direct photon spectra and elliptic flow in 2.76 tev pb-pb...

Direct Photon Spectra and Elliptic Flow in 2.76 TeVPb-Pb Collisions from ALICE

Martin Wildeon behalf of the ALICE Collaboration

Westfälische Wilhelms-Universität Münster

December 5, 2012

Helmholtz Research School

H-QM Quark Matter Studies

Martin Wilde 12-5-12 p.1

Table of Contents

1 Direct Photon Production and the ALICE Detector

2 Part I: Direct Photon SpectraAnalysis StrategyDetection of Converted Photons and π0sInclusive Photon Results in pp and Pb-PbDecay Photon Background CalculationDirect Photon Results in pp and Pb-Pb

3 Part II: Direct Photon v2Analysis StrategyInclusive Photon v2 AnalysisDecay Photon v2Direct Photon v2


Direct Photons in pp and Pb-Pb Collisions

(a) Quark-gluon Comptonscattering

(b) Quark-Anti-quark annihilation

(c) Fragmentation photons(bremsstrahlung)

Direct Photons - DefinitionPhotons that are not produced by particle decays

Prompt Photons: In pp and Pb-Pb

Calculable within NLO pQCD

Predominant source in pp

Signal scales with numberof binary collisions in Pb-Pb

Fragmentation photons may be modified byparton energy loss in the medium

Measurement of direct photons in pp is an ideal test for pQCD


Direct Photons in Pb-Pb CollisionsAdditional sources of direct photons in Pb-Pb collisions

Jet-Medium Interactions:

Scattering of hard partons withthermalized partons

In medium (photon) bremsstrahlungemitted by quarks

Thermal Photons:

Scattering of thermalized particles

QGP: qq̄ → gγ and qg → qγ (+NLO)

HHG (hot hadronic gas): Hadronic interactions(e.g. π+π− → γρ0)

Exponentially decreasing but dominant at low pT

Photons leave medium unaffected, an ideal probe to study HI collisions

√sNN

= 5.5TeV,charge particle rapidity density: 3000

2006 J. Phys. G: Nucl. Part. Phys. 32 1295


The ALICE Detector and Data Sample

pp,√s = 7TeV:

Data sample:3.54× 108 events(min. bias)

Monte Carlo:Pythia-Perugia0 and Phojet

Pb-Pb, √sNN

= 2.76TeV:

Data sample: 17× 106 min. bias events

Monte Carlo: Hijing(min. bias plus enriched events with highp

Tπ0s)

Photons are measured via their conversionproducts in ITS and TPC

Martin Wilde Direct Photon Production and the ALICE Detector 12-5-12 p.5

Part I: Direct Photon Spectra

Martin Wilde Part I: Direct Photon Spectra 12-5-12 p.6

General Strategy of the Analysis

Subtraction Methodγdirect = γinc − γdecay = (1− γdecay

γinc) · γinc

Inclusive photons: measure all photons that are produced

Decay photons: calculated from measured particle spectra with photon decaybranches (π0, η, ...)

Double Ratioγincπ0 /

γdecayπ0

paramu γinc

γdecayif > 1 direct photon signal

→ advantage of ratio method: cancellation of uncertainties

Photons and π0s (and η) are measured via conversion methodπ0 → γγ, γ → e+e−


Photon Reconstruction with ITS and TPCSecondary Vertex Algorithm - V0 Particles

Charged tracks with large impactparameter are paired

Candidates with a small DCA→ V0 candidate

Most abundant particle species:K 0s , Λ, Λ̄ or γ

Photon conversion probability in|η| < 0.9 up to R = 180 cm at 8.5%

Cuts on the decay topology of photons and electron track properties→ Purity at 90% at 2GeV/c for 0-40% Pb-Pb events

Background is mainly combinatorial - Strange particle contribution negligible


Photon Corrections and Invariant Cross Section for pp

(GeV/c)T

p1 10

)3 c-2

(pb

GeV

3dp

σ3 dE

410

510

610

710

810

910

1010

1110

1210

3.5%± uncertainty ppσγInclusive

= 7 TeVspp,

Raw γ spectrum in pp and Pb-Pb corrected for:

purity (P)efficiency (E)conversion probability (C)

and secondary photon candidates subtracted

Inclusive photon cross section in pp: E d3σdp3 = 1

2πσMBOR

Nevents

1pT

PCE

Nγprim

∆y∆pT

Main sources of uncertainty:

Material budget of thedetector ∼ 4.5%

Efficiency estimation by cutvariationspT< 5GeV: pp ∼ 3%, Pb-Pb ∼ 6%

pT> 5GeV: pp ∼ 6%, Pb-Pb ∼ 15%

e.g. geometrical cuts, detectorPID, sharing of tracks betweensec. vertices


Inclusive Photon Invariant Yield in Pb-Pb

(GeV/c)T

p1 10

)2 c-2

(G

eVdy

Tdp

TpN2 d

ev

. Nπ21

-710

-610

-510

-410

-310

-210

-110

1

10

210

310

(0-40%)γInclusive (40-80%)γInclusive

= 2.76 TeVNNs Pb-Pb,

Two centrality selections: 0-40% and 40-80%(central and peripheral)


π0 and η Reconstruction via Conversion

)2

(GeV/cγγM0.1 0.15 0.2

2E

ve

nts

/2 M

eV

/c0

2

4

6

8

10

310×

PCM

<0.8 GeV/cT

: 0.6<p0π

)2 (GeV/cγγM0.4 0.5 0.6 0.7

2E

ve

nts

/8 M

eV

/c

0

2

4

6

310×

<1.8 GeV/cT

: 1.4<pη

Neutral pion and η (pp only) based on converted photons

Measurement based on identical set of photons as usedfor photon results

Inv. mass calculated for all photon pairs in an event

Combinatorial background obtained via mixed eventtechnique

Raw π0 spectrum obtained by peak integration

Efficiency and acceptance estimated with MCsimulations

For more details see:

pp at TeV: Phys. Lett. B 717, 162 (arXiv:1205.5724)

Pb-Pb and pp at 2.76TeV: published soon, similarmethod


Cocktail Generator

(GeV/c)T

p0 2 4 6 8 10 12 14 16

0 π /

deca

yγ

-510

-410

-310

-210

-110

1

10

210

310γall decay

)γ-e+ (eγγ → 0π)γγ0π,γ-e+,eγ-π+π (γγ → η

)γη (γ0π → ω)γγ, γω (γρ →’ η)γω, γ0π (γη → φ

)γη, γ0π (γ-π+π → ρ

= 7 TeVspp,

1/08/2012

Decay photon spectra are obtained via calculation

Based on a fit to measured π0 and η (in pp)

Other meson spectra obtained via mT-scaling

Incorporated mesons: π0, η, η′, ω, φ and ρ0

mT-Scaling:

Same shape of cross sections,f (m

T), of various mesons

E d3σm

dp3 = Cm · f (mT

)

Meson (Cm) Mass Decay Branch B. Ratioπ0 134.98 γγ 98.789%

e+e−γ 1.198%η 547.3 γγ 39.21%

π+π−γ 4.77%(0.48) e+e−γ 4.9 · 10−3

ρ0 770.0 π+π−γ 9.9 · 10−3

(1.0) π0γ 7.9 · 10−4

ω 781.9 π0γ 8.5%(0.9) ηγ 6.5 · 10−4

η′ 957.8 ρ0γ 30.2%ωγ 3.01%

(0.25) γγ 2.11%φ 1019.5 ηγ 1.3%

π0γ 1.25 · 10−3(0.35) ωγ < 5%

Phys. Rev. C (arXiv:1110.3929)


Direct Photons in pp Collisions at 7 TeV

(GeV/c)T

p0 2 4 6 8 10 12

)0 π/de

cay

γ)/

(0 π/

inc

γ(

0.8

1.0

1.2

1.4

1.6

1.8

2.0

Direct photon double ratio)

decayγ/

direct,NLOγNLO prediction: 1 + (

T = 0.5 to 2.0 pµfor

In the ratio uncertaintiesrelated to:

normalization

π0 measurement

rec. efficiency

partially or exactly canceled

The NLO double ratio prediction is plotted as RNLO = 1 +γdirect,NLO

γdecaycocktail

Measurement is consistent with the expected direct photon signal

Direct photon signal in pp at 7TeV is consistent with zero


Double Ratio - Pb-Pb 2.76 TeV - peripheral

(GeV/c)T

p0 1 2 3 4 5 6 7 8

)0 π/de

cay

γ)/

(0 π/

inc

γ(

0.8

1.0

1.2

1.4

1.6

1.8


decayγ/

direct,pp,NLOγ

collNLO prediction: 1 + (N


= 2.76 TeVNNs40-80% Pb-Pb,

Double ratio for peripheral events shows no excess at any value of pT

Measurement is consistent with the expected direct photon signal

pp NLO predictions scaled with Ncoll


Double Ratio - Pb-Pb 2.76 TeV - central

(GeV/c)T

p0 2 4 6 8 10 12 14

)0 π/de

cay

γ)/

(0 π/

inc

γ(

0.5

1.0

1.5

2.0

2.5

3.0


decayγ/

direct,pp,NLOγ



= 2.76 TeVNNs0-40% Pb-Pb,

Clear extra yield of 20% for pT< 2GeV/c

Ncoll scaled pp NLO in agreement with high pTdirect photons


Double Ratio - Pb-Pb 2.76 TeV - central

[GeV/c]T

p1 1.5 2 2.5 3 3.5 4 4.5 5

0.6

0.8

1

1.2

1.4

1.6

1.8

PHENIX Preliminary

=200GeVNN

sAu+Au

background

0π/γ / measured

0π/γRun4

*γ+1 from incl.γ/directγ

0 - 20%(150 MeV/c shifted)

0 - 10%

[GeV/c]T

p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

]-2

in

v. y

ield

[(G

eV

/c)

dir

ec

tγ

-610

-510

-410

-310

-210

-110

1

10

incl.*γ/direct

*γ • inclγ = directγ

AB T•pQCD

thermal

pQCD + thermal

Au+Au (0-20%)directγ

PHENIX Preliminary

a) b)

(GeV/c)T

p0 2 4 6 8 10 12 14

)0 π/de

cay

γ)/

(0 π/

inc

γ(

0.5

1.0

1.5

2.0

2.5

3.0


decayγ/

direct,pp,NLOγ




Clear extra yield of 20% for pT< 2GeV/c

Ncoll scaled pp NLO in agreement with high pTdirect photons

Similar to low pTdirect photon observation by PHENIX

arXiv:nucl-ex/0605005


Results of Pb-Pb Direct Photons at 2.76 TeV

(GeV/c)T

p0 2 4 6 8 10 12 14

)2 c-2

(G

eVdy

Tdp

TpN2 d

ev

. Nπ21

-710

-610

-510

-410

-310

-210

-110

1

10

210

310

Direct photons


Direct Photon Spectrumfor central Pb-Pb events

Spectrum derived fromdouble ratio by:

γdirect = (1− γdecay

γinc) · γinc

NLO predictions in agreement with spectrum (pT> 4GeV/c)

At low pT(< 2.2GeV/c) spectrum fitted with an exponential

→ slope parameter T = 304± 51stat+syst MeV

Intermediate region: superposition of low and high pTdirect photons

(GeV/c)T

p0 2 4 6 8 10 12 14

)2 c-2

(G

eVdy

Tdp

TpN2 d

ev

. Nπ21

-710

-610

-510

-410

-310

-210

-110

1

10

210

310


Direct photons (scaled pp)

T = 0.5 to 2.0 pµDirect photon NLO for

MeVstat+syst

51±/T), T = 304 T

exp(-p×Exponential fit: A


Conclusions I: Direct Photon Spectra

Statistical analysis of direct photons based on converted photons via doubleratio

With current uncertainties no significant direct photon signal in pp andperipheral Pb-Pb

Direct photon signal is consistent with expectation from NLO pQCD

In central Pb-Pb:Low p

Tdirect photon signal, exponential in shape

Similar excess measured at RHIC interpreted as thermal signal

Slope parameter:TALICE = 304± 51stat+syst MeV (0-40%)TPHENIX = 221± 19stat ± 19syst MeV (0-20%)arxiv:0804.4168 PRL 104 (132301) 2010


Part II: Direct Photon v2

Martin Wilde Part II: Direct Photon v2 12-5-12 p.19

What can we learn from direct photon v2?

0.0 1.0 2.0 3.0 4.0 5.0 6.0pT (GeV/c)

0.00

0.04

0.08

0.12

0.16

0.20

v 2(pT)

1.0 fm/c

0.4 fm/c0.2 fm/c

Au+Au@200 AGeV

0.6fm/c

0.8 fm/c

Thermal Photons

b = 6 fm

QM0.5xHMQM+HM

ideal hydro th. photon v2 for different QGP formation times τ0

arXiv:0809.0548 [nucl-th]

Initial azimuthal asymmetry in coordinate space in non-central A+A⇒ asymmetry in momentum space

dNdφ = 1

2π

(1 + 2

∑n≥1 vn cos(n(φ−ΨRP

n )))

v2: elliptic flow, collectiveexpansion at low pT

v2 at high pT : path lengthdependence of in-medium partonenergy loss

Thermal Photon v2

Constrains onset of direct photonproductionEarly production → small flowLate production → hadron-likeflow


General Strategy of the v2 Analysis

(GeV/c)T

p0 2 4 6 8 10 12 14

)0 π/de

cay

γ)/

(0 π/

inc

γ(

0.5

1.0

1.5

2.0

2.5

3.0


decayγ/

direct,pp,NLOγ




Direct photon v2 obtained via comparison between measured and calculateddecay photon v2

vdirect γ2 =

R·v inc γ2 −vdecay γ

2R−1

Factor R represents the direct photondouble ratio

R · v inc γ2 : weighted inclusive photon v2 due to extra photons compared to

background

vdecay γ2 : calculated decay photon v2 from cocktail calculation


Inclusive Photon v2 Analysis

Centrality (%)0 10 20 30 40 50 60 70 80 90

Re

so

lutio

n

0

0.2

0.4

0.6

0.8

17/07/2012

=2.76 TeVNN

sPbPb,

V0A 3subeventV0C 3subevent

ALI−PERF−31433

ALI-PERF-43616

v2 given by the reaction plane

v2 = 〈cos(2(φ−ΨRP2 ))〉

Extracted via this formula or by a fit

Relation of RP to EP:

v2 =vEP2

〈cos(2ΨEP2 −ΨRP

2 ))〉 =vraw2

resolution

Event Plane angle determined by usingthe VZERO detector

VZEROA: 2.8 < η < 5.1

VZEROC: −3.7 < η < −1.7

Reaction plane resolution obtained by thethree sub-event method


Inclusive Photon v2 Results 0-40%

ALI-PREL-43604 ALI-PREL-43612

Magnitude of v2 increases with decreasing centrality

Similar v2 to hadrons

Expected behavior, main contributions are decay photons


Cocktail Simulation and Decay Photon v2

ALI-PERF-43620

Spectra of other mesons withphoton decay branches obtainedby mT scaling

Assumption: vπ0

2 = vπ±

2

v2 of various mesons (X)calculated via KET (quarknumber) scaling from vπ±

2

Decay photon vX2 obtained by

cocktail calculation

vX2 (pXT ) = vπ±

2

(√(KEX

T+ mπ

±)2 − (mπ

±)2)

with:KET = mT − m =

√p2T

+ m2 − m


Comparison of Inclusive and Decay v2

ALI-PREL-43608

Above 3GeV/c inclusive photonssignificantly smaller than decayphotons

→ Direct photon v2 contributionwith vdirect

2 < v inc2

Below 3GeV/c consistent withinuncertainties

→ Either contribution of directphotons with similar v2 or nodirect photons


Direct Photon v2 0-40% and Conclusions II

ALI-PREL-43588

Significant direct photon v2 forpT < 3GeV/c measured

Magnitude of v2 comparable tohadrons

Result points to late productiontimes of direct photons after flowis established

Large inverse slope parameter oflow pT direct photon spectrumfavours earlier production times

Similar direct photon v2 resultsseen by PHENIX


Backup Slides


Denominator Ratio: Cocktail Generator Pb-Pb Results

(GeV/c)T

p0 2 4 6 8 10 12 14

)2 c-2

(G

eVdy

Tdp

TpN2 d

ev

. Nπ21

-1310

-1110

-910

-710

-510

-310

-110

10

310

510

γall decay )γ-e+ (eγγ → 0π

)γγ0π,γ-e+,eγ-π+π (γγ → η)γη (γ0π → ω

)γγ, γω (γρ →’ η)γω, γ0π (γη → φ


= 2.76 TeVNNs0-40% Pb-Pb, 26/07/2012

(GeV/c)T

p0 2 4 6 8 10 12 14

0 π /

deca

yγ

-510

-410

-310

-210

-110

1

10

210

310γall decay





26/07/2012

(GeV/c)T

p0 2 4 6 8 10 12 14

)2 c-2

(G

eVdy

Tdp

TpN2 d

ev

. Nπ21

-1310

-1110

-910

-710

-510

-310

-110

10

310

510

γall decay )γ-e+ (eγγ → 0π

)γγ0π,γ-e+,eγ-π+π (γγ → η)γη (γ0π → ω

)γγ, γω (γρ →’ η)γω, γ0π (γη → φ


= 2.76 TeVNNs40-80% Pb-Pb, 26/07/2012

(GeV/c)T

p0 2 4 6 8 10 12 14

0 π /

deca

yγ

-510

-410

-310

-210

-110

1

10

210

310γall decay




= 2.76 TeVNNs40-80% Pb-Pb,

26/07/2012


Combined Fit for Direct Photons

(GeV/c)T

p0 2 4 6 8 10 12 14

)2 c-2

(G

eVdy

Tdp

TpN2 d

ev

. Nπ21

-710

-610

-510

-410

-310

-210

-110

1

10

210

310

Direct photons

Hagedorn

Exponential

Hagedorn+Exponential

/T)T

+ B exp(-pn/nC)T

(1+p1 π2

A = dy

Tdp

Tp

N 2d π21

51 MeV±T = 299


Combined fit (Hagedorn + Exponential) gives similar result for the inverseslope parameter T as for the exponential only fit


Systematic Cut Studies ppCut Variations for γ and π0:

Cut Name Std. value Variation 1 Variation 2 Variation 3

Electron dEdx -4,5σ -4,4σ -3,4σ -Pion dEdx 1,-10σ 2,1σ 2,0.5σ 2,0.5σ

Min. p e+/e− 0.4 GeV/c 0.4 GeV/c 0.4 GeV/c 0.3 GeV/cFind. Cls. TPC 0.35 0.6 - -Photon χ2 20 30 10 -

qt 0.05 0.07 0.03 -min. pt e+/e− 50MeV/c 75MeV/c 100MeV/c -photon η, π0 y 0.9, 0.8 0.8, 0.7 1.2, 0.9 -

min. R 5 cm - 180 cm 2.8 cm - 180 cm 10 cm - 180 cm -

V0s with shared electrons rejectedPurity for different centralities usedTOF and α cut not used for ppR cut already considered for material budget

π0 yield extraction:

Three different integration windowsDifferent Numbers of mixed events for bg, different mixed event bins(n V0s, n tracks)

Cocktail simulation:

Two different fitsVariation of the mt scaling factors (η measured)


Systematic Cut Studies Pb-PbCut Variations for γ and π0:

Cut Name Std. value Variation 1 Variation 2 Variation 3

Electron dEdx -3,5σ -4,5σ -2.5,4σ -Pion dEdx 3,-10σ 2.5,-10σ 3.5,-10σ 3,-10σ

Min. p e+/e− 0.4 GeV/c 0.4 GeV/c 0.4 GeV/c 0.3 GeV/cFind. Cls. TPC 0.6 0.7 0.35 -Photon χ2 10 5 20 -

qt 0.05 0.03 0.07 -min. pt e+/e− 50MeV/c 75MeV/c 100MeV/c -photon η, π0 y 0.75, 0.7 0.9, 0.8 0.8, 0.7 -

min. R 5 cm - 180 cm 2.8 cm - 180 cm 10 cm - 180 cm -α meson central 0.65 1.00 - -α meson peripheral 0.8 1.00 - -

TOF -5,-5σ -3,-5σ -2,-5σ -

V0s with shared electrons rejectedPurity for different centralities used

π0 yield extraction:

Three different integration windowsDifferent Numbers of mixed events for bg, different mixed event bins(n V0s, n tracks)

Cocktail simulation:

Two different fits, with and without blast waveVariation of the mt scaling factors


PHENIX Direct Photon v2 Results

0 2 4 6 8 10 12

-0.05

0

0.05

0.1

0.15

0.2

0.25

(c)

2 vdir.γ

0 2 4 6 8 10 12

-0.05

0

0.05

0.1

0.15

0.2

0.25

(b)

2 vinc.γ

|=1.0~2.8)η (|RXNE.P.

|=3.1~3.9)η (|BBC

E.P.

0 2 4 6 8 10 12

-0.05

0

0.05

0.1

0.15

0.2

0.25

(a)

2 v0π

2v

[GeV/c]T

p


direct photon spectra and elliptic flow in 2.76 tev pb-pb...

Documents