“search for multi-lepton events with the zeus detector at ...parenti/dtesi/dtalk.pdf · luca...
TRANSCRIPT
Università degli Studi di PadovaDottorato di Ricerca in Fisica – XVI Ciclo
“Search for multi-lepton events with the
ZEUS detector at HERA”
and“VCMVD: an algorithm for MVD tracking
at TLT”Dottorando: Andrea Parenti Tutore: Dott. Luca Stanco
Coordinatore: Prof. Attilio Stella
Padova, 18 Dicembre 2003Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.1/92
Outline – Part I
• The experimental setup• Introduction and motivations• Multi-electron events search• Di-muon events search
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.2/92
The HERA ColliderHERA is a particle accelerator
located in Hamburg...
ZEUS
...and collides e±p at√s =
300/318 GeV:E =820 (920) GeVpE =27.5 GeVe
In 1992-2000 HERA sup-
plied the following inte-
grated luminosity:Years Lumi [pb−1] Type
1992-94 2.19 e−p
1994-97 70.92 e+p
1998-99 25.20 e−p
1999-00 94.95 e+p
• In 1998√s was raised from 300 GeV to 318 GeV.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.3/92
The HERA ColliderHERA kinematics:
• Centre-of-mass energy:s = (P + k)2.
• Transferred 4-momentum, squared:Q2 = −(k − k′)2.
• Bjorken-x: x = Q2/(2P · q).• Inelasticity: y = (P · q)/(P · k).
• Proton-photon mass, squared:W 2 = (P + q)2.
Electron-Proton scattering•• Neutral Current: ep→ eX
• Charged Current: ep→ νX
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.4/92
The ZEUS Detector
ZEUS is a general purpose detector:
The components used in the anal-yses are:
• Central Tracking Detector(CTD),
• Uranium Calorimeter (CAL),
• Muon Chambers (B/RMUON+ FMUON),
• Luminosity Monitor.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.5/92
Di-lepton production at HERALepton pair production is a pure QED process, well predictable in
the Standard Model (SM) context.
• Main contribution: Bethe-Heitler (BH) process.
• Other contributions: Cabibbo-Parisi (CP) and Drell-Yan (DY).
Feynmann diagrams:l
l
l
Drell−Yan
ee
ee
l
l
pX
p
l
l
X
X
Cabibbo−Parisi
X
l
l
l
e
e
p
e
pe
Xp
Bethe−Heitler
ee
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.6/92
Backgrounds
• Vector meson decays: Υ → l+l− (BR ∼ 1-2%).
• Single lepton production:• Heavy quark decays: q → q′W ⋆ → q′lν;
• W-production;
• τ decays.
• Backgrounds to multi-electron search:processes where hadrons or photons aremisidentified as electrons:• Neutral Current DIS;
• QED-Compton.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.7/92
Motivations
• Test QED & photon spectrum of the proton.
• Cross-check H1 excesses:High Mass multi-electrons at H1:
10−3
10−2
10−1
1
10
102
0 50 100 150
2e
H1 Data 115 pb −1
Pair Production(GRAPE)
DIS+Compton
M12
(GeV)
Even
ts
1
10
0 50 100 150
M12
(GeV)
Even
ts
10−3
10−2
10−1
3e
Selection Data MC
2e, M12 > 100 GeV 3 evts 0.30±0.04
3e, M12 > 100 GeV 3 evts 0.23±0.04(see hep-ex/0307015)
Isolated leptons with P/T :H1 Data
All SM processes
SM error
Signal (W+Z prod)
−
10−2
10−1
1
10
102
0 50 100 150 200
MT
/ GeV
Even
ts
Ndata
= 18
NSM
= 12.4 1.7
Combined Electron and Muon
+
Type Data MC
e+p 10e 9.9 ± 1.3
8µ 2.56 ± 0.44
e−p 1e 1.69 ± 0.22
0µ 0.37 ± 0.06
(see Phys. Lett. B561, 241)
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.8/92
Multi-electron search
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.9/92
Event Selection
XY View ZR View
Zeus Run 33753 Event 4607 date: 24-09-1999 time: 23:52:03ZeV
is
E=200.12 GeV =106.49 GeVt
E = 49.86 GeVzE-p =144.54 GeVfE = 55.58 GeVbE= 0.00 GeVrE = 3.35 GeV
tp = -0.69 GeVxp = 3.27 GeVyp =150.26 GeVzp
phi= 1.78 = 0.43 nsft = 0.27 nsbt =-100.00 nsrt = 0.39 nsgt= 55.05 GeVeE = 1.22eθ = 0.15eφ = 0.17
e,DAx = 0.27
e,DAy
2=4531.09 GeVe,DA2Q
PreselectionFast selection and reduction of
data. Selection criteria:
• Trigger selection: di-electron ORNC-DIS OR High-ET ;
• No. of CTD tracks: 1 ≤ NTrkVtx ≤ 9;
• Transverse momentum in CTD:Pi PT,i > 3 GeV.
Electron selection• EM electron finder (CTD+CAL info);
• Isolation cuts:• Econe < 0.3 GeV in R=0.3 (ηφ),• NTrk = 0 in R=0.4 (ηφ);
• Ee > 10 GeV (θ ≤ 164) OREe > 5 GeV (θ > 164).
Event Selection• Vertex: |z| < 50 cm;
• two electrons in 17 < θ < 164;
• Ee1T > 10 GeV;
• Ee2T > 5 GeV.
Two classes: “2e” (2 electrons)
and “3e” (3 electrons) events.Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.10/92
Data to MC comparison1996-2000 data taking L=120.49 pb−1
Selection Data All SM di-ele MC NC-DIS MC QED-C MC
All 2e+3e 209 235.82 207.00 18.72 10.10
M12 > 50 GeV 23 35.01 30.14 3.40 1.47
M12 > 100 GeV 1 1.12 0.91 0.08 0.13
10-2
10-1
1
10
102
0 20 40 60 80 10010
-3
10-2
10-1
1
10
102
0 50 100 150
10-2
10-1
1
10
0 1 2 3
ZEUS 2e+3e
ETe1 (GeV)
ZEUS 1996-00GRAPE+NC+QEDCNC+QEDCQEDC
Ee1 (GeV)
θe1 (rad) φe1 (rad)
1
10
-2 0 2
10-2
10-1
1
10
102
0 20 40 60 80 100
10-2
10-1
1
10
102
0 50 100 150
1
10
0 1 2 3
ZEUS 2e+3e
ETe2 (GeV)
ZEUS 1996-00GRAPE+NC+QEDCNC+QEDCQEDC
Ee2 (GeV)
θe2 (rad) φe2 (rad)
1
10
-2 0 2
10-2
10-1
1
10
102
0 50 100 150
10-1
1
10
-2 0 2
10-1
1
10
102
0 1 2 3
ZEUS 2e+3e
M12 (GeV) abs(θe1-θe2) (rad)
φe1-φe2 (rad)
ZEUS 1996-00GRAPE+NC+QEDCNC+QEDCQEDC
(E-Pz)ee (GeV)
10-2
10-1
1
10
102
0 20 40 60
E − Pz permits to separate Bkg from signal.
Monte Carlo describes data reasonably well.Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.11/92
Cross section: sample selectionA clean ep → eeX sample is needed for
cross section measurement.
“γγ” selection• 2 electrons in 17 < θ < 164;
• Ee1T > 10 GeV and Ee2
T > 5 GeV;
• M12 > 5 GeV;
• E−Pz < 45 GeV (cuts NC-DIS and QED-C).
ResultsSelection Data All SM di-ele MC
All γγ 51 52.86 52.71
M12 > 50 GeV 3 2.88 2.80
M12 > 100 GeV 0 0.09 0.08
10-3
10-2
10-1
1
10
0 20 40 60 80 100
10-2
10-1
1
10
0 1 2 3
10-3
10-2
10-1
1
10
0 20 40 60 80 100
ZEUS γγ
ETe1 (GeV)
ZEUS 1996-00GRAPE+NC+QEDCNC+QEDCQEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
10-2
10-1
1
10
0 1 2 3
10-3
10-2
10-1
1
10
0 50 100 150
10-2
10-1
1
-2 0 2
10-1
1
10
0 1 2 3
ZEUS γγ
M12 (GeV) abs(θe1-θe2) (rad)
φe1-φe2 (rad)
ZEUS 1996-00
GRAPE+NC+QEDC
NC+QEDC
QEDC
(E-Pz)ee (GeV)
10-3
10-2
10-1
1
10
0 20 40 60
Nic
eag
reem
ento
fdat
aan
dM
C.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.12/92
Total cross section
Period Purity Acceptance σDATA (pb) σMC
1996-97 0.825 0.234 1.66 ± 0.43+0.18−0.14 1.55
1998-99 0.773 0.292 0.82 ± 0.41+0.13−0.12 1.66
1999-00 0.790 0.283 1.73 ± 0.31+0.22−0.17 1.66
1996-00 1.62 ± 0.23+0.21−0.16
1996-97: cross-section is measured at√s = 300 GeV.
1996-00: cross-section is referred to√s = 318 GeV.
First uncertainty is statistical, second is systematic.
Main systematics: CAL energy scale simulation (+9.4−4.1%), isolation cut (+7.3%).
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.13/92
Differential cross section
0
0.1
0.2
0.3
20 30 40 50 60
10-2
10-1
20 30 40 50 60
0
0.1
0.2
10 15 20 2510
-2
10-1
10 15 20 25
0
0.1
0.2
0.3
0.4
0.5 1 1.5 2 2.5
M12 (GeV)
Acc.
M12 (GeV)
dσ/
dM12 (pb/GeV)
PTe (GeV)
Acc.
PTe (GeV)
dσ/
dPTe (pb/GeV) ZEUS 1996-00
GRAPE MC
θe (rad)
Acc.
θe (rad)dσ/
dθ e (pb/rad)
1
0.5 1 1.5 2 2.5
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.14/92
Discussion• Multi-electron events sought in ZEUS 1996-2000 data.
• Data distributions compared to SM: good description.
• The excess found by H1 is not confirmed.
• Total and differential cross sections were measured:
agreement with SM in shape and normalisation.
• Outlook: reduce systematic uncertainties, analyse 1994-95.
Phase space
• 2 electrons in 17 < θ < 164
• Ee1T > 10 GeV and Ee2
T > 5 GeV
• M12 > 5 GeV
• E − Pz < 45 GeV
Total cross section
σ(ep→ eeeX) = 1.62 ± 0.23+0.21−0.16 pb
σMC = 1.66 pb
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.15/92
Di-muon search
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.16/92
Event Selection
PreselectionFast selection and reduction of
data. Selection criteria:
• Trigger selection: B/RMUON activityOR FMUON activity;
• No. of CTD tracks: 2 ≤ NTrkVtx ≤ 7;
• Transverse momentum in CTD:Pi PT,i > 1.5 GeV.
Muon selection• Muon finding:
• GLOMU: CTD + CAL + B/RMUI,• MPMATCH2: CTD + FMUON,• CTD + CAL (MIPs);
• Isolation cut: DµTrk > 1 in ηφ;
• PT > 5 GeV.
Event Selection• Vertex: |z| < 50 cm ANDp
x2 + y2 < 0.5 cm;
• 2 muons (1 matched to chambers);
• cosmic muons rejection:• acollinearity: cos Ω > −0.995,• timing cut.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.17/92
Backgrounds [pb/GeV]
µ t/dP
σd
10−3
10−2
10−1
1
10
10−3
10−2
10−1
1
10
)/
DATA
EW(GRAPE)
(LPAIR)γ
SM
γµµττγγ
cc + cbb
µµΥµµ0
Z
[GeV]µt
P
10 20 30 40 50 60 70 80
σSM
σ −
σ(
−1
0
←←
←←
1
2
3
H1 Muon Pair Analysis
Backgrounds (Υ, heavy quarks
and τ decays) were analysed by H1:
found to be negligible.
In my analysis they were ne-
glected.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.18/92
Comparison of data to MC1996-2000 data taking L=101.47pb−1
Selection Data di-mu MC
All 218 241.14
Mµµ > 50 GeV 6 3.91
Mµµ > 100 GeV 0 0.23
10-1
1
10
102
0 10 20 30 40
1
10
102
0 1 2 3 4 5
10-1
1
10
0 1 2 3
PTµ1 (GeV)
ZEUS 1996-00
GRAPE MC
Dµ1Trk
θµ1 (rad) φµ1 (rad)
10-1
1
10
0 2 4 6
10-1
1
10
102
0 10 20 30 40
1
10
102
0 1 2 3 4 5
10-1
1
10
0 1 2 3
PTµ2 (GeV)
ZEUS 1996-00GRAPE MC
Dµ2Trk
θµ2 (rad) φµ2 (rad)
10-1
1
10
0 2 4 6
10-1
1
10
-2 0 210
-1
1
10
102
0 1 2 3
10-1
1
10
-1 -0.5 0 0.5 1
∆θ (rad) abs(∆φ) (rad)
ZEUS 1996-00
GRAPE MC
cos(Ω) Mµµ (GeV)
10-1
1
10
102
0 25 50 75 100 125
Three di-muons crowd at Mµµ ≃ 50 GeV: genuine events.
Monte Carlo describes data reasonably well.Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.19/92
Total cross section
Period Purity Acceptance σDATA (pb) σMC
1996-97 0.953 0.436 4.65 ± 0.56+0.35−0.31 5.31
1998-99 0.956 0.438 4.97 ± 0.98+0.29−0.27 5.21
1999-00 0.968 0.427 4.81 ± 0.45+0.35−0.33 5.32
1996-00 4.79 ± 0.33+0.35−0.32
1996-97: cross-section is measured at√s = 300 GeV.
1996-00: cross-section is referred to√s = 318 GeV.
First uncertainty is statistical, second is systematic.
Main systematics: muon chamber eff. (+5.4−4.8%), isolation cut (+4.0%).
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.20/92
Differential cross section
00.10.20.30.4
20 40 60 80 10010
-3
10-2
10-1
20 40 60 80 100
0
0.2
0.4
10 20 30 40 50
10-2
10-1
1
10 20 30 40 50
0
0.2
0.4
1 2
Mµµ (GeV)
Acc.
Mµµ (GeV)
dσ/
dM
µµ (pb/GeV)
PTµ (GeV)
Acc.
PTµ (GeV)
dσ/
dPTµ (pb/GeV) ZEUS 1996-00
GRAPE MC
θµ (rad)
Acc.
θµ (rad)dσ/
dθ µ
(pb/rad)
1
1 2
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.21/92
Discussion• Di-muon events sought in ZEUS 1996-2000 data.
• Data distributions compared to SM: good description.
• 3 events found at Mµµ ∼ 50 GeV: genuine di-muons.
• Total and differential cross section were measured:
agreement with SM in shape and normalisation.
• Outlook: reduce systematic uncertainties, analyse
background (bb, cc, Υ, τ decays).
Phase space
• 2 muons in 15 < θ < 164
• P µT > 5 GeV
• Mµµ > 5 GeV
Total cross section
σ(ep→ eµµX) = 4.79±0.33+0.35−0.32 pb
σMC = 5.32 pb
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.22/92
VCMVD: MVD tracking at ZEUSThird Level Trigger
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.23/92
Outline – Part II
• Introduction and motivations• The ZEUS Micro-Vertex Detector (MVD)• Description of the method• Results• Discussion
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.24/92
Introduction: track parameterisation
Internal detectors are plunged in a magnetic field :charged particles follows helix-like trajectories.
Helices are characterised by 5 parameters ai and areference point (x0, y0):
DH
φ1
Hφ
(x ,y )0 0
(0,0) x
y
R
s=0
(Q=+1)
(α,β)• a1 = φH : azimuthal angle at s = 0;
• a2 = Q/R: helix curvature;
• a3 = Q DH : closest approach to(x0, y0);
• a4 = ZH : z coordinate at s = 0;
• a5 = cot θ; θ is the polar angle at s = 0.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.25/92
MVD: motivationsF charm
2 measurement
MVD will allow better charmtagging (higher efficiencyand larger kinematicalrange).
Measurement ofF charm
2 and gluon density.
Photoproduction of jets
Resolved PHPDirect PHP
MVD will allow better tag-ging of resolved and directcharm/beauty photoproduc-tion and thus better eventselection and cross-sectionmeasurement.
Exotic processes
Heavy states are boostedforward.
daughters’ decaylenghts may be detectedeven if particles are short-lived.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.26/92
VCMVD: motivations
• After Micro-Vertex Detector (MVD) installation in 2001the ZEUS tracking package (VCTRAK) needed to beupdated to benefit from the new detector
• A “fast” algorithm was needed for the use in the ThirdLevel Trigger (TLT)
• VCMVD (VCtrak+MVD) has been developed to refineVCTRAK tracks by means of MVD hits
• Effects: great improvement in a4, a3 precision.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.27/92
The ZEUS MVD
Barrel Section
• 3 layers of silicon detectors;
each layer measures 2 views: rφ and z;
• intrinsic resolution: ∼ 20 µm;
• angular coverage (3 layers):22 < θ < 159.
Forward wheels
• 4 wheels of silicon detectors;
• extend coverage to θ = 7.
Only BMVD is used in the package.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.28/92
VCMVD: the method
• Tracks are found in CTD by
VCTRAK;
• MVD hits are reconstructed;
• VCTRAK tracks are propagated
from CTD to MVD;
• MVD hits are collected on the
outer layer;
• track parameters are updated;
• the same procedure is repeated for
middle and internal layers.
Cluster 1
Cluster 2
Cluster 3
Lad
der
Beam−Pipe
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.29/92
Results: efficiency
Geometrical coverage
• 2 layers: 72 in φ;
• 3 layers: 288 in φ;
Expected intersections:
〈n〉 ≃ 2.8/track/view.
Assigned clusters
• Single muon MC: 〈n〉 = 2.9;
• bb MC: 〈n〉 = 2.4 ÷ 2.5;
• Data (offline): 〈n〉 = 2.0.
0
200
400
600
800
1000
1200
0 2 40
200
400
600
800
1000
1200
0 2 4
IDEntriesMeanRMS
3000 2000
2.895 0.6675
2003/12/10 15.13
n(Rφ)
IDEntriesMeanRMS
3100 2000
2.918 0.6934
n(z)
IDEntriesMeanRMS
3050 2000
5.814 1.292
n(Rφ)+n(z)
0
200
400
600
800
1000
1200
0 5 100
500
1000
1500
2000
2500
3000
3500
4000
4500
0 2 40
500
1000
1500
2000
2500
3000
3500
4000
4500
0 2 4
IDEntriesMeanRMS
3000 9810
2.457 1.018
2003/12/10 15.16
n(Rφ)
IDEntriesMeanRMS
3100 9810
2.412 1.016
n(z)
IDEntriesMeanRMS
3050 9810
4.869 1.883
n(Rφ)+n(z)
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0 5 100
200
400
600
800
1000
0 2 40
200
400
600
800
1000
0 2 4
IDEntriesMeanRMS
3000 2738
2.016 1.110
2003/12/10 15.18
n(Rφ)
IDEntriesMeanRMS
3100 2738
1.993 1.062
n(z)
IDEntriesMeanRMS
3050 2738
4.009 2.016
n(Rφ)+n(z)
0
200
400
600
800
1000
0 5 10
b−b Monte Carlo_
Single mu MC Real Data
4
3
8
2
4
2
6
2
3
4
2
3
4
6
4
8
2
3
2
4
6
4
23 3
4
4
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.30/92
Results:DH and ZH
SAMPLE: Single muon Monte Carlo, pT = 0.5 ÷ 1.5 GeV
Results
• σ(DH): 650 µm (VCTRAK) → 300 µm (VCMVD);
• σ(ZH): 3.2 mm (VCTRAK) → 0.26 mm (VCMVD).
IDEntriesMeanRMS
1900 2000 0.3010E-01 0.1065E-01
2003/12/09 15.26
σ(a3) (cm)
Evt
s
Before MVD Fit
After MVD Fit
0
100
200
300
400
500
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14
IDEntriesMeanRMS
1950 2000 0.2627E-01 0.1469E-01
2003/10/04 18.50
σ(a4) (cm)
Evt
s
Before MVD Fit
After MVD Fit
0
200
400
600
800
1000
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
(a3)σ (a4)σ
2003/12/12 18.07
PT (GeV)
σ(a3
) (c
m)
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
0 1 2 3 4 5 6 7 8
b−b Monte Carlo_
• σ(DH) → 100 µm at high-PT .
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.31/92
Results: Execution TimeThe execution time is evaluated as fraction of VCTRAK
(CTD tracking package) time.
• Monte Carlo: 40% (single muon) → 11% (bb).
• Data (offline): 20 ÷ 25%.
IDEntriesMeanRMS
110 2000
0.3800 0.2919E-01
2002/12/24 17.45
T(VCMVD)/T(VCTRAK)
Evt
s
0
200
400
600
800
1000
0 0.2 0.4 0.6 0.8 1 1.2 1.4
IDEntriesMeanRMS
110 374
0.2023 0.1881
2002/12/24 17.52
T(VCMVD)/T(VCTRAK)
Evt
s
0
10
20
30
40
50
60
70
80
0 0.2 0.4 0.6 0.8 1 1.2 1.4
IDEntriesMeanRMS
110 999
0.1143 0.4663E-01
2002/12/24 17.50
T(VCMVD)/T(VCTRAK)
Evt
s
0
100
200
300
400
500
0 0.2 0.4 0.6 0.8 1 1.2 1.4
_b−b Monte CarloSingle muon MC Real data
Absolute execution time increases steeply for VCTRAKMulti-leptons at HERA – MVD tracking at ZEUS TLT – p.32/92
Results:PT and probabilitiesSAMPLE: Single muon Monte Carlo, pT = 0.5 ÷ 1.5 GeV
PT measurement
IDEntriesMeanRMS
1700 2000 0.8236E-02 0.1336E-02
2003/10/29 14.58
σ(PT) (GeV)
Evt
s
Before MVD Fit
After MVD Fit
0
50
100
150
200
250
300
350
0 0.002 0.004 0.006 0.008 0.01 0.012 0.014 0.016 0.018
• σ(PT , VCTRAK) = 11 MeV.
• σ(PT , VCMVD) = 8 MeV.
Probability of ai:
0
50
100
150
0 0.2 0.4 0.6 0.8 10
50
100
150
0 0.2 0.4 0.6 0.8 1
0255075
100
0 0.2 0.4 0.6 0.8 10
200
400
0 0.2 0.4 0.6 0.8 1
VCMVD2002/12/24 17.47
Prob(χ2)
Evt
s
Prob(χ2)
Evt
s
Prob(χ2)
Evt
s
Prob(χ2)
Evt
s
Prob(χ2)
Evt
s
0
100
200
300
0 0.2 0.4 0.6 0.8 1
a1 a2
a3 a4
a5
Definition of proba-bility:
P
(
(afitν − atrue
ν )2
σ2fit
, 1
)
Flat probabilities: fitting procedure is
correct.
Peaks at 0: hint of non gaussian tails.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.33/92
Discussion• A new vertex detector has been installed in 2001 at ZEUS
• The VCMVD tracking package has been developed to
update VCTRAK parameters
• VCMVD package has been tested offline: very good results
in execution time and tracking improvement
• VCMVD has been implemented in the online software: data
are coming for further testing
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.34/92
Conclusions
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.35/92
ConclusionsPart I
• HERA collisions have been analysed in the search for
multi-electron and di-muon events
• The data are nicely described by the SM simulation
• The excess observed by H1 in high-mass di/trielectrons is
not confirmed
Part II
• A tracking package, VCMVD, has been developed for use of
MVD in TLT
• VCMVD behaves well on (offline) data and MC
• Online data are coming for testingMulti-leptons at HERA – MVD tracking at ZEUS TLT – p.36/92
ADDITIONAL SLIDES
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.37/92
The ZEUS detector
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.38/92
The Central Tracking Detector (CTD)• Cylindrical wire drift chamber, 72 layers of
sense wires divided into 9 super-layers.
• -100 cm < z < 104 cm, 15 < θ < 164.
• Filled with Ar, CO2 and C2H6 (ethane) inthe proportion 85:5:1.
• Wires are tilted by 45 to compensate theLorentz angle.
The resolution on pT for tracks with pT > 150 MeV, constrained to theinteraction vertex and passing three CTD super-layers is
σ(pT )
pT
= 0.0058 · (pT /GeV) ⊕ 0.0065 ⊕ 0.0014
(pT /GeV).
The CTD also supplies information on the energy loss dE/dx.Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.39/92
The Uranium Calorimeter (CAL)• Longitudinally divided into three
parts (RCAL/BCAL/FCAL).
• Surrounds the tracking devicesand the solenoid.
• Consists of 3.3 mm thick depleteduranium plates alternated with2.6 mm thick organic scintillator.
• Compesation: same response forelectrons and hadrons.
Electromagnetic energy resolution
σ(E)
E=
18%√
E/GeV⊕ 2%
Hadronic energy resolution
σ(E)
E=
35%√
E/GeV⊕ 1%
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.40/92
The Barrel/Rear Muon Chambers
• Large area to be covered ⇒ modular structurebased on the chamber.
• Two layers of LST tubes placed on both sides ofa honeycomb support structure.
• x and y coordinates read by each LST plane(signal induced on conductive strips orthogonalto the wires, glued outside the plane).
• Cells filled with Ar, CO2 and C4H10 (isobutane).
• Chambers have different shape and dimension,depending on their position.
⋆ BMUI-RMUI: chambers between the CAL and the yoke.
⋆ BMUO-RMUO: chambers outside the yoke.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.41/92
The Forward Muon Chambers
The FMUON detector consists of:
• 4 LST trigger planes, with digital ρ and φ
read-out (LST1–4);
• Two coverage planes of LSTs, in the large polarangle region (LW1–2);
• 4 planes of drift chambers (DC1–4);
• Two large toroidal magntes, providing a 1.7 T fieldfor momentum separation and measurement inthe small polar angle region.
FMUO
LST4 LST3 LST2
LW1 LW2
DC4 DC3 DC2
FMUI
DC1
LST1
⋆ FMUI: LST1–DC1, inside the yoke.
⋆ FMUO: other chambers, outside the yoke.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.42/92
The Lumi Monitor
• Photon calorimeter atz = −(104 ÷ 107) m.
• Resolution:
σ(E)
E=
23%√
E/GeV.
• Photons from ep→ epγ are counted.
• Luminosity is extracted usingBethe-Heitler formula for σ(ep→ epγ).
• Precision: σ(L)/L = 1 ÷ 2%.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.43/92
Past results
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.44/92
Multi-electrons at H1Data taking: 1994-00 Lumi used: 115.2 pb−1
Event Selection• Two “central” (20 < θ < 150), “isolated”, electrons
• PT cut: P e1T > 10 GeV, P e2
T > 5 GeV
Event Classification• “2e”: Only 2 central electrons
• “3e/4e”: Additional “isolated” electrons (also Forward and Rear)
“γγ” subsample• Just 2 opposite charge electrons: ep → e+e−X
• E − Pz < 45 GeV (ie y < 0.82, Q2 < 1 GeV2)
“Cleaner” sample; scattered electron is lost in beam–pipe; both detected electronscome from interaction
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.45/92
Multi-electrons at H1
Different topology for “2e” and “3e” events:A “2e” event:
preliminary
PT=63 GeV
PT=62 GeV
Multi-electron Event M(12)=130 GeV
P e1T = 63 GeV, P e2
T = 62 GeV
M12 = 130 GeVHarder PT in “2e”
A “3e” event:
preliminary
PT=25 GeV
PT=16 GeV
PT=20 GeV
Multi-electron Event M(12)=118 GeV
P e1T = 25 GeV, P e2
T = 20 GeV
M12 = 118 GeVMore forward es in “3e”
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.46/92
Multi- e at H1: global variablesH1 Preliminary Multi-electron Analysis
0
20
40
60
0 50
2eH1 Data 115 pb-1
GRAPE
NC-DIS+Compton
E-Pz (GeV)
Eve
nts
10-2
10-1
1
10
10 2
0 20
2e
PT
miss (GeV)
Eve
nts
10-1
1
10
10 2
0 20 40
2e
PT
hadrons (GeV)
Eve
nts
0
10
20
0 50
3e
E-Pz (GeV)
Eve
nts
10-2
10-1
1
10
0 20
3e
PT
miss (GeV)
Eve
nts
10-1
1
10
0 20 40
3e
PT
hadrons (GeV)
Eve
nts
•G
RA
PE
:γγ
inte
ract
ion
+γ
&Z
0co
nver
sion
•N
C-D
IS+
Com
pton
:fa
ke“2
e”-”
3e”
even
ts
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.47/92
Multi- e at H1: electron variablesH1 Preliminary Multi-electron Analysis
10-2
10-1
1
10
10 2
0 50
2e
PTe1 (GeV)
Eve
nts
10-2
10-1
1
10
10 2
0 50
2e
PTe2 (GeV)
Eve
nts
10-1
1
10
0 20 40
3e
PTe1 (GeV)
Eve
nts
10-1
1
10
0 20 40
3e
PTe2 (GeV)
Eve
nts
10-1
1
10
0 20
3e
H1 Data 115 pb-1
GRAPENC-DIS+Compton
PTe3 (GeV)
Eve
nts
Three “2e” events with P e1T > 50 GeV (but low SM expectation)
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.48/92
Multi- e at H1: Mass distributionsH1 Preliminary Multi-electron Analysis
10-2
10-1
1
10
10 2
0 50 100 150
2e
H1 Data 115 pb-1
GRAPENC-DIS
+Compton
M12 (GeV)
Eve
nts
0
50
100
150
0 50 100 150 M12 (GeV)
PT
e1+
PT
e2 (
GeV
)
2eLMC(GRAPE)≅500×LDATA
10-2
10-1
1
10
0 50 100 150
3e
M12 (GeV)
Eve
nts
0
50
100
150
0 50 100 150 M12 (GeV)
PT
e1+
PT
e2 (
GeV
)
3e LMC(GRAPE)≅500×LDATA
M12 = Mass of two
highest PT electrons
Harder PT for “2e”
At M12 > 100 GeV
“2e” events:
3 found
0.25±0.05 expected
“3e” events:
3 found
0.23±0.04 expected
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.49/92
Multi- e at H1: γγ Cross Section
10−3
1
10−2
10−1
PT
e 1
1
21
2
(GeV)
dσ/
dP
Te
10
(p
b/G
eV)
15 20 25
M12
(GeV)
dσ/
dM
12 (
pb
/GeV
)10
−3
10−2
20 40 60
e p e e+e
− X
PT
e ≥ 10 GeV, PT
e ≥ 5 GeV
20o ≤ θe ,e ≤ 150
o
y ≤ 0.82, Q2 ≤ 1 GeV
2
10−3
H1 Data
PT
hadrons (GeV)
dσ/
dP
Thad
ron
s
10
(p
b/G
eV)
−2
SM (GRAPE)
10−1
0 10 20
• Extracted from “γγ”sample
• Good Agreement withSM
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.50/92
Multi-electrons at H1: Overview
Selection DATA SM GRAPE NC-DIS + Compton
Visible 2e 105 118.2±12.8 * 93.3±11.5 25.0±5.5
Visible 3e 16 21.6±3.0 21.5±3.0 0.1±0.1
Visible 4e or more 0 0.1±0.0 0.1±0.0 0.0±0.0
γγ → e+e− subsample 41 48.3±6.1 46.4±6.1 1.9±0.9
Visible 2e M(12) >100 3 0.25±0.05 0.21±0.04 0.04±0.03
Visible 3e M(12) >100 3 0.23±0.04 0.23±0.04 0.00±0.00
* Statistical ⊕ Systematic Uncertainty
DATA agree with SM at low M12
Excess in DATA at high M12
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.51/92
Multi-electrons at ZEUSData taking: 1994-00 Lumi used: 130.5 pb−1
Event Selection• “Good Vertex”: |Zvtx| < 50 cm
• Two “central” (17 < θ < 164) electrons: Ee1T > 10 GeV, Ee2 > 10 GeV
A “2e” event
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.52/92
Multi- e at ZEUS: electron variables
ZEUS
10-2
10-1
1
10
10 2
0 20 40 60 80 100ET,1 (GeV)
even
ts ZEUS (prel.) 94-00GRAPE+NC+QEDCNC+QEDCQEDC
a)
10-1
1
10
10 2
0 1 2 3θe,1 (rad)
even
tsb)
10-2
10-1
1
10
10 2
0 20 40 60 80 100ET,2 (GeV)
even
ts c)
10-1
1
10
10 2
0 1 2 3θe,2 (rad)
even
ts
d)
Good Agreement
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.53/92
Multi- e at ZEUS: Mass distributionZEUS
10-2
10-1
1
10
10 2
0 20 40 60 80 100 120 140M12 (GeV)
even
ts
ZEUS (prel.) 94-00
GRAPE+NC+QEDC
NC+QEDC
QEDC
Two events with M12 > 100 GeV; expected 1.2±0.1Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.54/92
Multi- e at ZEUS: Overview
Type Data SM GRAPE NC-DIS Compton
2e sample
2e 191 213.9±3.9 * 182.2±1.2 23.9±3.7 7.8±0.5
Ee1T > 30 GeV 6 5.7±0.3 4.4±0.2 0.9±0.2 0.4±0.1
M12 > 100 GeV 2 0.77±0.08 0.47±0.05 0.12±0.06 0.18±0.03
3e sample
3e 26 34.7±0.5 34.7±0.5 - -
Ee1T > 30 GeV 2 1.43±0.08 1.43±0.08 - -
M12 > 100 GeV 0 0.37±0.04 0.37±0.04 - -
* Only Statistical Error
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.55/92
Di-Muons at H1
Data taking: 1999-00Lumi used: 70.9 pb−1
Muon Selection• Reconstructed track in both CTD
and Muon Detectors
• Angular region: 20 < θ < 160
• For low momentum muons:CTD Track + Calorimeter MIP
Run 276837 Event 89128 Class: 8 10 16 20 22 23 24 27 29 Date 6/02/2002
Z
R
X
Y
-180.
-90.
0.
90.
180. -4.-2.
0.2.
4.
7
3.5
0
E[G
eV]
(DC
LU)
ϕ[degree]
η
Event Selection• Two muons: P µ1
T > 2.00 GeV, P µ2T > 1.75 GeV
• Invariant mass cut: Mµµ > 5 GeV
• Muon Isolation: DµTrk,jet > 1.0 in ηφ (or Dµ
Trk,jet > 0.5 if P µT > 10 GeV)
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.56/92
Di-Muons at H1: Cross section [
pb
/GeV
]µ,µ
/dM
σd
10-4
10-3
10-2
10-1
1
10
[p
b/G
eV]
µ,µ/d
Mσ
d
10-4
10-3
10-2
10-1
1
10H1 DATA (Prelim.) SM (GRAPE)
(LPAIR)γ γµµ → ττ → γ γ
c + cbbµµ → Υµµ → 0Z
[GeV]µ,µM
S
Mσ
) /
SM
σ -
σ( -1
0
1
2
3
10 10020 506
H1 Muon Pair Analysis
[p
b/G
eV]
µ t/d
Pσ
d
10-3
10-2
10-1
1
10
[p
b/G
eV]
µ t/d
Pσ
d
10-3
10-2
10-1
1
10
H1 DATA (Prelim.) SM (GRAPE)
(LPAIR)γ γµµ → ττ → γ γ
c + cbbµµ → Υµµ → 0Z
[GeV]µtP
10 20 30 40 50 60 70 80
S
Mσ
) /
SM
σ -
σ( -1
0
1
2
3
H1 Muon Pair Analysis
•M
ain
cont
ribut
ion:γγ
inte
ract
ion
•S
mal
lcon
trib
utio
ntoµ
+µ−
from
:Υ
andττ
deca
ys
Total Cross–Section: σ = 46.5±1.3±4.7 pb
Good Agreement with SM: σ(GRAPE) = 46.2 pbMulti-leptons at HERA – MVD tracking at ZEUS TLT – p.57/92
Di-Muons at H1: Cross section [
pb
/GeV
]µ,µ
/dM
ela
σd
10-4
10-3
10-2
10-1
1
10H1 DATA (Prelim.)
SM (GRAPE)
[GeV]µ,µM20 40 60 80 100 120
S
Mσ
) /
SM
σ -
σ( -1
0
1
2
3
) µµ ep→H1 Muon Pair Analysis (ep
Elastic
[p
b/G
eV]
µ,µ /d
Min
elσ
d
10-4
10-3
10-2
10-1
1
10H1 DATA (Prelim.)
SM (GRAPE)
[GeV]µ,µM20 40 60 80 100 120
S
Mσ
) /
SM
σ -
σ( -1
0
1
2
3
X)µµ e→H1 Muon Pair Analysis (ep
Inelastic
•E
last
ican
dIn
elas
ticse
para
ted
byta
ggin
gpr
oton
rem
nant
Inelastic Cross–Section: σinel = 20.8±0.9±3.3 pb
Good Agreement with SM: σinel(GRAPE) = 21.5 pbMulti-leptons at HERA – MVD tracking at ZEUS TLT – p.58/92
Di-Muons at ZEUS
Data taking: 1997-00Lumi used: 105.2 pb−1
Muon Selection• Track in CTD (PT > 5 GeV)+ MIP in CAL
• Angular region: 20 < θ < 160
ZEUS
10-2
10-1
1
10
10 2
20 40 60 80 100 120 140 160 180Mµµ (GeV)
even
ts
ZEUS (prel.) 97-00
GRAPE
Event Selection• Two muons: 1µ matched to muon chambers
• Muon Isolation: Ntrks(Rηφ < 1) = 0
• Good Vertex: |Zvtx| < 40 cm,q
X2vtx + Y 2
vtx < 0.5 cm
• Acollinearity: cos(Ω) > −0.995
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.59/92
Multi-electron search
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.60/92
The electron taggingThe EM electron finder analyses energy deposits in the CAL to
distinguish electromagnetic from hadronic clusters.Sketch of the algorithm:
• Calorimeter cells are grouped in clusters;
• 4 quantities are evaluated for each cluster:• fraction of hadronic energy,• fraction of electromagnetic energy outside two highest-energy modules,• fraction of energy outside two highest-energy modules,• non-electron energy in a R=0.8 cone in ηφ;
• when a CTD track matches the cluster, are evaluated:• θTrk − θCal,• φTrk − φCal,• 1/PTrk − 1/ECal;
• a probability is computed using the above defined quantities.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.61/92
Variable reconstruction
Very nice correlation of true and reconstructed variables:
0
0.5
1
1.5
2
2.5
3
0 1 2 30
0.5
1
1.5
2
2.5
3
0 1 2 3
0
0.5
1
1.5
2
2.5
3
0 1 2 3
GRAPE MC
θe1 (rad) - true
θ e1 - reco
θe2 (rad) - true
θ e2 - reco
φe1 (rad) - true
φ e1 - reco
φe2 (rad) - true
φ e2 - reco
0
0.5
1
1.5
2
2.5
3
0 1 2 3
0
5
10
15
20
25
30
35
40
0 10 20 30 400
5
10
15
20
25
30
35
40
0 10 20 30 40
-1-0.8-0.6-0.4-0.2
00.20.40.60.8
1
-1 -0.5 0 0.5 1
GRAPE MC
PTe1 (GeV) - true
PTe1 - reco
PTe2 (GeV) - true
PTe2 - reco
cosΩ - true
cos
Ω -reco
M12 (GeV) - true
M12 - reco
0102030405060708090100
0 20 40 60 80 100
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.62/92
Variable reconstruction
Resolutions:
• σ(θ)/θ = 0.5%;
• σ(PT )/PT = 10%;
• σ(cos Ω)/ cos Ω = 2%;
• σ(M12)/M12 = 10%.
Excellent resolution for angles.
Good resolution for energies.
0
5
10
15
20
-0.02 0 0.02 0.04 0.060
5
10
15
20
-0.02 0 0.02 0.04 0.06
0
5
10
15
20
-0.5 0 0.5 10
5
10
15
20
-0.5 0 0.5 1
0
5
10
15
20
-0.05 0 0.05 0.1
GRAPE MCIDEntriesMeanRMS
622 45309
0.1249E-02 0.5522E-02
(θe1,reco-θe1e,true)/θe1,true
IDEntriesMeanRMS
623 45309
0.6508E-03 0.5317E-02
(θe2,reco-θe2,true)/θe2,trueIDEntriesMeanRMS
620 45309
0.4555E-01 0.9085E-01
(Pe1
T,reco-Pe1
T,true)/Pe1
T,true
IDEntriesMeanRMS
621 45309
-0.8825E-02 0.1027
(Pe2
T,reco-Pe2
T,true)/Pe2
T,true
IDEntriesMeanRMS
650 45309
0.1214E-02 0.1733E-01
(cosΩreco-cosΩtrue)/cosΩtrue
IDEntriesMeanRMS
617 45309
0.1739E-01 0.7944E-01
(M12,reco-M12,true)/M12,true
0
10
20
-0.5 0 0.5 1
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.63/92
Trigger chainDi-electron chain
• Elastic di-e: good vertex + few CTD tracks + 2 electrons.
• Inelastic di-e: good vertex + 2 electrons + high mass.
Neutral Current DIS chain
• good vertex + 1 electron + large ET + large Q2 + large
E − Pz.
High-ET chain
• high ET in the CAL.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.64/92
Data to MC comparison: 2e
1996-2000 data taking L=120.49 pb−1
Selection Data All SM di-ele MC NC-DIS MC QED-C MC
All 2e 183 197.78 169.11 18.57 10.10
M12 > 50 GeV 17 27.47 22.68 3.32 1.47
M12 > 100 GeV 1 0.67 0.46 0.08 0.13
10-2
10-1
1
10
102
0 20 40 60 80 100
10-2
10-1
1
10
0 1 2 3
10-3
10-2
10-1
1
10
102
0 20 40 60 80 100
ZEUS 2e
ETe1 (GeV)
ZEUS 1996-00GRAPE+NC+QEDCNC+QEDCQEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
1
10
0 1 2 3
10-2
10-1
1
10
102
0 50 100 150
10-1
1
10
-2 0 2
10-1
1
10
102
0 1 2 3
ZEUS 2e
M12 (GeV) abs(θe1-θe2) (rad)
φe1-φe2 (rad)
ZEUS 1996-00GRAPE+NC+QEDCNC+QEDCQEDC
(E-Pz)ee (GeV)
10-2
10-1
1
10
0 20 40 60
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.65/92
Data to MC comparison: 2e
1996-97 period
10-2
10-1
1
10
0 20 40 60 80 100
10-2
10-1
1
10
0 1 2 3
10-3
10-2
10-1
1
10
0 20 40 60 80 100
ZEUS 2e
ETe1 (GeV)
ZEUS 1996-97GRAPE+NC+QEDCNC+QEDCQEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
10-1
1
10
0 1 2 3
1998-99 period
10-3
10-2
10-1
1
10
0 20 40 60 80 100
10-3
10-2
10-1
1
10
0 1 2 3
10-3
10-2
10-1
1
10
0 20 40 60 80 100
ZEUS 2e
ETe1 (GeV)
ZEUS 1998-99 (e-)GRAPE+NC+QEDCNC+QEDCQEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
10-3
10-2
10-1
1
10
0 1 2 3
1999-2000 period
10-2
10-1
1
10
0 20 40 60 80 100
10-2
10-1
1
10
0 1 2 3
10-1
1
10
0 20 40 60 80 100
ZEUS 2e
ETe1 (GeV)
ZEUS 1999-00 (e+)GRAPE+NC+QEDCNC+QEDCQEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
1
10
0 1 2 3
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.66/92
Data to MC comparison: 3e
1996-2000 data taking L=120.49 pb−1
Selection Data All SM di-ele MC NC-DIS MC QED-C MC
All 3e 26 38.03 37.88 0.15 0.00
M12 > 50 GeV 6 7.54 7.46 0.08 0.00
M12 > 100 GeV 0 0.44 0.44 0.00 0.00
10-1
1
10
0 20 40 60 80 100
10-1
1
0 1 2 3
10-1
1
10
0 20 40 60 80 100
ZEUS 3e
ETe1 (GeV)
ZEUS 1996-00
GRAPE+NC+QEDC
NC+QEDC
QEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
10-1
1
0 1 2 3
10-1
1
10
0 50 100 150
10-1
1
-2 0 2
10-1
1
10
0 1 2 3
ZEUS 3e
M12 (GeV) abs(θe1-θe2) (rad)
φe1-φe2 (rad)
ZEUS 1996-00
GRAPE+NC+QEDC
NC+QEDC
QEDC
(E-Pz)ee (GeV)
10-1
1
10
0 20 40 60
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.67/92
Data to MC comparison: 3e
1996-97 period
10-1
1
0 20 40 60 80 100
10-1
1
0 1 2 3
10-1
1
0 20 40 60 80 100
ZEUS 3e
ETe1 (GeV)
ZEUS 1996-97
GRAPE+NC+QEDC
NC+QEDC
QEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
10-1
1
0 1 2 3
1998-99 period
10-2
10-1
1
0 20 40 60 80 100
10-2
10-1
1
0 1 2 3
10-2
10-1
1
0 20 40 60 80 100
ZEUS 3e
ETe1 (GeV)
ZEUS 1998-99 (e-)GRAPE+NC+QEDCNC+QEDCQEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
10-2
10-1
1
0 1 2 3
1999-2000 period
10-1
1
10
0 20 40 60 80 100
10-1
1
0 1 2 3
10-1
1
10
0 20 40 60 80 100
ZEUS 3e
ETe1 (GeV)
ZEUS 1999-00 (e+)
GRAPE+NC+QEDC
NC+QEDC
QEDC
θe1 (rad)
ETe2 (GeV) θe2 (rad)
10-1
1
0 1 2 3
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.68/92
Cross section: Method
• Ng : MC events generated;
• Nr : MC events reconstructed;
• Nrg : MC events generated AND
recontsructed;
• Nd: events selected in data.
Definitions• efficiency: e = Nr
g /Ng ;
• purity: p = Nrg /Nr ;
• acceptance: a = Nr/Ng .
Cross sectionTotal:
σMC =Ng
L
σDATA =Nd
L a
Differential:
dσMC
dx=
Ng
L ∆x
dσDATA
dx=
Nd
L a ∆xMulti-leptons at HERA – MVD tracking at ZEUS TLT – p.69/92
Cross section: combination of periods
Correction for√
s
To refer cross-section to√s = 318 GeV:
σ318x =
σMC99−00
σMCx
σDATAx
x = 1996-97 or 1998-99
Combination of periods
σ96−00 =
∑
x σ318x Lx
∑
x Lx
x = 1996-97, 1998-99 or1999-2000
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.70/92
Systematic uncertainties
Type Variation Effect
Lumi measurement 2% 2%
CAL energy scale 5% +9.4−4.1%
Prob. cut in EMPgrand>0.01→0.1
Pcal>0.1→0.2 −4.8%
Isolation cut Econe < 0.3 → 0.2 GeV +7.3%
Mass cut 8% 0%
Total +13−10%
Statistical uncertainty: ±14%.
-1
-0.5
0
0.5
1
10 20 30 40 50 60 70 80 90
-1
-0.5
0
0.5
1
10 20 30 40 50 60 70 80 90
Variation
Variation
M12 (GeV)
Variation
-1
-0.5
0
0.5
1
10 20 30 40 50 60 70 80 90
UCAL Energy Scale
Probability cut in EM
Isolation cut
Systematic UncertaintyStatistical Uncertainty
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.71/92
Cross section: Tables
Bin definition Pur. Acc. dσDATAdx
(pb/[x]) dσMCdx
(pb/[x])
15 < M12 < 25 GeV 0.723 0.309 (4.2 ± 1.1 ± 1.0)E−02 5.7E−02
25 < M12 < 40 GeV 0.840 0.330 (5.2 ± 1.0 ± 0.6)E−02 4.5E−02
40 < M12 < 60 GeV 0.803 0.206 (1.50 ± 0.57+0.34−0.32)E−02 1.38E−02
10 < P eT < 15 GeV 1.154 0.223 (2.20 ± 0.42+0.28
−0.20)E−01 1.94E−01
15 < P eT < 20 GeV 1.073 0.236 (5.6 ± 2.1+1.1
−1.0)E−02 4.9E−02
20 < P eT < 25 GeV 0.924 0.239 (1.13 ± 0.93+0.45
−0.43)E−02 1.69E−02
0.349 < θe < 1.105 rad 0.831 0.381 (6.6 ± 1.4+1.0−0.9)E−01 7.2E−01
1.105 < θe < 1.862 rad 0.828 0.334 (5.1 ± 1.4 ± 0.6)E−01 5.6E−01
1.862 < θe < 2.618 rad 0.681 0.198 (8.1 ± 2.2+1.2−1.0)E−01 6.8E−01
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.72/92
Di-muon search
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.73/92
Muon tagging: GLOMUThe GLOMU package tags muons in the barrel/rear region.
It combines info from inner B/RMUON, CTD, CAL.
Sketch of the algorithm:
• CTD tracks with high momentum and close to vertex are collected;
• B/RMUI tracks are collected;
• MIP-like deposits are searched in the CAL;
• a match is attempted in θ-φ for these objects;
• a cut χ2 < 20 is set to accept a match.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.74/92
Muon tagging: MPMATCH2 and MIPsThe MPMATCH2 package tags muons in the forward region.
It combines info from inner FMUON and CTD.Sketch of the algorithm:
• tracks are reconstructed in the FMU detector;
• these tracks are backwards propagated to the CAL surface;
• CTD tracks are extrapolated to the same surface;
• a match is attempted of the two sets of tracks;
• track parameters are refitted.
To raise efficiency in muon finding, a matching of CTD tracks and
MIP-like deposits in the CAL is attempted.
• CTD tracks are extrapolated to the CAL and matched to MIP-likedeposits.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.75/92
Variable reconstruction
Very nice correlation of true and reconstructed variables:
0
0.5
1
1.5
2
2.5
3
0 1 2 30
0.5
1
1.5
2
2.5
3
0 1 2 3
0
0.5
1
1.5
2
2.5
3
0 1 2 3
GRAPE MC
θµ1 (rad) - true
θ µ1 - reco
θµ2 (rad) - true
θ µ2 - reco
φµ1 (rad) - true
φ µ1 - reco
φµ2 (rad) - true
φ µ2 - reco
0
0.5
1
1.5
2
2.5
3
0 1 2 3
02.5
57.510
12.515
17.520
22.525
0 5 10 15 20 250
2.55
7.510
12.515
17.520
22.525
0 5 10 15 20 25
-1-0.8-0.6-0.4-0.2
00.20.40.60.8
1
-1 -0.5 0 0.5 1
GRAPE MC
PTµ1 (GeV) - true
PTµ1 - reco
PTµ2 (GeV) - true
PTµ2 - reco
cosΩ - true
cos
Ω -reco
Mµµ (GeV) - true
Mµµ - reco
05101520253035404550
0 10 20 30 40 50
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.76/92
Variable reconstruction
Resolutions:
• σ(θ)/θ = 0.3%;
• σ(PT )/PT = 10÷ 15%;
• σ(cos Ω)/ cos Ω = 1%;
• σ(Mµµ)/Mµµ = 10%.
Excellent resolution for angles.
Good resolution for energies.
0
20
40
60
-0.02 0 0.020.040.060
20
40
60
-0.02 0 0.020.040.06
010203040
-0.5 0 0.5 10
20
40
-0.5 0 0.5 1
0
20
40
60
-0.05 0 0.05 0.1
GRAPE MCIDEntriesMeanRMS
622 59166
0.8319E-04 0.3287E-02
(θµ1,reco-θµ1,true)/θµ1,true
IDEntriesMeanRMS
623 59166
0.6389E-04 0.2935E-02
(θµ2,reco-θµ2,true)/θµ2,true
IDEntriesMeanRMS
620 59166
0.5338E-01 0.1417
(Pµ1T,reco-P
µ1T,true)/P
µ1T,true
IDEntriesMeanRMS
621 59166
-0.2024E-01 0.9417E-01
(Pµ2T,reco-P
µ2T,true)/P
µ2T,true
IDEntriesMeanRMS
650 59166
0.9365E-03 0.1277E-01
(cosΩreco-cosΩtrue)/cosΩtrue
IDEntriesMeanRMS
617 59166
0.2000E-01 0.1123
(Mµµ,reco-Mµµ,true)/Mµµ,true
0
20
40
-0.5 0 0.5 1
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.77/92
Trigger chainB/RMUON chain
• Inner chambers trigger: (few CTD tracks + B/RMUI signal)
OR (B/RMUI signal matched to CTD tracks and MIP-like
deposits).
• Outer chambers trigger: Signal in B/RMUI and B/RMUO +
CAL activity and CTD tracks
• Both chains: a GLOMU matching
FMUON chain
• few CTD tracks + matching (FMU and MIP-like deposits) +
two CTD tracks with high invariant mass
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.78/92
B/RMUI Efficiencies
B/RMU inefficiencies are not simulated in MC
Efficiencies were measured (M. Turcato, A. Bertolin, M. Corradi)
MethodDi-muon events are selected (J/ψ,Bethe-Heitler, ...) by
• A hit in muon chambers
• 2 CTD tracks withM > 2.5 GeV
• Acollinearity: cosΩ > −0.95
The efficiency is
e = No. of GLOMU tracks
No. of CTD tracks
ResultsThe ratio R =
e(DAT A)e(MC)
is used to weight
Monte Carlo samples
Region Data/MC ratio
Years 1996-97 Years 1998-2000
BMUI 0.854 ± 0.061 0.709 ± 0.042
RMUI 0.819 ± 0.068 0.853 ± 0.050
BMUI cut: 4.25 < P µT < 10 GeV
RMUI cut: 5 < P µT < 10 GeV
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.79/92
Di-muon events
Three events found
around Mµµ = 50 GeV
All of them look like gen-
uine di-muons
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.80/92
Systematic uncertainties
Type Variation Effect
Lumi measurement 2% 2%
B/RMUI Efficiency 7.5% in 1996−976.0% in 1998−00
+5.4−4.8%
cos Ω cut cosΩ > −0.995 → −0.985 +2.8%
Isolation cut DµTrk > 1 → 2 +4.0%
Prob. cut χ2<20→10 (GLOMU)P >0.01→0.05 (MPMATCH2)
−0.4%
Total +7.2−6.6%
Statistical uncertainty: ±6.9%.
-1-0.8-0.6-0.4-0.2
00.20.40.60.81
20 40 60 80-1
-0.8-0.6-0.4-0.2
00.20.40.60.81
20 40 60 80
-1-0.8-0.6-0.4-0.2
00.20.40.60.81
20 40 60 80
Variation
Variation
Mµµ (GeV)
Variation
Mµµ (GeV)
Variation
-1-0.8-0.6-0.4-0.2
00.20.40.60.81
20 40 60 80
Ωcos( ) Cut
Prob. Cut
B/RMUI Efficiency
Isolation Cut
Systematic uncertainty
Statistical uncertainty
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.81/92
Cross section: Tables
Bin definition Pur. Acc. dσDATAdx
(pb/[x]) dσMCdx
(pb/[x])
10 < Mµµ < 20 GeV 0.968 0.437 (3.44 ± 0.28+0.24−0.22)E−01 3.70E−01
20 < Mµµ < 30 GeV 0.977 0.412 (8.0 ± 1.4+0.8−0.7)E−02 10.7E−02
30 < Mµµ < 45 GeV 0.928 0.388 (2.02 ± 0.58+0.32−0.31)E−02 2.40E−02
45 < Mµµ < 100 GeV 0.787 0.437 (2.04 ± 0.91 ± 0.65)E−03 1.90E−03
5 < P µT < 6.5 GeV 0.949 0.401 1.53 ± 0.16+0.11
−0.10 1.55
6.5 < P µT < 8 GeV 1.009 0.437 (7.6 ± 1.1+0.6
−0.5)E−01 8.3E−01
8 < P µT < 13 GeV 0.994 0.447 (2.17 ± 0.31+0.23
−0.22)E−01 2.72E−01
13 < P µT < 50 GeV 0.885 0.473 (7.9 ± 2.1+1.3
−1.2)E−03 10.3E−03
0.262 < θµ < 0.912 rad 0.963 0.339 2.36 ± 0.42+0.18−0.17 2.81
0.912 < θµ < 1.562 rad 0.977 0.523 1.65 ± 0.22+0.13−0.12 1.83
1.562 < θµ < 2.212 rad 0.970 0.516 1.83 ± 0.23+0.14−0.13 1.69
2.212 < θµ < 2.862 rad 0.963 0.385 1.41 ± 0.24+0.14−0.13 1.84
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.82/92
MVD tracking at ZEUS TLT
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.83/92
Track parameterisation
DH
φ1
Hφ
(x ,y )0 0
(0,0) x
y
R
s=0
(Q=+1)
(α,β)
Coordinates
x(φ) = x0 +
(
a3 +1
a2
)
sin a1 −sinφ
a2
y(φ) = y0 −(
a3 +1
a2
)
cos a1 +cosφ
a2
z(φ) = a4 −a5
a2
(φ− a1)
Pathlegth
s = −φ− a1
a2
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.84/92
The ZEUS MVD: design
Sensor design
300
1
14 12
20
120
12
1
1,5
n-substrate
p+ inter-strip
p+ readout-strip
backplane(aluminum)
n+
aluminum-readout-
strip
SiO2
Si3N4
• 64x64 mm n-type siliconsensors
• p+ strips every 20 µm
• a strip over 6 is read-out
BMVD designr- ϕ readoutr- ϕ readout
electronicselectronics
to front-end
60 mm
60 m
m
r-z readoutr-z readout
to front-end
512 channels 512 channels
• Two sensors form a“half-module”
• Two half-modules form a“module”, 2.2% X0 thick
• A module measures rφ and z
coordinates
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.85/92
The ZEUS MVD: performancesTest beam
Cluster resolution on a single detec-
tor (6 GeV beam), with different re-
constuction algorithms.
Example: c-quark selection
Using D∗ tagging:
eff ≃ 1% @ pur ≃ 30%
Using MVD (simulation):
eff ≃ 10–20% @ pur ≃ 30%
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.86/92
The clustering procedureProcedure
• Pedestal and common mode
are subtracted;
• Consecutive strips above
threshold are clustered;
• Centre and width: accordingly
to centre of gravity:
〈ξ〉 =1
Q
∑
i
Qiξi
σ2(ξ) =1
Q
∑
i
Qi(ξi − 〈ξ〉)2
Results
IDEntriesMeanRMS
1000000 2000
1.061 0.1247
2003/10/25 16.31
BMVD clusters / MC hits
Evt
s
0
200
400
600
800
1000
1200
1400
0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2
IDEntriesMeanRMS
3201 12781 0.8390E-02 0.1683E-02
2003/10/25 16.31
Clusters Width (cm)
Evt
s
0
500
1000
1500
2000
2500
3000
3500
4000
4500
0 0.005 0.01 0.015 0.02 0.025 0.03
Left: Eff =
(No. of meas. hits)/(No. of GEANT hits)
Right: Cluster widths
Results obtained on a single µ MC
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.87/92
Multiple scattering
Dead materials:
• Ladders (rectangular boxes);
• Support tube (cylinder);
• Beam pipe (ellipsoidal cylinder).
D is the traversed material (g/cm2)
Scattering angle
δ2D =0.0136
p/GeV
√D (1 + 0.0038 lnD)
δRMS =δ2D
| sin θ|Effect on covariance matrix
cov(a1, a1) → cov(a1, a1) + δ2RMS
cov(a1, a3) → cov(a1, a3)+Q s δ2RMS
cov(a3, a3) → cov(a1, a3)+(s δRMS)2
Minor effect on the other covariances
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.88/92
Energy lossBethe-Bloch Formula:
−dE
dx= Kz2Z
A
1
β2
[
1
2ln
2mec2β2γ2Tmax
I2− β2 − δ
2
]
βγ
−d
E/d
x (
Ge
V g
−1 c
m2)
W/O Density Corrections
With Density Corrections
10−3
10−2
10−1
1 10 102
103
104
where:
• K = 3.07075 · 10−4 GeV cm2 mol−1
• Incident particle:• z is the charge number• M is the mass• βc is the particle speed• γ = 1/
√1 − β
• Tmax = 2mec2β2γ2
1+2γme/M
• Medium:• Z/A: atomic / mass number• I is the mean excitation energy• δ is the density effect correction
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.89/92
Pattern Recognition
Cluster 1
Cluster 2
Cluster 3
Lad
der
Beam−Pipe
• Each helix is propagated inwards;
• intersections with ladders are found;
• the closest BMVD hit is collected;
• track parameters are updated;
• the procedure is iterated for the
remaining layers;
• external layer only: hits are preferrd
which leads to hits in the internal lay-
ers.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.90/92
Multiple assignmentT
rack
1
Tra
ck 2
Cluster 1
Cluster 2
Ladder
The same hit may be assigned to
more tracks.
If one hit is present:
• hit is assigned to the closest track
If two or more hits are present:
• assignment is done to minimise
(Trk1 − Clun)2/σ2 + (Trk2 − Clum)2/σ2
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.91/92
Parameter FitMeasurement
• Fm ±σm: measurement No. m
• f(m;~a): expected position
Before fit
χ20 of old fit (minimised wrt ~a).
A measurement Fm is added.
• χ2(~a) = χ20 + [Fm−f(m;~a)]2
σ2m
• Bµ(~a) = −12
∂χ2(~a)∂aµ
• Uµν(~a) = −Bµ(~a)∂aµ
χ2(~a) is no more minimised.
χ2 Minimisation
Parameter ~a is updated by minimis-
ing χ2: ∂χ2(~a+δ~a)∂aµ
= 0
δaµ = U−1µν (~a) Bν(~a)
After fit
• ~anew = ~a+ δ~a;
• Bµ(~anew) = 0;
• Uµν(~anew) = Uµν(~a);
• χ2(~anew) = χ2(~a) −Bµ(~a)δaµ.
Multi-leptons at HERA – MVD tracking at ZEUS TLT – p.92/92