phenix upc

Report
J/y Photoproduction in
ultra-peripheral Au+Au collisions
at √sNN =200 GeV measured by
RHIC-PHENIX
TAKAHARA, Akihisa
for the PHENIX Collaboration
(CNS, University of Tokyo and RIKEN)
1
Characteristics of
Ultra Peripheral Collisions
b>2R
no nuclear overlap. Possibility to
study g- induced reactions
Coherence condition:
wavelength > nucleus size . Very
low photon virtuality
Z 1e
A
b
v~c
b > 2R
A
Z 2e
Weizsacker-Williams (EPA):
•
The electromagnetic field is
equivalent to a large flux of quasireal photons, and can be calculated
per (Fermi)-Weizsacker-Williams:
RHIC max. photon
Wggmax  6GeV
energies(EPA):3 GeV~g/R
WgNmax  34GeV
RHIC, LHC LgA/LAA
2
Physics motivation for UPC J/y
gluon distribution in Nuclei is not same free proton
RHIC UPC J/y
Q2=2.5GeV2
x~0.01
Theoretical predictions
And 2004 result
•Direct Measurement of gluon distributions at low-x
•search for Nuclear shadowing
3
coherent and incoherent distribution
Strikman, Tverskoy, Zhalov, PLB
626 p. 72-79、2005
•Strikman et al calculate that quasi-elastic
(incoherent) J/y cross-section comparable to
coherent production
•Incoherent J/y produced from photo-nucleon
interaction
•Much larger t distribution expected for
incoherent
•Both process emit only J/y and neutron. pT
distribution is important to divide them
•Strikman’s predictions say at central rapidity,
coherent process is dominant, but at forward
rapidity, incoherent process is dominant at
PHENIX
n incoherent
coherent
n incoherent
coherent
4
RHIC-PHENIX
Luminosity
Au+Au(200GeV) : 2 x 1026 [cm-2s-1]
p+p (500GeV): 2 x 1031 [cm-2s-1]
2007-RUN
•for central
•AuAu
•200GeV
•~530/μb
2010-RUN
•for forward
•AuAu
•200GeV
~800/μb
Central arm (|y|<0.35)
electron (using 2007 data)
Forward arm (1.2<|y|<2.2)
muon(using 2010 data)
5
Signal from UPC J/y and its trigger
Coherent UPC
n Signal from UPC J/y→ll
1. a lepton pair without any other tracks
e+
2. 50~60% UPC events are associated with
nuclear break up
e- PHENIX UPC trigger
1. BBC_VETO(reject nuclear overlap)
2. EMCAl(for central)/Muon track(for
50-60%
forward)
3. ZDC detect at least a neutron
BBC
BBC(3<|y|<3.9)
BBC is main vertex detector of PHENIX
1st condition means we can’t use it
PHENIX MB :4kHz
PHENIX ERT2x2(EMcal):8kHz
to reduce trigger rate,
3rd condition was required
6
UPC J/y+Xn measurement at
PHENIX Central arm(|y|<0.35,X>1)
p
e+
e-
g
+
Offline analysis cuts
1. |collision vertex determined from tracks reconstructed in the PAD
chambers|< 30cm
2. number of tracks==2
3. North or south BBC charge==0
4. energy deposit of ZDC>30GeV(just confirm there are no noise
trigger)
7
UPC J/y+Xn(y>0)Yn(y<0) measurement
at PHENIX Forward arm(X>1,Y>1)
5 interaction length→
to get vertex information,
both side ZDC fire was required
Offline analysis cuts
1. |vertex determined from
ZDC|<30 cm
2. number of tracks==2 (in
central and forward)
3. North or south BBC charge==0
4. energy deposit of ZDC>30GeV
1. Just noise cut
5. Both RXNP charge <1000a.u.
2010 run forward UPC mass dist
(North 1.2_<y<2.2)
Clear J/ψ peak
8
Real data for dielectron(|y|<0.35)
Dimass distribution
Unlike
like
ZDC energy
dipT distribution
pT(GeV/c)
• Clear J/y peak
• Only unlikesign pair (over 2 GeV)
• Clear Coherent(low pT) peak
9
Real data for dimuon(1.2<|y|<2.2)
North
Unlike
like
dipT distribtuion
2.7 GeV<dimass<3.5 GeV
North
South
South
Unlike
like
• Clear J/y peak
• Only unlike sign pair (2 GeV>mass)
• Coherent(low pT) peak is not so clear
• But still 0 pT peak
10
Neutron emission for dimuon J/y
North ZDC
South ZDC
Dimuon
North
• Same side:1300GeV
• Opp side :700GeV
opp side
Multi Photon
excitation
Dimuon
South
• If there was nuclear over lap, the asymmetry can’t be
explained
• Suggest UPC (incoherent) process
Same side
Multi photon
excitation
+Nuclear break up
by recoiled
neutron
11
Comparison with pp for dielctron
Dimass distribution
Unlike
like
dipT distribution
Unlike
Like
~1GeV/c peak
Even pp(ncoll=1 limit),
• J/y events are associated with additional tracks
• At pT ~2.5 GeV/c, unlike/like~2
• UPC and PP J/y pT distribution is different
Number of tracks in central arm
UPC(without central track cut)
PP
Normalized at 2
12
Comparison with pp for dimuon
Dimass distribution
Unlike
like
dipT distribution
Unlike
like
Even pp(ncoll=1 limit),
• J/y events are associated with additional
tracks
• At pT ~2.5 GeV/c, unlike/like~2
• UPC and PP J/y pT distribution is different
• Central multiplicity distribution suggest little
most peripheral contamination
• About 3 J/y per arm.
Number of tracks in central arm(not muon arms)
UPC(without central track cut)
PP(0track/non 0tracks~30%)
Normalized at 0
13
Contamination form diffractive process
• Diffractive J/y should have just 2 tracks
• UPC like events !
• Typically, diffractive collisions /all pp~30%
• 10k J/y was generated by PYTHIA (pp 200GeV,msel2(minimum byas))
• 0/10k J/y was generated by diffractive process
• Can be neglected
14
Background sources
for dielectron(|y|<0.35)
Rapidity distribution
STARLIGHT simulation
for gg->dilepton
In central(|y|<0.35) region, γγ->dielectron is main background source
“Simulated continuum curve +Gaussian” fit
J/y gaus +trig,detecter Acc xeff(mass)x exp
• Exp slope was fixed by simulation
15
Background sources
for dimuon(|y|<0.35)
• HERA:gp->J/y measurement
y(2s)/J/y=7%
Eur. Phys. J. C 24, 345–360 (2002)
• gg->dimuon can be neglected this
rapidity region
•
Expected most peripheral contamination
•
doesn’t have enough statistics to explain
all background
North dipT <0.5GeV/c
J/y
y(2s)
Total background
Background by UPC process are suggested
• UPC ccbar
• gAu->dipion->dimuon
16
Integrated cross section
J/y +Xn Central 2004 & 2007
sys errors
Acc(over all) 5%
simulation 12%
Lumi 4%
ERT 0.2%
Njpsi 1.4%
BBC 1.6%
2004+2007
2004 PHENIX
76  31 (stat) 15 (syst) b
2007 PHENIX
J. Nystrand, Nucl. Phys. A 752(2005)470c; A.J. Baltz,
S.R. Klein, J. Nystrand, PRL 89(2002)012301; S.R.
Klein, J. Nystrand, Phys. Rev. C 60(1999)014903
M. Strikman, M. Tverskoy and M. Zhalov, Phys. Lett.
B 626 72 (2005)
V. P. Goncalves and M. V. T. Machado,
arXiv:0706.2810 (2007).
Yu. P. Ivanov, B. Z. Kopeliovich and I. Schmidt,
arXiv:0706.1532 (2007).
17
J/y+Xn Invariant Yield
@(-0.35<y<0.35)
theoretical
calculations for Coherent (@y=0)
•no shadowing 113μb
•DS
10 μb
•EKS
83μb
•Kopeliovich GBW 61μb
•Kopeliovich KST 54μb
•EPS
53μb
•Strikman impulse 40 μb
•Strikman glauber 30 μb
•We can see both coherent and incoherent distribution
•46.7 ±13μb for pT < 0.4GeV(upper limit of coherent)
•compatible with calculations including strong suppression of gluons at low x
18
J/y +Xn(y>0)Yn(y<0)
•The pT distributions at forward rapidity shows that incoherent
process is very visible at forward (can’t see coherent peak)
•There are no theoretical predictions with XnYn condition
19
Summary and Outlook
Summary
• PHENIX measured J/ ψ photo-production yield and its
pT dependence in a broad rapidity region.
– characteristics of UPC signals is obviously different from pp
• - J/ψ +Xn result at mid-rapidity is consistent with
calculations suggesting strong gluon shadowing
• - important contribution from incoherent processes in
J/ψ+Xn(y<0)Yn(y>0) at forward rapidity, looking
forward for calculations for this exclusive process
Outlook
• New vertex detectors will help in the further study of
UPC events at RHIC.
20
21
22
23
24
gg->dimuon distribution
• gg->dimuon distribution is very
sharp in this region
• Because of edge of detector,
Acceptance x efficiency is very low
at Y=1.2
25
Detail of fitting
J/y gaus +trig,detecter Acc xeff(mass)x exp
• Exps lope was fixed by simulation
26
Detail of fitting
Divide into pT bins
• Due to hadron suppresser
Fitting function
Acceff(dimass)x
Gaus1(J/y)
Gaus2(J/y tail)
Gaus3(y(2s))
+exp(background)
North dipT <0.5GeV/c
J/y
y(2s)
Total background
• Shape of gaus 1&2 was fixed to pp data
• Gaus3/Gaus(1+2) is fixed to 7%
27

similar documents