1

Surface versus volume emissions in photon-hadron correlations

Han-Zhong Zhang

Institute of Particle Physics, Huazhong Normal University, China

Collaborators: J. Owens, E. Wang and X.-N. Wang

ATHIC 2008, Tsukuba Oct. 14, 2008

I. IntroductionII. Isolated photonIII. Gamma-tagged jet emissions in A+A collisionsIV. Conclusions

2

I. Introduction

Jet quenching:

The hard jet loses a significant amount of its energy

via gluon radiation induced by multiple scattering.

hadrons

q

q

hadrons

leadingparticle

leading particle

N-N collision

hadrons

q

q

hadrons

Leading particle suppressed

leading particle suppressed

A-A collision

X.-N.Wang and M.Gyulassy, Phys.Rev.Lett.68,1480(1992)

What happens for a jet propagating inside QGP?

3

Three kinds of hard probes of QGP

1) Single jet Single hadron spectra

2) Dijet Hadron-triggered away-side hadron spectra

3) Gamma-jet Photon-triggered away-side hadron spectra

Single jet Dijet Gamma-jet

?H.Z. Zhang, J.F. Owens, E. Wang and X.-N. Wang , PRL 98(2007)212301

Motivation

4

Gamma-jet by NLO pQCD parton model

LO : 22 tree level: NLO corrections: 22 one loop

+23 tree level

FFsdPDFsTd ABAA

GamT

JetT pp

GamTp

JetTp

GamT

JetT

GamT

JetT

GamT

JetT

pp

pp

pp

,

,GamTp

1JetTp

2JetTp

),( 21 JetT

JetT

JetT ppofOnep

therefore leading to hadrons with transverse

momentum larger than that of the photons

5

FFs modification due to jet quenching

The jet energy loss in a 1D expanding system:

),,(0

000

0

1

nrbd

dL

dEE g

L

dc

)/5.7/()6.1/( 02.1

001

EEdL

dE

d

Energy loss parameter

(a parameterization form of theory calculations) Enke Wang , X. -N. Wang , PRL87(2001)142301)

(X. -N. Wang , PRC70(2004)031901)

r

),()],(/),()[1(),,( 20/

/2'0/

'2'0

/

'/2

/ cch

Lggh

c

gcch

c

cLccch zDezD

z

zLzD

z

zeEzD

,//),/( ''cTgcTcTc EpLzEppz

With energy loss without loss

0000

/1),,,(0

0

nrb

dLg

L

The modified fragmentation functions:

6

II. Isolated photon

Direct photon sources in p+p:

1) Prompt photon from hard scattering

Annihilation Compton 23 tree level 22 one-loop Annihilation Compton

LO NLO

22 tree level

2) Fragmentation or e. g.

bremsstrahlung contributions

7J. F. Owens, Rev. Mod. Phys. 59, 465(1987);

H. Baer, J. Ohnemus, and J. F. Owens, Phys. Rev. D. 42, 61(1990)

Most accompanying hadrons arewithin a cone of half-angle coneR

Isolation Cut (IC)

Generally, the bremsstrahlung photons are always accompanied by nearly collinear hadrons on the same side.

22 )()( coneR

Jet

GammaAn “isolation” cut (IC) are often applied on the electromagnetic signal to separate the prompt photons from other sources.

8

NLO results for prompt photons in p+p

Because of IC selected at

RHIC, most fragmentation

contributions from parton

jets are taken out.

Data with IC:

1.0/.,5.0 TT

cone pEradR

PHENIX, PRL 98 (2007) 012002

“the measured photon samples … are expected to be isolated from parton jet activity.”

9

If we only consider the events where the photon has no nearly collinear hadrons accompanying on the same side, high p_T photon/photon-hadron will be dominated by annihilation and Compton processes.

Focus on isolated photons now

In order to give a simple and clear jet quenching picture in A+A, we will only consider such gamma-jet events in which the photons are isolated or only contributed by annihilation and Compton processes.

Theory prompt photons Ex. IC photons

10

Turbide, Gale, Jeon, Moore, Phys. Rev. C. 72 (2005) 014906

High p_T prompt photon dominates in central Au+Au

Other sources in AA:conversion photons;induced photons;thermal photons

With isolation cut applied, the “clean” photons are separated from the “unclean” EM signals in A+A collisions.

11

III. Gamma-tagged jet emissions

D_AA gives such a fragmentation function of the final parton jets, which is just weighted with the invariant cross sections of the correlated photons.

Sometimes we call D_AA as the photon-triggered hadron fragmentation functions

abABbaAB

abABbahAB

ThTT

TABT

hhTT

hABThT

T

hAB

ABTAB

dPDFsTdxrdxbddd

FFsdPDFsTdxrdxbddd

ppzwhere

dydpddydp

ddydydpdpdpddydydp

dz

dN

NzD

ˆ

,ˆ

,/

/

/1)(

22

22

The per-trigger photon-hadron spectra

12Data from “Matthew Nguyen for PHENIX, talk at QM2008”

dydp

dNdy

dydpdydp

dNdydpdy

T

pp

hhTT

hpp

hhT

Per-trigger yield as a

function of the p_T of

the triggered photon:

NLO pQCD results describe

the behavior of the data for

photon-hadron production

in p+p collisions.

Gamma-triggered hadrons in p+p:

13

Gamma-triggered hadrons in A+A:

Data from “A. Hamed for STAR, talk at QM2008 and HP2008”

Within the same energy loss formalism as that in our previous studies on single/dihadron spectra in A+A collisions.

Simultaneous fit of single, dihadron and photon-hadron productions.----Another evidence of jet quenching!

14

Qualitatively, Iaa in small z_Tregion is slightly more sensitiveto epsilon_0 than Iaa in large z_Tregion. LO

Per-trigger yield for photon-hadron in central Au+Au

NLO

T

hAA

AATAA

Tpp

TAATAA dz

dN

NzDwhere

zD

zDzI

1)(,

)(

)()(

Nuclear modification factor:

Energy loss parameter,which is introduced to describe the uncertainty of the medium density.

00

15

T

hTT ppz

NLO Nh > 0 at z_T>1: surface emissionnolossch

Llossch

LAAch DeDeD /

//

// )1(

At large z_T:the contributions with energy loss vanish due to jet quenching,dominated by the contributions without energy loss.

9.0Tz

0

For LO, the jet’s energy can’t exceed the gamma’s energy, no contributions for z_T>1 region.

For NLO, because of 2->3 processes, have z_T>1 contributions.

However 2->2 (tree level + one loop) dominate.

16

T

hTT ppz

For small z_T: Volume emission

nolossch

Llossch

LAAch DeDeD /

//

// )1(

At small z_T: both contribute.The jets near the center with energy loss dominate.

3.0Tz

17

The averaged distance <L> for the gamma-triggered parton jets passing through the quark matter.

Surface versus Volume emission

Small zt probes the matter deeper than large zt, so more sensitive.

Surface emission

Volume emission

3.0Tz

9.0Tz

18

Single hadron Dihadron Photon-hadron

More sensitive probe?

NLO

Comparing small-zt gamma-jets to single jets and dijets

19

Comparing small-zt gamma-jets to single jets

Gamma-jet Single jetsmall zT

Gam-jets for small zt probes the matter deeper than single jets.

3.0Tz 9.0Tz

20

Comparing small-zt gamma-jets to dijets

Gamma-jet Dijetsmall zT

Because of punch-through jets for dihadrons, it is not sure whether small-zt gam-jets are more sensitive than dijets.

3.0Tz

9.0Tz3.0Tz 9.0Tz

21

Hadron-triggered FFs are greater than gamma-triggered FFs

)()(

)()(

ThAuAuT

hhAuAu

ThppT

hhpp

zDzD

zDzD

e.g. Trig=8GeV, zt=0.5

hadr:8 jet:12 jet:12 assoc: 4

gamm:8 jet:8 assoc: 4

p+p: Per trigger

Why hadron-triggered FFs are greater than gamma-triggered FFs in pp/AA?

22

)()(

)()(

ThAuAuT

hhAuAu

ThppT

hhpp

zDzD

zDzD

e.g. Trig=8GeV, zt=0.5

hadr:8 12 12 assoc: 4

gamm:8 8 6 assoc: 4

Au+Au: Per trigger

Volume emission

Tangential ~pp

Why hadron-triggered FFs is greater than gamma-triggered FFs in pp/AA?

Hadron-triggered FFs are greater than gamma-triggered FFs

23

IV. Conclusions

1) With isolation cut applied, the “clean” photons are separated from the “unclean” EM signals in A+A collisions. With the “clean” photons tagged, the back-to-back jets are “clearly” stared for checking their energy loss.

2) Within the same energy loss formalism, our numerical results for photon-hadron fit data well, providing another evidence of jet quenching.

3) The suppression factor for hadrons with large z_T is controlled mainly by the surface emission of the gamma-jet events, while small z_T region will be volume emission bias.

4) Gamma-jets for small z_T region probe the dense matter deeper than those for large z_T region, so the gamma-jets for small z_T region are slightly more sensitive to the dense matter properties.

Thanks for your attention!