MeasurementOf)DirectPhoton)Higher)Order)) Azimuthal ...utkhii.px.tsukuba.ac.jp/report/2014/Mizuno_D/Predefence_photon_vn_Sanshiro.pdfSanshiro)Mizuno) High)Energy)Nuclear)Physics)Group)

Sanshiro Mizuno High Energy Nuclear Physics Group

Pre-‐defense Session Jan. 23th 2015

Measurement Of Direct Photon Higher Order Azimuthal Anisotropy

In √sNN=200GeV Au+Au Collisions at RHIC-‐PHENIX

PreDefence (M.Sanshiro) 2

Outline

ü Introduc9on •  Direct photon •  MoOvaOon

ü Analysis •  PHENIX detector •  Analysis method

ü Results and Discussion •  Neutral pion vn •  Direct photon vn •  The probability of resolving photon puzzle

ü Summary

2015/1/23

Introduc9on PreDefence (M.Sanshiro) 3 2015/1/23


Quark-‐Gluon Plasma at heavy ion collision

LaTce-‐QCD calculaOon predicts ε ≈ 1GeV/fm3

T ≈ 170MeV

Quarks and gluons are predicted to move freely in the state of extremely high temperature and dense ma[er.

QGP has been created at high energy heavy ion collision. •  RHIC at BNL •  LHC at CERN

P.R.D 80, 014504(2009)

RHIC

2015/1/23


History of collision and photon emission

ThermalizaOon Hydrodynamics Freeze-‐out HadronizaOon

Direct photon analysis All photons except for those originaOng from hadron decays. •  emi[ed during all stages of the collisions •  don’t interact with the medium

QGP Hadron Gas (HG) Collision

Hadron Photon

Hard sca[ering

2015/1/23

(GeV/c)Tp0 2 4

γR

1

1.2

1.4

2015/1/23 PreDefence (M.Sanshiro) 6

(GeV/c)T

p0 5 10

γR

1

2

30-20 %

CalorimeterExternal conversionvirtual photon

The excess of direct photon

P.R.L. 94, 232301 arXiv:1405.3940 P.R.L. 104, 132301

R� = Ninc./Ndec.

The excess of direct photon has been measured via several methods. The virtual photon method and external conversion photon method are sensiOve to low pT region. Less than 4 GeV/c, direct photons are included by 20 % in inclusive photon.


Direct photon pT spectra

arXiv:1405.3940

The pT spectra in Au+Au collision is enhanced compared with that in p+p collision scaled by the number of binary collisions less than 4 GeV/c.

The excess of pT spectra is fi[ed and effecOve temperature is extracted. (Freeze out temperature ≈ 100MeV)

Photons in low pT are mainly radiated from very hot medium at early 9me of collisions.

Au+Au Centrality Effec9ve temperature

0% -‐ 20% 239 ± 25 ± 7 (MeV)

20% -‐ 40% 260 ± 33 ± 8 (MeV)

40% -‐ 60% 225 ± 28 ± 6 (MeV)

arXiv:1405.3940(2014)

2015/1/23

8

Azimuthal anisotropy (Ellip9c flow)

PreDefence (M.Sanshiro)

) R.P.

Ψ -φ=(φd-1 0 1

)R

.P.

Ψ -φd(

dN

210

310

v2

Reac9on Plane

charged parOcle dφ distribuOon

•  anisotropic pressure gradient in parOcipant zone (IniOal state) •  QGP expansion (hydrodynamic moOon, η/s)

(η is shear viscosity and s is entropy density) •  hadron producOon mechanism (coalescence)

(1) : Ini9al geometry is converted into final azimuthal anisotropy (2) : (expected to be) sensiOve to η/s.

v2 = hcos {2(�� R.P.)}i

2015/1/23


Photon emiang angle dependence

Parton Photon Par9cipant zone Angular dependence of photon sources •  IniOal hard sca[ering : v2≈0 •  Medium induced : v2≤0 •  Jet fragmentaOon : v2≥0 •  RadiaOon from expanding medium : v2>0

The measurement of photon azimuthal anisotropy is a powerful probe to idenOfy the photon sources.

2015/1/23


P.R.L. 109, 122302(2012)

High pT : very small v2 It is consistent with expectaOon that photons produced in the iniOal hard sca[ering are dominant.

Low pT : Comparable to hadron v2 at around 2 GeV/c

Ellip9c flow of direct photon

2015/1/23


Direct photon puzzle Thermal radiaOon photons are dominant in low pT region. Ellip9c flow : Photon should have small v2, since it includes one from early stage having small v2,

-‐> Photons are dominantly emi[ed at late stage. pT spectra : Emi[ed from very hot medium (Teff≈240MeV).

-‐> Photons are dominantly emi[ed at early stage. There is a discrepancy, and it is called “direct photon puzzle”. There is no models to explain both observables simultaneously.

2015/1/23

Ψ3

Ψ2

Ψn : Par9cipant Plane

Third order azimuthal anisotropy (v3)

PreDefence (M.Sanshiro)

v3 is originaOng from fluctuaOon of the shape of parOcipants. The higher order flow is expected to constrain the iniOal geometry calculaOng model and η/s of QGP.

12 2015/1/23

N(�� n) / 1 + 2

Xvn cos {n(�� n)}

vn = hcos {n(�� n)}i


Why direct photon v3 is measured?

B

Radial flow effect (blue shij effect) : It makes effecOve temperature higher than true temperature. Photons from late state are dominant. v2>0 : v3>0 Large magne9c field : DirecOon of magneOc field is strongly related with Ψ2(R.P.) but not with Ψ3. v2>0 : v3≈0 v3 measurement could provide addiOonal constraint on photon producOon mechanism.

P.R.C 89, 044910 (2014)

Effec9v

e Tempe

rature

2015/1/23

True Temperature


My ac9vity

2014 (D3):Data taking shil and Detector expert

2013 (D2):Data taking shil and Detector expert & TOF calibraOon

2012 (D1):Data taking shil and Detector expert & TOF calibraOon Talk ATHIC 2012

Talk ATHIC 2014

Talk HIC HIP

Talk JPS spring

Talk JPS fall

Talk JPS spring

Poster QM 2012

Talk QM 2014

Talk WPCF 2014

Iden9fied par9cle azimuthal anisotropy

Neutral pion and direct photon azimuthal anisotropy

Poster & Talk : Analysis

2015/1/23

Analysis PreDefence (M.Sanshiro) 15 2015/1/23


PHENIX detector CNT |η| <0.35

1.0< |η| <2.8

ReacOon Plane detector (RxNP), MPC, BBC are used for measuring Event Plane. Photons and π0 are detected by EMCal in CNT.

vn = hcos {n(�� n)}i2015/1/23


Centrality determina9on

Centrality : The size of parOcipant zone is classified by mulOplicity in BBC. Beam-‐Beam Counter (BBC) : Measures charged parOcles.

BBC charge sum 0 500 1000 1500 2000

Cou

nt

10

210

0 % 100 %

2015/1/23


Event plane and resolu9on

n =

1

ntan

�1⇣P

wi sinn�iPwi cosn�i

⌘

Event plane is determined for each harmonic “n”.

Reac9on Plane detector(RxN)

Muon Piston Calorimeter (MPC)

RxN(In) RxN(Out) RxN(I+O)

MPC BBC RxN(In)+MPC

vtruen

= vobs.n

/Res( n

)

Res( n

) = hcos {n( true

n

� obs.

n

)}i

2015/1/23

)2invariant mass(GeV/c0 0.1 0.2 0.3

Cou

nt

0

500

1000

1500310×

Real eventMixed eventExtracted signal

20-30 %<2.5(GeV/c)T2.0<pγ + γ → 0π

<0.18γγ0.10<m


Photon reconstruc9on Pad chamber : space point of charged parOcle track Electromagne9c calorimeter(EMCal) Photons are reconstructed •  Energy threshold : E > 0.2GeV •  Shower shape : χ2 < 3 •  Charged parOcle rejecOon at PC3 : √(dz)2+(rTsin(dφ))2 > 6.5 π0 (-‐>γ+γ) reconstrucOon •  Asymmetry cut : |E1-‐E2|/(E1+E2)<0.8 •  Photons are detected in same sector •  Invariant mass of γ+γ

Mass =p

2E1E2(1� cos ✓)

2015/1/23

) 2

Ψ - φ | ( = φ Δ| 0 0.5 1 1.5

Cou

nt

4000

5000

6000

310×

20-30 %<2.5(GeV/c)T2.0<p

])}nΨ -φcos{n(n[2v0N

(GeV/c)T

p0 5 10 15

nv

-0.05

0

0.05

0.1

0-20 %RxN(I+O)

2 vinc.γ

3 vinc.γ


Inclusive photon vn measurement

The method of extracOng vn 1. vn=<cos{n(φ-‐Ψn)}> 2. dN/dφ is fi[ed by N0(1+2vncos{n(φ-‐Ψn)}) Systema9c uncertainty •  Photon selecOon •  vn measuring method •  Event plane determinaOon

2015/1/23

) 2

Ψ - φ | ( = φ Δ| 0 0.5 1 1.5

Cou

nt

2000

2500

3000

310×

20-30 %<2.5(GeV/c)T2.0<p

])}nΨ -φcos{n(n[1+2v0N

(GeV/c)T

p0 5 10 15

nv

-0.05

0

0.05

0.1

0-20 %RxN(I+O)

2 v0π

3 v0π


Neutral pion vn measurement

The number of π0 is counted with respect to the event plane. The azimuthal distribuOon is fi[ed by the equaOon. Systema9c uncertainty •  Photon selecOon •  π0 selecOon •  Event plane determinaOon

N0(1 + 2vn cos {n(�� n)})

2015/1/23


Hadronic decay photon It is impossible to idenOfy photons originaOng from hadron decay experimentally. Decay photon pT spectra and vn should be simulated by Monte-‐Carlo simulaOon. Par9cle Data Group

2015/1/23

(GeV/c) T

p0 5 10 15 20

Tdy

dpdNT

pπ21

-1210

-910

-610

-310

1

310410

0-20 % PHENIX results

Estimated spectra

1.0× 0π 1.0× ±π

0.1× η 0.01× ω

1.0×pion 0.1× η 0.01× ω 0.001× ρ 0.0001×' η

(GeV/c) T

p0 5 10 15 20

Dat

a/Eq

uatio

n

0.5

1

1.5

2 0π±πη

ω


Meson pT spectra es9ma9on

Since it is difficult to measure mesons except for pion, the other mesons pT spectra are esOmated by mT scaling from pion experimental data.

pT,meson

=q

p2T,pion

+M2meson

�M2pion

P.R.C 69,034909 P.R.L. 101,232301 P.R.C 82,011902 P.R.C 84,044902

2015/1/23

(GeV/c) T

p0 5 10 15

2v

0

0.05

0.12v

0-20 %RxN(I+O)

pionη

ω

ρ

'η

(GeV/c) T

p0 5 10 15

3v

0

0.05

0.13v

0-20 %RxN(I+O)

pion

η

ω

ρ

'η


Meson vn es9ma9on The number of cons9tuent scaling

It is known that hadron vn as a funcOon of KET are scaled by the number of consOtuent quark. Meson vn is esOmated from pion vn.

pT,meson

=

r⇣qp2T,⇡

+M2⇡

�M⇡

+Mmeson

⌘2�M2

meson

arXiv:1412:1038

2015/1/23

(GeV/c) Tp0 5 10 15

Tdy

dpdNT

pπ21

-510

-210

10

410

710

10100-20 %decay photon

0π from dec.γη from dec.γω from dec.γρ from dec.γ

'η from dec.γdec.γall

(GeV/c) Tp0 5 10 15

) ico

ntrib

utio

n ra

tio (

R

-410

-310

-210

-110

1

(GeV/c) Tp0 5 10 15

2v

0

0.05

0.1

0.152

decay photon v


Hadronic decay photon vn

Decay photon vn is simulated from meson input. Systema9c uncertainty •  Propagated from pion pT spectra •  Propagated from pion vn •  Event plane determinaOon

vdec.n =X

i

Rivdec.,in

2015/1/23

(GeV/c)Tp0 2 4

γR

1

1.2

1.4

(GeV/c)T

p0 5 10

γR

1

2

30-20 %

CalorimeterExternal conversionvirtual photon

(GeV/c)T

p0 5 10

2v

-0.05

0

0.05

0.1 2v

0-20 %RxN(I+O)

inclusive photondecay photon

(GeV/c)T

p0 5 10 3v

-0.05

0

0.05

0.1 3v

0-20 %RxN(I+O)

inclusive photondecay photon


Direct photon vn

⌫dir.n =R�⌫inc.n � ⌫dec.n

R� � 1

P.R.L. 94, 232301 arXiv:1405.3940 P.R.L. 104, 132301

Systema9c uncertainty •  Propagated from inclusive photon vn •  Propagated from decay photon vn •  Propagated from Rγ •  Event plane determinaOon

R� = Ninc./Ndec.

2015/1/23

Results & Discussion Neutral pion vn


|η|

Electromagne9c calorimeter (CNT) |η| < 0.35 Reac9on Plane detector (Inner) 1.5 < |η| < 2.8 Reac9on Plane detector (Outer) 1.0 < |η| < 1.5 MPC 3.1 < |η| < 3.8 BBC 3.0 < |η| < 3.9

2015/1/23

(GeV/c)T

p0 5 10

2v

0

0.05

0.1

0.15

0.2

0-10 %

2Neutral pion v

(GeV/c)T

p0 5 10

2v0

0.05

0.1

0.15

0.2

40-50 %

(GeV/c)T

p0 5 10

3v

-0.4

-0.2

0

0-10 %

3Neutral pion v

|<3.8ηRxN(In)+MPC:1.5<||<1.5ηRxN(Out):1.0<|

(GeV/c)T

p0 5 10

3v

-0.4

-0.2

0

40-50 %

(GeV/c)T

p0 5 10

4v

0

0.1

0.2

0.3

0-10 %

4Neutral pion v

(GeV/c)T

p0 5 10

4v

0

0.1

0.2

0.3

40-50 %


The event plane dependence of neutral pion vn

Hadrons in high pT are originated from jet fragmentaOon. •  Study jet property In peripheral event v2 and v4 : posiOve v3 : negaOve In the v2 case, there is the difference blue and red

2015/1/23

〉 part N〈0 100 200 300

2in

tegr

ated

v

-0.2

0

0.2

0.4 2Neutral pion v

< 15 GeV/cT

6 < p

|<3.8ηRxN(In)+MPC:1.5<|

|<1.5ηRxN(Out):1.0<|

〉 part N〈0 100 200 300

3in

tegr

ated

v

-0.2

0

0.2

0.4 3Neutral pion v

< 15 GeV/cT

6 < p

〉 part N〈0 100 200 300

4in

tegr

ated

v

-0.2

0

0.2

0.4

4Neutral pion v

< 15 GeV/cT

6 < p


Event plane dependence of π0 vn in high pT

Di-‐jet makes v2,4 >0 and v3≈0. Energy loss of path length dependence : vn > 0 It can be understood in central event. NegaOve v3 and the difference of red and blue in v2 are not understood.

2015/1/23


Jet effect in event plane determina9on

⌘ = � lnn

tan✓

z

o

EMCal : |η|<0.35 RxN(Out) : 1.0<|η|<1.5 RxN(In) : 1.5<|η|<2.8 MPC : 3.1<|η|<3.8 Jet

Event plane determinaOon is much affected by jet effect in peripheral event due to small mulOplicity. The trend is different in v2 & v4 and v3 due to the iniOal geometry.

v2&v4 > 0 : v3 < 0

z

2015/1/23

θ

0 200 400 600 800 1000 1200 14001

10

210

310

410

510


a Mul9phase transport model (AMPT) event generator (HIJING) + parton cascade (ZPC) + hadronizaOon + hadron cascade P.R.C 72, 064901 3 M events including Jet > 20 GeV are analyzed. Centrality is defined by the number of parOcles in 3.1<|η|<3.9. Pions are detected in |η|<0.35. Event Plane is measured in 1 : 1.5<|η|<2.8 (RxN(In)) 2: 1.0<|η|<1.5 (RxN(Out)) 3: 1.0<|η|<2.8 (RxN(I+O)) 4: 3.1<|η|<3.8 (MPC) 5: 3.1<|η|<3.9 (BBC) 6: 1.0<|η|<1.5 + 3.1<|η|<3.9 (RxN(In)+MPC) Number of parOcle

Coun

t

3.1<|η|<3.9

2015/1/23

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.32pion v

40-60 %

Event with Jet>20GeVExperimental RxN(In)+MPCExperimental RxN(Out)

(GeV/c)T

p0 5 10

3v

-0.2

-0.1

0

0.1

0.23pion v

40-60 %

AMPT RxN(In)+MPCAMPT RxN(Out)


Pion vn simulated by AMPT

Event plane dependence is found in v2, it is similar to the experimental measurement. The v3 shows similar trend less than 5 GeV/c.

2015/1/23

Results & Discussion Direct photon vn

PreDefence (M.Sanshiro) 33 2015/1/23

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0-20 %Neutral pionDirect photon

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

20-40 %

(GeV/c)T

p0 5 10

3v

-0.05

0

0.05

0.1

0-20 %

(GeV/c)T

p0 5 10

3v

-0.05

0

0.05

0.1

20-40 %

(GeV/c)T

p0 5 10

4v

0

0.1

0.2

0-20 %

(GeV/c)T

p0 5 10

4v

0

0.1

0.2

20-40 %


The results of neutral pion and direct photon vn

•  In low pT region Direct photon has non-‐zero and posiOve v2 and v3. The strength is comparable to the hadron v2 and v3. •  In high pT region Direct photon vn is close to zero.

2015/1/23

〉 part N〈0 100 200 300

2in

tegr

ated

v

-0.1

0

0.1

0.2

Neutral pionDirect photon

(RxN(I+O))2v<10 GeV/cT6<p

〉 part N〈0 100 200 300

3in

tegr

ated

v

-0.1

0

0.1

0.2 (RxN(I+O))3v

<10 GeV/cT

6<p

〉 part N〈0 100 200 300

4in

tegr

ated

v

-0.1

0

0.1

0.2

(RxN(I+O))4v

<10 GeV/cT6<p


Centrality dependence of γdir. and π0 in high pT

γdir. v2 and v4 < π0 v2 and v4 γdir. v3 ≈ 0 The direct photon vn is close to zero. The difference between γdir. and π0 is understood that photon includes prompt photon which vn≈0.

2015/1/23

Blast wave model is based on hydrodynamic model Parameterizing the expanding medium at freeze-‐out temperature pT spectra : azimuthal anisotropy : 6 parameters Freeze-‐out temperature : Tf average velocity : <ρ> transverse anisotropy : ρ2, ρ3 spaOal anisotropy : s2, s3 6 parameters are defined by hadron observables. Photon spectra and vn are predicted as massless parOcle (mass = 0).


dN

pT dpT/

Zd�I0(↵T )K1(�T )

vn(pT ) =

Rd� cos (n�)In(↵T )K1(�T ){1 + 2sn cos (n�)}R

d�I0(↵T )K1(�T ){1 + 2sn cos (n�)}

↵T (�) = (pT /Tf ) sinh (⇢(�))

�T (�) = (mT /Tf ) cosh (⇢(�))

⇢(�) = ⇢0(1 + 2⇢n cos (n�))

h⇢i =Rr(⇢0 ⇥ r/R

max

)drRr dr

Blast wave model


Photon observables predicted blast wave model

(GeV/c) T

p0 1 2 3 4

dy TN

/dp

2)d Tpπ

(1/2

-510

-310

-110

10

310

spectraT

PIDed charged hadron pCentrality:0-20%

0.27[MeV]±=107.37fT0.00± =0.65〉ρ〈

10× ±π 5× ±K 1× pp

(GeV/c) T

p0 1 2 3 4

2v

0

0.05

0.1

0.15

2PIDed charged hadron v

0.000±=0.0212ρ

0.000±=0.0332s

Fitting range

Extrapolation

(GeV/c) T

p0 1 2 3 4

3v

0

0.05

0.1

0.15

3PIDed charged hadron v

0.000±=0.0183ρ

0.001±=0.0063s

(GeV/c) T

p0 1 2 3 4

dy TN

/dp

2)d Tpπ

(1/2

-510

-310

-110

10

310

spectraT

Direct photon p

CalorimeterConversion method

(GeV/c) T

p0 1 2 3 4

2v

0

0.05

0.1

0.15

2Direct photon v

(GeV/c) T

p0 1 2 3 4

3v

0

0.05

0.1

0.15

3Direct photon v

Photon observables are well described with parameters defined from hadron. The pT spectra is described though temperature is very low (Teff. =240 MeV). It could be due to strong radial flow (like blue shil correcOon).


The probability of understanding photon puzzle

Blast wave model well describes photon pT spectra and vn at freeze-‐out temperature (TFO ≈ 100 MeV). EffecOve temperature could be affected by strong radial flow. The possibility that photons from early stage are not dominant. Photon pT spectra and vn are calculated with blue shil effect.

T 0 = T

s1 + �

1� �

T’ : apparent temperature β  : velocity of photon source T < T’

2015/1/23


Photon pT spectra and vn with blue shij effect Assump9on of photon source •  temperature decreases with the Ome : T(t) •  velocity increases with the Ome : β(t) •  azimuthal anisotropy increases with the Ome vn(pT, t) •  thermal photon momentum distribuOon : pT spectra and vn at final state are calculated as : EffecOve temperature is taken via fiTng by exponenOal equaOon to pT spectra. The difference with experimental measurement is esOmated as :

(Vobs. � V

cal.)/E(stat.� sys.)

nfin.(pT ) =

Zdtn(pT , t) vfin.n (pT ) =

Rdtn(pT , t)vn(pT , t)R

dtn(pT , t)

2015/1/23

Vobs. : experimental measurement E : error of Vobs. Vcal. : calculaOon

n(pT , t) =pT

exp (pT /T (t))� 1


time0 0.5 1

N(t)

0

0.5

1

1.5

2

time0 0.5 1

Prob

abili

ty d

ensi

ty

0

0.02

0.04Basic assumption

T,t)dp

Tn(p∫N(t)=

(t)=B tβ

) tT

,t)=C(pT

(pnv

time0 0.5 1

Tem

pera

ture

0

0.1

0.2

0.3 Temperature : T(t)

time0 0.5 1

Velo

city

0

0.2

0.4

0.6 (t)βVelocity :

time0 0.5 1

Azi

mut

hal a

niso

trop

y

0

0.05

0.1 ,t)T

(pnazimuthal anisotropy : v

Basic assump9on for yield, velocity, and anisotropy

The temperature is decreased from 300 MeV to 100 MeV. The Ome is defined by temperature.

2015/1/23

(GeV/c) T

p0 1 2 3 4 5

spe

ctra

Tp

-610

-410

-210

1

210310 spectra

TThermal photon p

=235(MeV)eff.T

without blue shift correction

Experimental measurement7±25±=239eff.T

=0.15eff. Tσ

(GeV/c) T

p0 1 2 3 4 5

spe

ctra

Tp

-610

-410

-210

1

210310

arXiv:1405.3940<2 GeV/c

Tfit range 0.6<pextrapolation

spectraT

Thermal photon p=265(MeV)eff.T

with blue shift correction

=-1.00eff. Tσ

(GeV/c) T

p0 1 2 3 4 5

2v

0

0.05

0.1

0.15 nDirect photon vn

Calculated photon v

(0-20 %)2

Thermal photon v

=3.742 vσ

(GeV/c) T

p0 1 2 3 4 5

3v0

0.05

0.1

0.15 (0-20 %)

3Thermal photon v

=1.513 vσ

< 3 GeV/cT

is estimated in pn vσ


EffecOve temperature with blue shil is higher than without correcOon. The vn are much underesOmated.

Calcula9on with basic assump9on

2015/1/23

time 0 0.5 1

Prob

abili

ty d

ensi

ty

0

0.02

0.04

,t)T

n(patT

dp∫N(t)=a=0a=1/4a=1/2

a=1a=2a=4

time 0 0.5 1

)βVe

loci

ty(

0

0.2

0.4

0.6 b(t)=B tβ

b=1/4b=1/2b=1b=2b=4

time 0 0.5 1

Azi

mut

hal a

niso

trop

y0

0.05

0.1 c,t)=C tT

(pnv

c=1/4c=1/2c=1c=2c=4


Addi9onal assump9on •  Yield dependence Since photon source expands, the yield is assumed to get large with Ome. •  Velocity dependence •  Azimuthal anisotropy dependence

T : V ∨ : ∧ T’ : V’

N(t) =

ZdpT t

an(pT , t)

�(t) = B · tb

vn(pT , t) = C(pT ) · tc

2015/1/23

time 0 0.5 1

Prob

abili

ty d

ensi

ty0

0.02

0.04

Probability density


(GeV/c) T

p0 1 2 3 4 5

spe

ctra

Tp

-610

-410

-210

1

210a=0 : 264.88[MeV]a=1/4 : 260.99[MeV]a=1/2 : 257.32[MeV]a=1 : 250.75[MeV]a=2 : 240.07[MeV]a=4 : 226.16[MeV]

(GeV/c) T

p0 1 2 3 4 5

2v

0

0.05

0.1

0.15T

dp/T)-1}T

/{exp(pa t∫N(t)=a=4a=2a=1a=1/2a=1/4a=0

(GeV/c) T

p0 1 2 3 4 5

3v

0

0.05

0.1

0.15

pT spectra and vn with rela9ve yield dependence

Photons from late stage : low temperature & large vn The both of effecOve temperature and vn are affected.

N(t) =

ZdpT t

an(pT , t)

2015/1/23

time 0 0.5 1Ve

loci

ty0

0.2

0.4

0.6 (t)βVelocity :

(GeV/c) T

p0 1 2 3 4 5

spe

ctra

Tp

-610

-410

-210

1

210b=1/4 : 324.93[MeV]b=1/2 : 294.54[MeV]b=1 : 264.88[MeV]b=2 : 244.39[MeV]b=4 : 235.77[MeV]

(GeV/c) T

p0 1 2 3 4 5

2v

0

0.05

0.1

0.15 b (t)=B tβ b=1/4b=1/2b=1b=2b=4

(GeV/c) T

p0 1 2 3 4 5

3v

0

0.05

0.1

0.15


pT spectra and vn with velocity dependence

EffecOve temperature significantly decreases with increasing “b”. The vn is a slightly affected.

2015/1/23

�(t) = B · tb

(GeV/c) T

p0 1 2 3 4 5

spe

ctra

Tp

-610

-410

-210

1

210264.88[MeV]

(GeV/c) T

p0 1 2 3 4 5

2v

0

0.05

0.1

0.15 c,t)=C tT

(pnv c=1/4c=1/2c=1c=2c=4

(GeV/c) T

p0 1 2 3 4 5

3v

0

0.05

0.1

0.15


pT spectra and vn with anisotropy dependence

Since pT spectra is not affected, effecOve temperature is not varied. The vn increases with “c” decreasing.

2015/1/23 time 0 0.5 1

Azi

mut

hal a

niso

trop

y0

0.05

0.1,t)

T(pnAzimuthal : v

vn(pT , t) = C(pT ) · tc


Comparison with experimental measurement

The differences (σTeff and σv2) are varies uniquely with the parameters “a”, “b”, and “c”. They are selected so that Teff. and v2 are comparable to the experimental measurement.

eff Tσ-4 -2 0 2

2 vσ

0

2

4

6 Basic assumptionyield dependencevelocity dependenceanisotropy dependence

(Vobs. � V

cal.)/E(stat.� sys.)

“c” geang large

“b” geang large

Vobs. : experimental measurement E : error of Vobs. Vcal. : calculaOon result

N(t) =

ZdpT t

an(pT , t)

�(t) = B · tbvn(pT , t) = C(pT ) · tc

(GeV/c) T

p0 1 2 3 4 5

spe

ctra

Tp

-610

-410

-210

1

210238.57[MeV]238.85[MeV]237.05[MeV]235.86[MeV]238.60[MeV]238.21[MeV]

(GeV/c) T

p0 1 2 3 4 5

2v

0

0.05

0.1

0.15 a : b : c0.00 : 2.86 : 0.020.25 : 1.98 : 0.030.50 : 1.77 : 0.04

(GeV/c) T

p0 1 2 3 4 5

3v

0

0.05

0.1

0.15 a : b : c1.00 : 1.47 : 0.062.00 : 1.04 : 0.094.00 : 0.63 : 0.14


pT spectra and vn with selected parameters

Parameter “b” is selected so that effecOve temperature is comparable to experimental measurement. The “c” is chosen to be comparable to v2.

2015/1/23

time0 0.5 1

N(t)

0

0.5

1

1.5

2

time0 0.5 1

Prob

abili

ty d

ensi

ty

0

0.02

0.04

a : b : c0.00 : 2.86 : 0.020.25 : 1.98 : 0.030.50 : 1.77 : 0.041.00 : 1.47 : 0.062.00 : 1.04 : 0.094.00 : 0.63 : 0.14

time0 0.5 1

Tem

pera

ture

0

0.1

0.2

0.3 Temperature

time0 0.5 1

Velo

city

0

0.2

0.4

0.6 Velocity

time0 0.5 1

Azi

mut

hal a

niso

trop

y

0

0.05

0.1azimuthal anisotropy


Selected dependence of velocity and anisotropy

The development of velocity and vn could be studied. The Ome dependence of vn of medium shows that vn is saturated early.

2015/1/23


Neutral pion and direct photon vn are measured in √sNN = 200GeV Au+Au collisions at RHIC-‐PHENIX experiment. Neutral pion v2, v3, and v4 show different pa[ern in high pT region. In central collisions, hadron shows non-‐zero vn.

It could be understood that jet path-‐length dependence. In peripheral collisions, v2 & v4 have posiOve while v3 shows negaOve.

It could be due to jet effect on determining event plane. Comparison of neutral pion and direct photon in high pT region. Hadron shows non-‐zero while photon vn is close to zero. It could be understood that photon includes prompt photons with vn≈0.

Summary (1)

2015/1/23


Strong radial flow effect could be the probability of explanaOon “photon puzzle”. High effecOve temperature could be understood that photon emi[ed from the medium having large radial velocity at late stage. Photon pT spectra and vn are calculated with blue shil effect. The Ome dependences are selected so that effecOve temperature and v2 are comparable to the experimental measurement. It is found that the medium azimuthal anisotropy is saturated early Ome of development.

Summary (2)

2015/1/23

PreDefence (M.Sanshiro) 51 2015/1/23

(GeV/c)T

p0 1 2 3 4 5

2v

0

0.1

0.2

0.3

2Direct photon v0-20 %

CalorimeterConversionHess et al. (a)Hess et al. (b)Linnyk et al.

(GeV/c)T

p0 1 2 3 4 5

2v

0

0.1

0.2

0.3

20-40 %arXiv1403.7558 (a)arXiv1403.7558 (b)arXiv1404.3714

(GeV/c)T

p0 1 2 3 4 5

2v

0

0.1

0.2

0.3

40-60 %


Model comparison of photon v2

(Orange) Transport model considering photons from hadron phase (Blue, red) Fireball model Hydrodynamic calculaOons (cyan, pink, and violet) including photons from late state, are much underesOmated.

PRC 84,054906 PRC 89,034908

Private communica9on

2015/1/23

(GeV/c)T

p0 1 2 3 4 5

2v

0

0.1

0.2


CalorimeterConversion

(GeV/c)T

p0 1 2 3 4 5

3v

0

0.1

0.2


Hess et al. (a)Hess et al. (b)Linnyk et al.

P.R.D 89,026013

arXiv1403.7558 (a)

arXiv1403.7558 (b)

arXiv:1404.3714


Model comparison of v2 and v3

Dark violet is based on magneOc field effect, upper limit is shown. Model calculaOons of photon v3 are much smaller than experimental data. The data of v3 may help to constrain parameters in model calculaOons.

P.R.D 89,026013 arXiv:1404.3714

PRC 84,054906 PRC 89,034908



2015/1/23

(GeV/c)T

p0 1 2 3 4 5

2v

0

0.05

0.1

0.15

(RxN(I+O)) inclusive photon2v0-20(%) Calorimeter

Conversion

(GeV/c)T

p0 1 2 3 4 5

3v

0

0.05

0.1

0.15

(RxN(I+O)) inclusive photon3v0-20(%)


Real photons from external photon conversion at the Hadron Blind Detector (HBD) readout plane are detected. •  Extend low pT limit Consistent inclusive photon vn well

External photon conversion method MHBD: Real track Mvtx : Measured track

MHB

D[GeV

/c2 ]

Mvtx[GeV/c2]

γ Invariant mass of e++e-‐ pair

2015/1/23


1) real photon converts to e+e-‐ in HBD backplane 2) default assumpOon: track come from the vertex 3) momentum of the conversion tracks will be mis-‐measured (see black tracks) 4) apparent pair-‐mass (about 12MeV) will be measured for phtons 5) assume the same tracks originate in the HBD backplane 6) re-‐calculate momentum and pair mass with this “alternate tracking model” 7) for true converted photons Matm will be around zero

Real track es9mated track

MHB

D[GeV

]

Mvtx[GeV]

External photon conversion method


Thermal photons are predicted sensi9ve to η/s

Hydrodynamic calculaOon predicts that direct photon is sensiOve to η/s of QGP than hadron.

v 2{SP}/v

3{SP}

pT(GeV/c)

arXiv:1403.7558v1 (2014)

2015/1/23

(GeV/c) T

p0 1 2 3 4

3/v 2v

0

2

4

6

0-20 %

±π

dir.γ

dir.γ/s(0.20) ηMCKLN+dir.γ/s(0.08) ηMCGlb+±π/s(0.20) ηMCKLN+±π/s(0.08) ηMCGlb+

(GeV/c) T

p0 1 2 3 4

3/v 2v

0

2

4

6

20-40 %theory arXiv:1403.7558

arXiv:1412.1038±π

(GeV/c) T

p0 1 2 3 4

3/v 2v

0

2

4

6

40-60 %


The ra9o of v2 to v3 in pT region

•  Photons don’t have strong centrality dependence at around 2-‐3 GeV/c

•  Pions increase from central to peripheral

Photon and pion show different centrality dependence.

π± : arXiv:1412:1038 Model : arXiv:1403.7558 Private communica9on

2015/1/23

time0 0.5 1

N(t)

0

0.5

1

1.5

2

time0 0.5 1

Prob

abili

ty d

ensi

ty

0

0.02

0.04Yield dependence

T,t)dp

Tn(pat∫N(t)=

(t)=B tβ

) tT

,t)=C(pT

(pnv

a=0a=1/4a=1/2a=1a=2a=4

time0 0.5 1

Tem

pera

ture

0

0.1

0.2


time0 0.5 1

Velo

city

0

0.2

0.4

0.6 (t)βVelocity :

time0 0.5 1

Azi

mut

hal a

niso

trop

y

0

0.05

0.1,t)

T(pnazimuthal anisotropy : v


Rela9ve yield dependence N(t) =

Ztan(pT , t)dpT

RelaOve probability density is modified.

2015/1/23

time0 0.5 1

N(t)

0

0.5

1

1.5

2

time0 0.5 1

Prob

abili

ty d

ensi

ty

0

0.02

0.04Velocity dependence

T,t)dp

Tn(p∫N(t)=

b(t)=B tβ

) tT

,t)=C(pT

(pnv

b=1/4b=1/2b=1b=2b=4

time0 0.5 1

Tem

pera

ture

0

0.1

0.2


time0 0.5 1

Velo

city

0

0.2

0.4

0.6 (t)βVelocity :

time0 0.5 1

Azi

mut

hal a

niso

trop

y

0

0.05

0.1,t)



The velocity dependence

The Ome dependence of velocity is modified.

2015/1/23

time0 0.5 1

N(t)

0

0.5

1

1.5

2

time0 0.5 1

Prob

abili

ty d

ensi

ty

0

0.02

0.04Azimuthal anisotropy dependence

T,t)dp

Tn(p∫N(t)=

(t)=B tβ

c) tT

,t)=C(pT

(pnv

c=1/4c=1/2c=1c=2c=4

time0 0.5 1

Tem

pera

ture

0

0.1

0.2


time0 0.5 1

Velo

city

0

0.2

0.4

0.6 (t)βVelocity :

time0 0.5 1

Azi

mut

hal a

niso

trop

y

0

0.05

0.1,t)



The azimuthal anisotropy dependence

2015/1/23


eff Tσ-4 -2 0 2

b

0

2

4

6

8

10a : b0 : 2.861/4 : 1.981/2 : 1.77

1 : 1.472 : 1.044 : 0.63

2 vσ0 2 4

c

0

0.5

1

1.5

2

a : b : c0 : 2.86 : 0.02a : b : c

1/4 : 1.98 : 0.03

a : b : c

1/2 : 1.77 : 0.04

a : b : c

1 : 1.47 : 0.06

a : b : c

2 : 1.04 : 0.09

a : b : c

4 : 0.63 : 0.14

Selec9ng b and c

2015/1/23


Medium effect (RAA)

RAA=1 not modified RAA≠1 medium effect Hadron less than unity -‐> medium effect

RAA

=(1/Nevt

AA

)d2NAA

/dpT

dy

< Ncoll

> /�inel

pp

⇥ d2�pp

/dpT

dy

photon RAA=1 : not modified

-‐> Emi[ed from iniOal hard sca[ering RAA>>1 : There are other photon sources which are not in p+p collisions.

2015/1/23


Event Plane resolu9on RxN(In) RxN(Out) RxN(I+O)

MPC BBC RxN(In)+MPC

vtruen

= vobs.n

/Res( n

)

Extracted from the correlaOon between the event plane angle measured by south and north. The vn measured with event plane should be corrected by the resoluOon.

=

qhcos {n( true

n

� obs.

n

)}i

hcos {n( South

n

� North

n

)}i

2015/1/23

(GeV/c) T

p0 5 10 15

2v

-0.2

0

0.2

0-20 %

|<3.8ηRxN(In)+MPC:1.5<||<1.5ηRxN(Out):1.0<|

2 vdir.γ

(GeV/c) T

p0 5 10 15

2v

-0.2

0

0.2

20-40 %

(GeV/c) T

p0 5 10 15

2v

-0.2

0

0.2

40-60 %

(GeV/c) T

p0 5 10 15

3v

-0.4

-0.2

0

0.20-20 %

3 vdir.γ

(GeV/c) T

p0 5 10 15

3v

-0.4

-0.2

0

0.220-40 %

(GeV/c) T

p0 5 10 15

3v

-0.4

-0.2

0

0.240-60 %


The event plane dependence of direct photon vn

2015/1/23


Charged hadron vn

Non-‐zero posiOve vn are observed. The trend of centrality dependence is not similar.

P.R.L. 107, 252301 (2011)

2015/1/23


Event Plane correla9on P.R.L. 107, 252301 (2011)

Ψ2 and Ψ3 are uncorrelated.

2015/1/23


Iden9fied charged par9cle vn

It is observed that all harmonics have mass ordering there are meson and baryon spliTng

arXiv:1412:1038

2015/1/23


The number of cons9tuent quark scaling (NCQ scale)

(a) : v2(KET)/nq (b) : vn1/n scaling (a)+(b) : vn(KET)/nqn/2

All parOcles are scaled by modified NCQ scaling.

arXiv:1412:1038

2015/1/23


π0 and γdir. v2 measurement by STAR

γdir. v2 in high ET region are consistent with 0 within systemaOc uncertainty, while π0 has posiOve v2.

Ahmed M. Hamed shown at QM

2015/1/23


photon vn measurement by ALICE

v2

It is also observed that γdir. v2 is posiOve in low pT at LHC-‐ALICE. v3 measurement is ongoing.

v3 γdec. v3 γinc. v3

arXiv:1212.3995v2 Friederike shown at QM

2015/1/23


The comparison of charged pion and hydro-‐model

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0-‐20% 20-‐40% 40-‐60%

pT(GeV/c) pT(GeV/c) pT(GeV/c)


v2

v3

π± : arXiv:1412:1038 Model : arXiv:1403.7558 Private communica9on

MCGlb+η/s(=0.08) MCKLM+η/s(=0.20)

2015/1/23

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3

0 1 2 3 40

0.1

0.2

0.3


The comparison of direct photon and hydro-‐model 0-‐20% 20-‐40% 40-‐60%



v2

v3 MCGlb+η/s(=0.08) MCKLM+η/s(=0.20)

π± : arXiv:1412:1038 Model : arXiv:1403.7558 Private communica9on 2015/1/23


(GeV/c) T

p0 5 10 15

2v

-0.2

-0.1

0

0.1

0.2

0.30-20 %

RxN(I+O)2v dir.γ

(GeV/c) T

p0 5 10 15

2v

-0.2

-0.1

0

0.1

0.2

0.320-40 %

(GeV/c) T

p0 5 10 15

2v

-0.2

-0.1

0

0.1

0.2

0.340-60 %

(GeV/c) T

p0 5 10 15

3v

-0.2

-0.1

0

0.1

0-20 %

RxN(I+O)3v

(GeV/c) T

p0 5 10 15

3v

-0.2

-0.1

0

0.1

20-40 %

(GeV/c) T

p0 5 10 15

3v

-0.2

-0.1

0

0.1

40-60 %


(GeV/c)T

p0 5 10

nv

0

0.1

0.2

0.3

0.40-10%

2Neutral pion v

(GeV/c)T

p0 5 10

nv

0

0.1

0.2

0.3

0.4Present analysis

P.R.C 88, 064910 (2013)

10-20%

(GeV/c)T

p0 5 10

nv

0

0.1

0.2

0.3

0.420-30%

(GeV/c)T

p0 5 10

nv

0

0.1

0.2

0.3

0.430-40%

(GeV/c)T

p0 5 10

nv

0

0.1

0.2

0.3

0.440-50%

(GeV/c)T

p0 5 10

nv

0

0.1

0.2

0.3

0.450-60%

Comparison of neutral pion v2 with previous results

They are consistent within systemaOc uncertainty.

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

0-20%

2Neutral pion v

E.P. : RxN(I+O)

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3 Present analysisP.R.L. 109, 122302 (2012)

20-40%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

40-60%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

0-20%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

20-40%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

40-60%


Comparison of neutral pion v2 with previous results


(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

0-20%

2Inclusive photon v

E.P. : BBC

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3 Present analysisP.R.L. 109, 122302 (2012)

20-40%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

40-60%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

0-20%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

20-40%

(GeV/c)T

p0 5 10

2v

0

0.1

0.2

0.3

40-60%


Comparison of inclusive photon v2 with previous results


(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2

0.250-10%

2Neutral pion v

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2

0.25Present analysis

arXiv:1412.1038 (2014)

10-20%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2

0.2520-30%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2

0.2530-40%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2

0.2540-50%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2

0.2550-60%


Comparison of neutral pion v2 with charged pion v2


(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.150-10%

3Neutral pion v

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15Present analysis

arXiv:1412.1038 (2014)

10-20%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.1520-30%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.1530-40%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.1540-50%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.1550-60%




(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.20-10%

4Neutral pion v

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2Present analysis

arXiv:1412.1038 (2014)

10-20%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.220-30%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.230-40%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.240-50%

(GeV/c)T

p0 1 2 3 4 5

nv

0

0.05

0.1

0.15

0.2




(GeV/c) T

p0 5 10 15

2v

-0.1

0

0.1

0.2

0-20 %

2Direct photon vE.P. : RxN(I+O)

(GeV/c) T

p0 5 10 15

2v

-0.1

0

0.1

0.2Present analysisP.R.L. 109, 122302 (2012)

20-40 %

(GeV/c) T

p0 5 10 15

2v

-0.1

0

0.1

0.2

40-60 %

(GeV/c) T

p0 5 10 15

2v

-0.1

0

0.1

0.2

0-20 %

2Direct photon vE.P. : BBC

(GeV/c) T

p0 5 10 15

2v

-0.1

0

0.1

0.2

20-40 %

(GeV/c) T

p0 5 10 15

2v

-0.1

0

0.1

0.2

40-60 %


Comparison of direct photon v2 with previous results


Documents

MeasurementOf)DirectPhoton)Higher)Order)) Azimuthal ...utkhii.px.tsukuba.ac.jp/report/2014/Mizuno_D/Predefence_photon_vn_Sanshiro.pdfSanshiro)Mizuno) High)Energy)Nuclear)Physics)Group)