragmentation [O(2S)]: 24×105 events (5 GeV < E jet <
200 GeV)
 ALICE offline framework AliRoot
 Generator: PYTHIA 6.214; PDF: CTEQ4L
 Luminosity: Lint = 10 pb -1
 Acceptance: two PHOS modules
 [-0.13, 0.13];  = [259, 301]
MYX@Nanjing
28 April 2008
9
Generated prompt photon from
-jet events
-jet in pp@14TeV
 Cross section of generated photons from -jet events in pp@14TeV
 Dashed lines are the simulation bins
MYX@Nanjing
28 April 2008
10
Generated decay photon from
jet-jet events
jet-jet in pp@14TeV
(π0→γ+γ)
 Cross section of generated photons from jet-jet events in
pp@14TeV
 Dashed lines are the simulation bins
MYX@Nanjing
28 April 2008
11
Photon identification (PID)
 We can discriminate , e and p0 from anything
else : based on:
 CPV : Charged particle identification
 TOF : Identification of massive low pT particles
 EMCA : Hadron rejection via shower topology (SSA)
MYX@Nanjing
28 April 2008
12
Photon PID Efficiency
 discriminate  and p0 (SSA) 30 GeV < E < 100 GeV
 High  identification efficiency, ~ 70%,
 Misidentification efficiency decreasing from 70% to 20%
true 70%
false 20%
 Not good enough
MYX@Nanjing
28 April 2008
13
Isolation cut (IC)
 Prompt  are likely to be produced
isolated
2
2
R

D


D

 Cone size
 pT threshold candidate isolated if:
 no particle in cone with pT > pT thres
 pT sum in cone, ΣpT < ΣpTthres

IP

TPC

pp collisions; R = 0.3, Σp thres = 2.0 GeV/c
T


Identification probability 98 %
Misidentification
3%

Signal/Background
MYX@Nanjing
R
candidate
PHOS
>10
28 April 2008
14
Isolation cut (IC)
 S/B:
 ~ 0.07 at pT =20 GeV/c (generated events)
 ~ 0.1 at pT =20 GeV/c (detected events)
 > 10 at pT = 20 GeV/c (after IC selection)
MYX@Nanjing
28 April 2008
15
Photon spectrum after one year
data taking
 Estimated counting statistics in one pp run for 2 PHOS
modules
 Systematic errors from misidentified p0
MYX@Nanjing
28 April 2008
16
Jet
Fragmentation function
 Construct jets within jet finder in PYTHIA;
 Calculate fragmentation function of these jets:
the distribution of charged hadrons as a function
of the fraction of jet momentum z = pT/ETjet
Jet
(0, 0)
R
R = (-0)2 + (-0)2 =1
IP
 Requirement: reconstruction of jet energies
MYX@Nanjing
28 April 2008
17
Tagging jet with photon
 Search identified prompt photon (PHOS) with
min
largest pT (E  > 20 GeV)
max
EMCal
 Search leading particle :
R
 -leading180º
leading
 Eleading > 0.1 E
 Reconstruct the jet:
TPC
 Particles around the leading inside a cone
 A more simpler method is …
MYX@Nanjing
IP

PHOS
28 April 2008
18
-hadron correlation
 Momentum imbalance variable
 z-h= -pT · pT / |pT|2
h
 In leading-order kinematics (s)
 z-h pT / pT
h
 According momentum conservation,
 pT = k = Eparton
 Therefore,
 (exp.) z-h z (th.)
MYX@Nanjing
28 April 2008
19
Kinematics conditions on CF
 Photon and hadron momenta cuts must be very
asymmetric:
pTcut >> pT hcut
 Photon must be produced directly from the
partonic process and not from a jet
fragmentation:
isolated and pT> 20GeV/c
 Photon – hadrons are back to back:
p/2 < D < 3p/2
MYX@Nanjing
28 April 2008
20
Experimental imbalance distribution
 Statistical errors correspond to one standard year of data taking with 2
PHOS modules.
 Systematic errors is contributed by decay photon contamination and
hadrons from underlying events.
 Imbalance distribution is equivalent to fragmentation function for z = 0.1
– 0.8
MYX@Nanjing
28 April 2008
21
Summary
 Tagging jets with direct prompt photons is the
unique approach to identify low energy jets (Ejet
< 50 GeV) in AA
 The medium modification on jets is best
measured in the jet fragmentation function
 The fragmentation function can be measured in
photon – charged hadrons correlations
 The feasibility of such a measurement with the
ALICE experiment has been evaluated in pp at 14
TeV
 Next time , different kinematics cuts and AA …
MYX@Nanjing
28 April 2008
22
Thanks for your attention!
MYX@Nanjing
28 April 2008
23
Back up
Hard Probes: an essential tool
 Hard probes involve high momentum (pT) or high
mass transfer:
 pT, M >> QCD : pertubative regime of QCD thus
calculable
 1/(pT, M) << tQGP formation: produced in early phase of
the collision
 pT, M >> T medium : they decouple from the medium
 Observe how the medium (AA) modifies the hard
probes as compared to its vacuum (pp)
properties
MYX@Nanjing
28 April 2008
25
Di-hadron correlation
Jet
Jet
• No back-to-back high pT correlation in central Au+Au
collisions compared to dAu or pp collisions: the hard
scattered parton looses energy via gluon radiation
when traversing the color dense medium
MYX@Nanjing
28 April 2008
26
Di-hadron correlation
Jet
Jet
• Back-to-back low pT correlation reflects the radiated
energy through the fragmentation of low pT gluons
MYX@Nanjing
28 April 2008
27
Performance
Central Barrel, pp
PHOS
• Momentum resolution in central barrel better than 4%
• Energy resolution in PHOS better than 1.5% for E > 10 GeV
MYX@Nanjing
28 April 2008
28
Imbalance distribution from
NLO (by F. Arleo)
 Within higher kinematics condition, the medium effects can be
measured by imbalance distribution (CF) instead of fragmentation
function (FF).
MYX@Nanjing
28 April 2008
29
Underlying events estimation
 Based on:
 Hadrons spatial distribution from underlying events (ue) is
isotropic:
ue (|-hadron |<0.5 p) ≅ ue (0.5p <|-hadron |<1.5 p)
 Strategy:
 Calculate ue contribution on the same side as photon
where there is no jet contribution
MYX@Nanjing
28 April 2008
30