alexander kup co for the d˜ collaboration · calor’04, perugia march 29 - april 2, 2004...
TRANSCRIPT
� �� �� �� �� �� � � � � � �� � �� �
�� � � � � � � �� �� ��
CALOR’04, Perugia
March 29 - April 2, 2004
Alexander Kupco
for the DØ collaboration
CEA Saclay, France
visitor from Institute of Physics, Academy of Sciences of the Czech Republic
Run II upgrade of the DØ detector
Tracking SystemTracking System: Silicon, Fiber Tracker,: Silicon, Fiber Tracker,Solenoid, Central & ForwardSolenoid, Central & Forward Preshowers Preshowers
ShieldingShielding
Fiber Tracker/Fiber Tracker/Preshower Preshower VLPC Readout SystemVLPC Readout System
NN SSMuon ToroidMuon Toroid
Muon Muon ScintillationScintillationCountersCountersForward Mini-Forward Mini-
Drift TubesDrift Tubes
PDTsPDTs
PlatformPlatform
CCCC
ECEC ECEC
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 2
Calorimeter
DO LIQUID ARGON CALORIMETER
1m
CENTRAL CALORIMETER
END CALORIMETER
Outer Hadronic (Coarse)
Middle Hadronic (Fine & Coarse)
Inner Hadronic (Fine & Coarse)
Electromagnetic
Coarse Hadronic
Fine Hadronic
Electromagnetic
� uniform and hermetic
- coverage up to � η � < 4.2
� nearly compensating
� fine segmentation
- ∆η � ∆ϕ = 0.1 � 0.1
(3rd EM layer: 0.05 � 0.05)
� particle energy resolution
e : σE = 15%√
E
� 0.3%
π : σE = 45%√
E
� 4%
� �� � � � � � � �
- shorter time between bunch crossings (396 ns) � faster trigger and readout electronics
- more material in front of calorimeter (magnet, new tracker) � new preshower detector
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 3
Jets
q
Tim
e
p p
q g
K
“part
on
jet”
“part
icle
jet
”“ca
lori
met
er j
et”
hadrons
CH
FH
EM
. calorimeter jet- calorimeter - main tool for jet measurement
- jet is a collection of towers
- geometrical definition (cone algorithms)
- pQCD motivated algorithms (kT )
. particle jet- after hadronization
- a spread of particles running roughly in thesame direction as the parton
. parton jet- parton hard scattering
- parton showers
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 4
Run II cone algorithm
. a collection of towers within a given cone R =√
∆2ϕ + ∆2η
- ϕ is the azimuthal angle
- pseudorapidity η is related to the polar angle ϑ : η = � log tan (ϑ/2)
. recombination scheme - E-scheme
Ejet =∑
towers
Ei , ~pjet =∑
towers
~pi
. cone axis - stable directions
- iterative procedure starts only fromseeds above some threshold
- using midpoints as an additional start-ing seeds
. splitting and merging overlapping jets
. accept only jets with ET > 8 GeV
eta
-4.7 -3
-2 -1
0 1
2 3
4.7
phi180
0
360
ET(GeV)
385
Bins: 481Mean: 2.32Rms: 23.9Min: 0.00933Max: 384
mE_t: 72.1phi_t: 223 deg
Run 178796 Event 67972991 Fri Feb 27 08:34:03 2004
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 5
Jet energy scale
correction of the jet energy measured on the detector level to the jetenergy on the particle level
Ejetptcl =
Ejetdet
� �
Rjet S
�� � �� ( � ) - energy not associated with the hard interaction (U noise,pile-up from previous crossings, additional pp interactions)
� � � ��� � � � (Rjet)
- calorimeter response to the jet
- EM part calibrated on Z � ee mass peak
- measured from ET balance in γ + jet events
� � �� � (S) - losses due to showering the energy in the calorimeterout of the jet cone
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 6
Offset - ( � � � cone)�� � �� ( � ) - energy not associated with the hard interaction (U noise,
pile-up from previous crossings, additional pp interactions)
|ηDetector |0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
)φ×η∆(G
eV/
TE
D
0.2
0.4
0.6
0.8
1
1.2
1.4
|ηDetector |0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
)φ×η∆(G
eV/
TE
D
0.2
0.4
0.6
0.8
1
1.2
1.4
L = 32.5E30
L = 19.5E30
L = 11.5E30
� average tower ET density
� ET=
∑
φ ET (η)
2π ∆η Nevents
� � ETmeasured in two dif-
ferent samples
- zero bias (accelerator clock)
- minimum bias (inelastic scat-
tering)
� offset for R = 0.7 cone is about 1 GeV
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 7
Jet response� deposited energy is different from the measured one
- calorimeter is not perfectly compensated : eπ
� 1.05 for E > 30 GeV
- dead material, module-to-module fluctuations, . . .
Missing ET projection method
- jet response determined fromthe pT imbalance in γ+jet events
~ETγ + ~ETrecoil = 0 (ideal)
Rγ~ETγ + Rrecoil
~ETrecoil = � ~ET� (real)
- after EM energy calibration from the
Z � ee mass peak (Rγ = 1)
Rrecoil = 1 +~nTγ · ~ET6
ETγ
- select nice, back-to-back, γ+jet events
Rjet = Rrecoil
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 8
EM scale
Di-Electron Mass (GeV)0 20 40 60 80 100 120
Eve
nts
/ 2
GeV
0
20
40
60
80
100
Number of Entries: 604
Peak Mass: 90.8 +- 0.2 GeV
Width: 3.6 +- 0.2 GeV
D0 Run II Preliminary
� determined from Z � e+e−
mass peak
� typical size of EM scale is between 1-5%
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 9
Response in central and forward cryostats
E’ [GeV]10 210
jet
R
0.65
0.7
0.75
0.8
0.85
0.9
CC
ECS
ECN
= 0.7coneR
Events are placed in various E ′ bins
E′ = ETγ cosh(ηjet)
- photon energy and jet direction are well
measured quantities
- smearing effects are minimazed
CC - jet in the central cryostat
ECS - jet in the south end-cap
ECN - jet in the north end-cap
� different response in cryostats � cryostat factors
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 10
Response
[GeV]detE10 210
jet
R
0.6
0.65
0.7
0.75
0.8
0.85
0.9
det log E1 + p0 p 0.013± = 0.425 0p 0.004± = 0.082 1p
= 0.7coneR
0.006± = 0.916 ECSFcr
0.007± = 0.948 ECNFcr
� mapping from E ′ to measuredjet energies Edet
� statistical error
< 1% for 20 � 250 GeV
� systematic error
� 4 � 5% for 20 � 250 GeV- variation of selection cuts
- closure tests on MC and different
data samples (Z + jet)
� for jets in the end-caps, corresponding cryostat factor is applied
� special treatment in the inter-cryostat region (ICR)
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 11
Response in ICR
Detector Eta0 0.5 1 1.5 2 2.5 3
Detector Eta0 0.5 1 1.5 2 2.5 3
Res
po
nse
/Fit
0.7
0.75
0.8
0.85
0.9
0.95
1
1.05
1.1 Data Response/Fit
15 < E_T < 22.5
22.5 < E_T <30.0
E_T > 30.0
Data Response/Fit
� η dependence studied in γ+jetevents
� expect log-linear dependence inuniform detector
Rjet(η) = a + b log[cosh(η)]
ICR correction
- obtained from the response deviation from the log-linear dependence
- applied in the region 0.5 < � η � < 1.5
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 12
Showering -(
� � cone
)
� � �� � (S) - losses due to showering the energy in the calorimeterout of the jet cone
Method
- study energy flow as a function ofdistance from the jet axis
r =√
(η � ηjet)2 + (φ � φjet)2
R = 0.7 R = 0.5
Central 0.99 0.92
ICR 0.96 0.89
Forward 0.94 0.85
� 5% systematic error
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 13
Overall jet energy correction
(GeV)uncorrE50 100 150 200 250 300 350 400 450 500
(GeV)uncorrE50 100 150 200 250 300 350 400 450 500
0.6
0.8
1
1.2
1.4
1.6
1.8
2
=0jet
η=0.7, coneR
Correction vs. E
(GeV)uncorrE50 100 150 200 250 300 350 400 450 500
(GeV)uncorrE50 100 150 200 250 300 350 400 450 500
0
0.05
0.1
0.15
0.2
0.25
0.3
total error
Stat. error
Correction error vs. E
|η|0 0.5 1 1.5 2 2.5
|η|0 0.5 1 1.5 2 2.5
0.8
1
1.2
1.4
1.6
1.8
2
=50 GeV uncorr
jet=0.7, EconeR
|ηCorrection vs. |
|η|0 0.5 1 1.5 2 2.5
|η|0 0.5 1 1.5 2 2.5
0
0.05
0.1
0.15
0.2
0.25
0.3
total error
Stat. error
|ηCorrection error vs. |
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 14
Jet T resolution
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80
200
400
600
800
1000
1200
1400
1600
1800
2000
2200
2400
< 100 GeVT 80 GeV < p
A� dijet events for E > 50 GeV
- select nice back-to-back dijet events
- pT asymmetry
� = � pT1
� pT2 �
pT1 + pT2
is directly related to pT resolution
σpT
pT= 2σA
- corrected for the soft radiation (additional jets below the 8 GeV recon-struction cut) and for jet imbalance on hadron level
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 15
Jet T resolution
)/2 [GeV]T2 + pT1(p0 50 100 150 200 250 300
0
0.05
0.1
0.15
0.2
0.25
0.3
T)/pT(pσ | < 0.5 (R=0.7) η |T)/pT(pσ
)/2 [GeV]T2 + pT1(p0 50 100 150 200 250 300
0
0.05
0.1
0.15
0.2
0.25
0.3
T)/pT(pσ| < 2. (R=0.7) η 1.5 < |T)/pT(pσ
σpT
pT=
√
√
√
√
N2
p2T
+S2
pT+ C2
N - noise term
S - sampling term
C - constant term
� larger value of resolution than in Run I� reason is being investigated now
- more dead material, larger noise
� we are working currently on deter-mination of jet response from γ+jetsample
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 16
Summary� Calibrating jet energies and understanding jet energy resolutions are es-
sential parts of the DØ Run II program and is needed by most physicsanalyses
� Although in Run II, the calorimeter is the same as in Run I, jet energyscale and resolutions need to be derived from the very beginning due tonew electronics and more dead material.
� Run II jet energy scale has been derived for data and MC for two differentcones, R = 0.5 , 0.7- currently, the error is 7% in data for central jets in the 30 � 250GeV region
- the goal is to reach the understanding of the scale on the level of 2-3%
� Run II resolutions larger than in Run I- improvements are expected from advanced algorithms combining different subdetector
information
� In future we expect further progress using: Z � bb, W � qq in tt events
Alexander Kupco, CEA Saclay, France CALOR’04, Perugia page 17