alan dion - tata institute of fundamental researchtheory.tifr.res.in/~qgpsat/dion.pdf · heavy...
TRANSCRIPT
Heavy Flavor at
Alan DionIowa State University
2008-02-14
Measuring Charm/Bottom at PHENIX
Direct ReconstructionD0 → K + π−
D0 → K + π−π0
Difficult without accurate ver-tex measurement (cτ ∼ 123µm)
Indirect MeasurementMeasure contribution from semilep-tonic decays of heavy flavor toelectron spectra.
Both single and pair spectra
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 2
Electron Identification
DetectorsTracking in drift chamber. Track matching to RICH andEMC.Ring size/shape in RICHE/p distribution from the EMC and DC
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 3
Electron Identification
E/p [c]0.4 0.6 0.8 1 1.2 1.4 1.60
2000
4000
6000
8000
10000
12000
14000
16000
< 2.5 GeV/cT
E/p for 2.0 GeV/c < p
total
Kaon decays,some photonconversions
associatedhadrons
vertex electrons,beampipe
conversions
flip
slide
North South
Hadronic BackgroundSome hadronic tracks are randomlyassociated with a ring in the RICH.These are statistically subtracted byswapping the north and south sides ofthe RICH in software.
Energy/Momentum DistributionThe E/p distribution gives strong evidence that we understand our eID. Kaonswhich decay far from the collision have mis-reconstructed momentum. Mosttracks passing cuts form a Gaussian centered at 0.98.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 4
Cocktail Method
MethodAll relevant background sources aremeasured.
Decay kinematics and photon conver-sion rate are simulated.
Background cocktail is subtractedfrom inclusive spectrum.
Performs well at high pT where sig-nal/background is large.
Not limited by statistics.
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9 10
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9 10
]3/c
-2p
[m
b G
eV3
/dσ3E
d-1210
-1110
-1010
-910
-810
-710
-610
-510
-410
-310
-210
-110
1
10
210 = 200 GeV (Run-5)sp+p @
)/2-+e+cocktail: (eeeγ → 0π
conversionγeeγ → ηeeγ →’ η
ee→ ρee0π → ω ee and → ω
eeη → φ ee and → φ contributionsγdirect
Ke3all electrons
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 5
Converter Method
MethodAdd material of known thicknessaround the beampipe and compare theelectron spectra with and without thematerial installed.
NHF =RγNinc −N converter
inc
Rγ − 1
Works best at low pT where photonicsources are significant
Limited by statistics of converter run
Converter method is used to normalizethe cocktail method
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9
±b
ackg
rou
nd
e
±h
eavy
-fla
vor
e
0
0.5
1
1.5
2
2.5
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 6
Electrons from Heavy Flavor in p+p)3 c
-2 (
mb
GeV
3/d
pσ3
E d
-1010
-910
-810
-710
-610
-510
-410
-310
-210
-110
(a)=200 GeVs)/2 + X at - + e+
(e→ p+p
PHENIX dataFONLL(total)
e)→FONLL(c e)→FONLL(b
e)→ c →FONLL(b
(GeV/c)Tp0 1 2 3 4 5 6 7 8 9 10
0.51
1.52
2.53
DA
TA
/FO
NL
L (b)
For pT between 4-6 GeV/c, the electron signal/background in PHENIX is about 2. Evenif PHENIX subtracted no electron background, they would still be below the STAR result!
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 7
Muon Sources
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 8
Single Muon Methodology
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 9
Muons from Heavy Flavor in p+p
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 10
Suppression of Charm in Au+Au
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9
)]-2
[(G
eV/c
)d
yT
dp
N2d
T
pπ21
-1410
-1310
-1210
-1110
-1010
-910
-810
-710
-610
-510
-410
-310
-210
-110
1
10
210
310
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9
±b
ackg
rou
nd
e
±h
eavy
-fla
vor
e0
0.5
1
1.5
2
2.5
4 10×Min-Bias 2 10×0-10%
1 10×10-20% 0 10×20-40% -1 10×40-60% -2 10×60-92%
/42mb-2 10×p+p = 200 GeVNNsAu+Au @
partN0 50 100 150 200 250 300 350
AA
R0
0.2
0.4
0.6
0.8
1
1.2
1.4 = 200 GeVNNsAu+Au @
> 0.3 GeV/cT
: p±e
> 3.0 GeV/cT
: p±e
> 4.0 GeV/cT
: p0π
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 11
Models for Charm Suppression
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9 10
AA
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
= 200 GeVNNsAu+Au @
0-10% centralPHENIX
STAR
DGLV
Col. Dis.
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9 10
AA
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
= 200 GeVNNsAu+Au @
0-10% centralPHENIX
STAR
q̂ = 4GeV 2/fm
q̂ = 10GeV 2/fm
q̂ = 14GeV 2/fm
(GeV/c)T
p0 1 2 3 4 5 6 7 8 9
AA
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6 PHENIX data/D enhancement factor of 5cΛ/D enhancement factor of 12cΛ
Lots of models, not much constraint fromheavy flavor suppression alone.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 12
Charm/Bottom Separation
The IdeaLook for D0 → eK with no PID on theKaon.
Reconstruct invariant mass for the electronhadron pair.
Subtract mass distribution from like-signcombination.
The Calculation
εdata ≡Ntag
NHeavy Flavor e±=Nc→tag +Nb→tag
Nc→e +Nb→e
εc ≡Nc→tag
Nc→e, εb ≡
Nb→tag
Nb→e
Nb→e
Nc→e +Nb→e=εc − εdata
εc − εb
Tagging efficiency determined fromsimulation. Production ratios of variousheavy flavor mesons gives the mainuncertainty.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 13
Charm/Bottom Separation
Bottom/charm ratio is in good agreement with FONLL.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 14
Electron Spectra from Charm and Bottom
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 15
Bottom Cross Section
σbb = 4.61± 1.31(stat)+2.57−2.22(sys)µb
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 16
D0 → K−π+π0 Reconstruction
Utilize the photon trigger to enhance statistics. Low pT is currently being worked onusing electron tagging.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 17
Dielectrons in p + p
)2 (GeV/ceem0 1 2 3 4 5 6 7 8
/GeV
) IN
PH
EN
IX A
CC
EP
TA
NC
E2
(c
ee d
N/d
mev
t1/
N
-1110
-1010
-910
-810
-710
-610
-510
-410
-310 = 200 GeVsp+p at DATA
|y| < 0.35 > 0.2 GeV/c
Tp
)2 (GeV/ceem0 1 2 3 4 5 6 7 8
/GeV
) IN
PH
EN
IX A
CC
EP
TA
NC
E2
(c
ee d
N/d
mev
t1/
N
-1110
-1010
-910
-810
-710
-610
-510
-410
-310
)2 (GeV/ceem0 0.2 0.4 0.6 0.8 1
/GeV
) IN
PH
EN
IX A
CC
EP
TA
NC
E2
(c
ee d
N/d
mev
t1/
N
-710
-610
-510
-410
-310 eeγ → 0π
eeγ → η
eeγ →’ η
ee→ ρ
ee0π ee & → ω
eeη ee & → φ
ee→ ψJ/
ee→’ ψ
ee (PYTHIA)→ cc
ee (PYTHIA)→ bb
ee (PYTHIA)→DY
sum
)2 (GeV/ceem0 0.2 0.4 0.6 0.8 1
/GeV
) IN
PH
EN
IX A
CC
EP
TA
NC
E2
(c
ee d
N/d
mev
t1/
N
-710
-610
-510
-410
-310
)2 (GeV/ceem0 1 2 3 4 5 6 7 8
/GeV
) IN
PH
EN
IX A
CC
EP
TA
NC
E2
(c
ee d
N/d
mev
t1/
N
-1110
-1010
-910
-810
-710
-610
-510 = 200 GeVsp+p at DATA - COCKTAIL
ee (PYTHIA)→ cc
ee (PYTHIA)→ bb
ee (PYTHIA)→DY
sum
|y| < 0.35 > 0.2 GeV/c
Tp
)2 (GeV/ceem0 1 2 3 4 5 6 7 8
/GeV
) IN
PH
EN
IX A
CC
EP
TA
NC
E2
(c
ee d
N/d
mev
t1/
N
-1110
-1010
-910
-810
-710
-610
-510
Dielectrons above and below J/ψ mass compared to charm, bottom and Drell-Yanfrom PYTHIA.
σbb = 3.9± 2.5(stat)+3−2(sys)µb
σcc = 518± 47(stat)± 135(sys)± 190(model)µb
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 18
Dielectrons in Au + Au
)2 (GeV/ceem0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
/GeV
) IN
PH
EN
IX A
CC
EP
TA
NC
E2
(c
ee d
N/d
mev
t1/
N -710
-610
-510
-410
-310
-210
-110 = 200 GeVNNsmin. bias Au+Au at eeγ → 0π
eeγ → ηeeγ →’ η
ee→ ρ
ee0π ee & → ωeeη ee & → φ
ee→ ψJ/ ee→’ ψ ee (PYTHIA)→ cc
sum ee (random correlation)→ cc
DATA|y| < 0.35
> 0.2 GeV/cT
p
Agreement with PYTHIA in charm region. But dielectrons in charm region should besuppressed since D mesons change direction in the medium. So maybe there is room forthermal component.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 19
Single Electron v2
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2In
clu
sive
Ele
ctro
n v
0
0.05
0.1
0.15
0.2
0.25
minimum-bias
PH ENIXPRELIMINARY
Au+Au√
sNN = 200 GeV
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2v
0
0.05
0.1
0.15
0.2
0.25
0-10%
10-20%
20-40%
40-60%
min-bias
v2 of inclusive electrons measured usingthe event plane from the new RXPNdetector.
Cocktail for photonic electron v2 takesmeasured hadron v2 as input.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 20
Single Electron v2
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2H
eavy
Fla
vor
Ele
ctro
n v
-0.1
-0.05
0
0.05
0.1
0.15
0.2
0.25
0-10%
PH ENIXPRELIMINARY
Au+Au√
sNN = 200 GeV
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2H
eavy
Fla
vor
Ele
ctro
n v
-0.1
-0.05
0
0.05
0.1
0.15
0.2
0.25
10-20%
PH ENIXPRELIMINARY
Au+Au√
sNN = 200 GeV
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2H
eavy
Fla
vor
Ele
ctro
n v
-0.1
-0.05
0
0.05
0.1
0.15
0.2
0.25
20-40%
PH ENIXPRELIMINARY
Au+Au√
sNN = 200 GeV
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2H
eavy
Fla
vor
Ele
ctro
n v
-0.1
-0.05
0
0.05
0.1
0.15
0.2
0.25
0.3
40-60%
PH ENIXPRELIMINARY
Au+Au√
sNN = 200 GeV
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 21
Single Electron v2
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2H
eavy
Fla
vor
Ele
ctro
n v
-0.1
-0.05
0
0.05
0.1
0.15
0.2
PHENIX Final Run4
PHENIX Preliminary Run7
minimum-bias
Run-7 was bad for electrons. The HBD didn’t work, we used a weaker magnetic field(for the HBD), and we had no helium bag (conversions in air). The systematic errors willimprove, but will stay larger than the Run4 result.
We have only analyzed half of the Run7 data so far.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 22
Charm Quarks Flow
[GeV/c]T
p0 1 2 3 4 5 6 7 8 9
AA
R
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
1.8
= 200 GeVNNsAu+Au @
0-10% central
Moore &Teaney (III) T)π3/(2
T)π12/(2
van Hees et al. (II)
Armesto et al. (I)
[GeV/c]T
p0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
2H
eavy
Fla
vor
Ele
ctro
n v
-0.1
-0.05
0
0.05
0.1
0.15
0.2
PHENIX Final Run4
PHENIX Preliminary Run7van Hees et al.
minimum-bias
The transport model from van Hess et. al. fits the data pretty well, and its resonancesnear the D mass gain some support from lattice calculations. It would be nice to see whateffect adding gluon radiation would have.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 23
Short-Term Outlook
D meson ReconstructionThis measurement should improve in the near future. Run 6 p+ pand Run 8 d + Au have not been looked at. But this is a verydifficult measurement.
e− µ correlationsSome electron-muon correlation analyses have begun. Azimuthalopening angle distributions look promising for another heavy flavormeasurement. e − µ can also be used as a trigger for the directreconstruction.
Cu + Cu DataWe still have Cu+Cu data to analyze. This is simply a manpowerissue.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 24
Long-Term Outlook
5 cm
55 cme-
e+
Hadron Blind DetectorWe have one more shot at gettingthis thing to work, and we think weknow how. The dielectron measure-ment from Run-9 is something to lookforward to.
Silicon Vertex DetectorWith decay positions measured accu-rately, direct D and B measurementwill be a breeze.
Heavy Flavor at PHENIX: 2008-02-14 Alan Dion 25