Top Physics with the ATLAS detector

1
Top Physics with the ATLAS detector
Arnaud Lucotte (IN2P3/LPSC Grenoble) on behalf
of the ATLAS collaboration
Outline Motivation
Top pair production Mass measurement
Cross-section measurement Single-top production
Single-top cross-section Search for
charged higgs Conclusion
2
Top Quark Physics Motivations

• Top Quark a tool for precise EW Sector studies
• A special role in the EW sector
• Top, W and Higgs masses enter radiative
  corrections in theoretical calculations of many
  observables
• ? Precision on (mt,mW) constrains mH
• Heaviest elementary particle known
• ? Yukawa couplings close to 1.0
• CKM Matrix element Vtb
• A special role in QCD the quark family
• Test of QCD
• Vast swath of phase space available to the decay
• ? short lifetime (lt tQCD28x10-25 s)
• ? Window on properties of bare quark
• Top Quark a probe to new physics
• A special role in (all) various extensions of the
  SM
• Searches for new (heavy) particles
• ? flavor/mass dependent couplings
• ? extra-bosons W,
• ? Higgs boson H
• A link to the (new) theory fundamentals

3
Top pair production Top mass and cross-section
Courtesy C. Timmermans
4
Top pair production decay

• Top pair production
• SM Total cross-section
• NLO calculations stt 835 pb 10pdf 6
  µ-scale
• Production via gluon-fusion (90) and quark
  anihilation
• Dependence in Top Mass dstt/stt 5 x dmt/mt
  
• Top pair decay modes
• SM Branching ratio
• BR(t?Wb) 1

s (fb)
10
90
mt (GeV/c2)
leptonjets
tt final state BR (pb) Nevt (10 fb-1)
tt ? (lv)b (jj)b 30 2.5x 106
tt?(lv)b(lv)b 5 400,000
tt?(jj)b (jj)b 44 3.7x 106
di-lepton
full-hadronic
5
1 or 2 weeks of the ATLAS data

• Top pair events in 300 pb-1
• Selection
• Missing ET, 1 lepton, 4 jets , NO b-tag !!
• ? efficiency 5.3
• ? Apply extra cut on hadronic W mass
• Expected performance

Atlas FullSim Preliminary
Atlas FullSim Preliminary
tt Wjets
tt Wjets
S/B 0.45
S/B 1.77
mjj-mWlt10 GeV/c2
MC_at_NLO
mjjb (GeV/c2)
mjjb (GeV/c2)
6
Top Mass measurements
7
Top Mass measurement motivations ...

• Precision measurements in the EW sector
• Boson masses relation
• Present Measurements
• ? constraint Higgs Mass
• mH lt 285 GeV/c2 (95CL)
• ? do not indicate any clear
  departure from SM (yet)
• ? equal weight to ?2 from
• ?mW 0.7 ?mt

hep-ph/0307177 LEPWWG05
15 MeV ? 2 GeV
s-top mass
mh2 mZ2 3GF mt4 ln Mt2
p2v2 mt2
hep-ph/0303092
8
Top Mass measurement motivations ...

• Precision measurements in the EW sector
• Boson masses relation
• Present Measurements
• ? constraint Higgs Mass
• mH lt 285 GeV/c2 (95CL)
• ? do not indicate any clear
  departure from SM (yet)
• ? equal weight to ?2 from
• ?mW 0.7 ?mt

hep-ph/0307177 LEPWWG05
?mt1 GeV/c2 ?mW15 MeV/c2
9
Top Mass using leptonjets Event Selection

• Strategy
• (1) leptonic top to tag the event
• 1 high-pT lepton
• high missing Energy
• at least 4 high-pT jets
• at least 1 high-pT b-tagged jet
• (2) hadronic top to measure the mass
• Identify b-quark jets
• ? 2 samples 2 btag (1 btag)
• W-boson reconstruction from jj
• ? in-situ light jet calibration
• Top quark reconstruction from jjb
• Event yields _at_ 10 fb-1

Processus s x BR (pb) e() Nevents
bb?lvjets 2.2 x 106 3x10-8 15
Wjets? lv jets 7.8 x 103 2x10-4 930
Zjets ?ll-jets 1.2 x 103 6x10-5 150
WZ ? lv jets 3.4 1x10-2 12
WW ? lv jets 17.1 7x10-3 10
ZZ ? ll- jets 9.2 5 5
Tt ? (lv)b (jj)b 250 3.5 87,000
Main background from wrong combinations in tt
events
10
Top Mass using leptonjets b-tagged jet

• b-tagging algorithms
• Several taggers
• Soft lepton tag
• Impact parameter 2D/3D
• Secondary Vertex
• Mass, charged fraction...
• Lifetime tag ...
• ? Combined likelihood/weight
• Performance

Atlas Preliminary
SV1IP3D jet weight
IP2D jet weight
b-jet
b-jet
u-jet
u-jet
Atlas Preliminary
jet weight
jet weight
Ru (50) Ru(60)
IP2D 160 55
IP3D 230 85
SV1IP3D 500 185
11
Top Mass using leptonjets b-tagged jet

• b-tagging algorithms
• Several taggers
• Soft lepton tag
• Impact parameter 2D/3D
• Secondary Vertex
• Mass, charged fraction...
• Lifetime tag ...
• ? Combined likelihood/weight
• Performance

Atlas Preliminary
etag vs rejection (light jet)
rejection vs pT
?
?
?
SV1IP3D
light-jet rejection
light jet rejection
IP3D
etag 60
?
SV1IP3D
?
?
?
?
?
IP3D
?
IP2D
?
?
?
?
?
?
Atlas Preliminary
?
?
light-jet pT(GeV/c)
b-jet efficiency
Ru (50) Ru(60)
IP2D 160 55
IP3D 230 85
SV1IP3D 500 185
12
Top Mass using leptonjets W-boson
reconstruction

• W-boson reconstruction
• Jet association
• Select (jj)-pair with minimum mjj mW
• Sample W-purity 66 (55) w/ efficiency
  3.2
• Light Jet calilbration
• In situ calibration with of W?jj
• Correct for jet energy
• Correct for jet direction

hep-ex/0403021
Nbtag 2
Nbtag 1
s 7.4 GeV/c2
s 7.4 GeV/c2
13
Top Mass using leptonjets Top reconstruction

• Top Quark reconstruction
• Association of hadronic W and b-jet
• Combination leading to the highest pTtop
• or that maximizes ?R(l,b) / minimizes ?R(b,W?jj)
  
• ? Top Purity 69 (65) w/ efficiency 1.2
  (2.5)
• Event Yield
• 30K (80K) events in 2 b-tag (1b-tag)

no-calibration
re-calibrated
s 10.6 GeV
s 12.4 GeV
14
Top Mass using leptonjets systematic
uncertainties

• Top mass uncertainty
• Main systematics
• Energy scale
• Knowledge at the 1 level
• ? light jets makes use of an in-situ
  calibration
• ? b-jets use of Zb jets
• ISR / FSR
• ISR affect the number of jets N(jet)

hep-ex/0403021
sources of uncertainty dmt(GeV/c2)
light jets energy scale 0.2
b-jet energy scale 0.7
Initial State Radiation 0.1
Final State Radiation 1.0
b-quark fragmentation 0.1
Combinatorial backgd 0.1
Total SYSTEMATIC 1.3
Total STATISTICAL 0.07
? mt for a 1 miscalibration
15
Top Mass using leptonjets systematic
uncertainties

• Top mass uncertainty
• Main systematics
• Energy scale
• Knowledge at the 1 level
• ? light jets makes use of an in-situ
  calibration
• ? b-jets use of Zb jets
• ISR / FSR
• ISR affect the number of jets N(jet)

hep-ex/0403021
sources of uncertainty dmt(GeV/c2)
light jets energy scale 0.2
b-jet energy scale 0.7
Initial State Radiation 0.1
Final State Radiation 1.0
b-quark fragmentation 0.1
Combinatorial backgd 0.1
Total SYSTEMATIC 1.3
Total STATISTICAL 0.07
20 of ? mt for mt(ISR ON)-mt(ISR OFF) mt(FSR
ON)-mt(FSR OFF)
16
Top Mass using leptonjets systematic
uncertainties

• Top mass uncertainty
• Main systematics
• Energy scale
• Knowledge at the 1 level
• ? light jets makes use of an in-situ
  calibration
• ? b-jets use of Zb jets
• ISR / FSR
• ISR affect the number of jets N(jet)

hep-ex/0403021
sources of uncertainty dmt(GeV/c2)
light jets energy scale 0.2
b-jet energy scale 0.7
Initial State Radiation 0.1
Final State Radiation 1.0
b-quark fragmentation 0.1
Combinatorial backgd 0.1
Total SYSTEMATIC 1.3
Total STATISTICAL 0.07
change in Peterson eb -0.006 / -0.0035
17
Top Mass using leptonjets systematic
uncertainties

• Top mass uncertainty
• Main systematics
• Energy scale
• Knowledge at the 1 level
• ? light jets makes use of an in-situ
  calibration
• ? b-jets use of Zb jets
• ISR / FSR
• ISR affect the number of jets N(jet)

hep-ex/0403021
sources of uncertainty dmt(GeV/c2)
light jets energy scale 0.2
b-jet energy scale 0.7
Initial State Radiation 0.1
Final State Radiation 1.0
b-quark fragmentation 0.1
Combinatorial backgd 0.1
Total SYSTEMATIC 1.3
Total STATISTICAL 0.07
change in backgd shape
18
Top Mass using leptonjets systematic
uncertainties

• Top mass uncertainty
• Main systematics
• Improvements
• Use of a kinematic fit on the entire tt event
• ? reconstruct hadronic / leptonic top
• Use Constraints event / event
• mjj mW mlv mW
• mjjb mlvb ? (X2,mtfit)

hep-ex/0403021
sources of uncertainty dmt(GeV/c2) dmt(GeV/c2)
light jets energy scale 0.2 0.2
b-jet energy scale 0.7 0.7
Initial State Radiation 0.1 0.1
Final State Radiation 1.0 0.5
b-quark fragmentation 0.1 0.1
Combinatorial backgd 0.1 0.1
Total SYSTEMATIC 1.3 0.9
Total STATISTICAL 0.07 0.12
?mt 1 GeV/c2 seems achievable provided 1
calibration Jet-energy scale
Systematicslimited measurement b-jet
energy scale FSR are dominant
19
Top Mass in the di-lepton channel

• Procedure
• (1) Selection yield _at_ 10 fb-1
• 2 high-pT leptons
• high missing Energy
• 2 high-pT jets
• ? 80,000 evts S/B 10
• (2) Reconstruct fully tt event
• Assess neutrinos momenta
• ? 6 eqs (SpT0,MlvmW,Mlvbmt)
• ? e 97 w/ Purity 73
• (3) Top mass determination
• Evt/evt mt ? solve system ? weight
• (using kinematics topology)
• All evts mean weight per mt
• ? mtfit mt w/ highest ltweightgt
• Performance with 10 fb-1
• Mass resolution
• s 13 GeV/c2

e,µ
v
V. Simak et al.
dmt
b-jet energy scale (1) 0.6
b-quark fragmentation 0.7
ISR / FSR modelisation 0.6
Parton Distr. function 1.2
Total SYSTEMATIC 1.6
STATISTICS method 0.3
20
Top mass in the di-lepton channel

• Procedure
• (1) Selection yield _at_ 10 fb-1
• 2 high-pT leptons
• high missing Energy
• 2 high-pT jets
• ? 80,000 evts S/B 10
• (2) Reconstruct fully tt event
• Assess neutrinos momenta
• ? 6 eqs (Spx,v0,MlvmW,Mlvbmt)
• ? e 97 w/ Purity 73
• (3) Top mass determination
• Evt/evt mt ? solve system ? weight
• (using kinematics topology)
• All evts mean weight per mt
• ? mtfit mt w/ highest ltweightgt
• Performance with 10 fb-1
• Mass resolution
• s 13 GeV/c2

generated mt 175 GeV
mean probability
mass hypothesis
V. Simak et al.
dmt
b-jet energy scale (1) 0.6
b-quark fragmentation 0.7
ISR / FSR modelisation 0.6
Parton Distr. function 1.2
Total SYSTEMATIC 1.6
STATISTICS method 0.3
21
Top mass in the full hadronic channel

• Procedure
• (1) Selection yield _at_ 10 fb-1
• at least 6 central high-pT jets
• 2 high-pT b-tagged jets
• ? 100,000 evts S/B 1/19 (QCD)
• most challenging channel
• (2) Analysis
• Jet energy variables (HT, ETb, ?Rjj..)
• Event shape variables
• Kinematic fit to mt
• form W? jj (X2W) t? Wb (X2t)
• Performance with 10 fb-1
• Event yields
• Keep events with
• pTtop 200 GeV/c
• 130 lt mjjb lt 200
• Resolution s 13 GeV/c2
• ? Signal 3,300 events
• ? S/B 18/1

q
?
q
?
hep-ex/0403021
dmt
light jet energy scale 0.8
b-jet energy scale 0.7
Initial State Radiation Final State Radiation 0.4 2.8
b-quark fragmentation 0.3
Background 0.4
Total SYSTEMATIC 3.1
Total STATISTICAL 0.2
22
Top pair production cross-section measurement

• Cross-section measurement
• Strategy
• Same pre-selection as
• for mt measurements
• Performance
• Uncertainty dsstat negl.
• Systematics dominated
• machine ?L/L 5
• b-tagging e mistag rates
• ISR/FSR, pdf, Jet energy scale
• Interpretations
• Test of QCD
• (mt,stt) measurements
• ? ?s/s 6
• ?stt/stttheo 10
• ? ?mt 3 GeV/c2
• Sensitivity of dstt/dMtt
• High mass resonance

Atlas Preliminary
Nevent _at_ 1033 ?s/sstat
1 month 70,000 0.4
1 year 300,000 0.2
tanß
mass (GeV/c2)
23
Single-top cross-section measurements
Courtesy C. Timmermans
24
Single-top cross-section measurements

• Motivations
• Properties of the Wtb vertex
• Determination of s(pp?tX), ?(t?Wb)
• Direct determination of Vtb
• Test of V-A, top polarization
• Probe to new physics
• Anomalous couplings, FCNC
• Extra gauge-bosons W (GUT, KK)
• Extra Higgs boson (2HDM)
• Single-top is one of the main background to
• Higgs physics with jets

25
Single Top cross-section Production _at_ LHC

• Production at the LHC
• All 3 contributing mechanisms in SM
• 1 high-pT lepton mET
• 1 high-pT forward jet
• 1 high-pT b-quark jet
• 1 high-pT lepton mET
• 2 high-pT jet
• 1 high-pT b-quark jet
• 1 high-pT lepton mET
• 2 high-pT b-quark jet

t-channel
sNLO155(t)95(t) 2509 pb
Wt channel
sLO 60 15 pb
S-channel
sNLO 10 0.7 pb
?theo 4-8
26
Single Top Event Selection

• Procedure
• (1) Select and tag event
• 1 high-pT lepton
• high missing Energy
• at least 2 high-pT jets
• at least 1 high-pT b-tagged jet
• (2) Discriminate vs non-top background
• Reconstruct a Top mass Mlvb
• Use event shape high HT or MTOT
• (3) Discriminate vs top backgrounds
• Number of b-jets
• Event topology

e,µ
b
b
v
q
s x BR (pb)
Wg ?(lv)b qb 54.2
Wt ? (jj) (lv)b 17.8
W ? (lv)b b 2.2
Wjets? lvjets 3,850
WQQ? lvQQ 66.7
WZ ?lvjets 3.4
WW ? lv jets 17.1
tt ? (lv)b (lv)b 38.2
tt ? (lv)b (jj)b 242.8
Main backgrounds ttbar 1/100 ,
?theo10 Wjets 1/2000, ?theo ?? ? Use
of DATA !
27
t-channel cross-section

• Selection
• Specific criteria
• Exactly 2 high-pT jets
• 1 high pT central b-jet
• 1 forward light jet ?gt2.5
• Reconstruct Top with mlvb mtgen lt 25
  GeV/c2
• Window in HT or Mtot

Preliminary
Preliminary
S/B 3 v(SB)/S 1.4
28
Wt channel cross-section

• Selection
• Specific criteria
• Exactly 3 high-pT jets
• 1 high pT central b-jet (pTgt 50 GeV/c)
• Reconstruct a W? jj with 60 ltmjj lt 90 GeV/c2
• Reconstruct leptonic Top mlvbmtlt 25
  GeV/c2
• Window in HT or Mtot

Preliminary
Preliminary
Preliminary
S/B 1/ 7 v(SB)/S 4
29
s-channel cross-section

• Selection
• Specific criteria
• Exactly 2 high-pT jets
• 2 high pT central b-jet (?lt2.5)
• Reconstruct Top with mlvb mtgen lt 25
  GeV/c2
• Window in HT or Mtot

Preliminary
Preliminary
S/B 0.2 v(SB)/S 6
likelihood analysis under development
30
s-channel with 30 fb-1 Why is it so
interesting ?

• Charged Higgs single-top
• Production mode in 2 HDM
• 5 higgs 3 neutral (A,h,H) 2 charged (H)
• Mass spectrum predicted in MSSM
• (Htb) couplings depends on mH and tan ß
• ? tb final state cross-sections are modified by
  an H

Preliminary
TopRex generator
mH (GeV/c2)
31
s-channel with 30 fb-1 Why is it so
interesting ?

• Charged Higgs single-top
• Production mode in 2 HDM
• 5 higgs 3 neutral (A,h,H) 2 charged (H)
• Mass spectrum predicted in MSSM
• (Htb) couplings depends on mH and tan ß
• ? tb final state cross-sections are modified by
  an H

Preliminary
Nb of events (30fb-1)
mH 250 tanß50 N 365 10
32
s-channel with 30 fb-1 Why is it so
interesting ?

• Charged Higgs single-top
• Production mode in 2 HDM
• 5 higgs 3 neutral (A,h,H) 2 charged (H)
• Mass spectrum predicted in MSSM
• (Htb) couplings depends on mH and tan ß
• ? tb final state cross-sections are modified by
  an H

Preliminary
Nb of events (30fb-1)
mH 300 tanß50 N 500 15
33
s-channel with 30 fb-1 Why is it so
interesting ?

• Charged Higgs single-top
• Production mode in 2 HDM
• 5 higgs 3 neutral (A,h,H) 2 charged (H)
• Mass spectrum predicted in MSSM
• (Htb) couplings depends on mH and tan ß
• ? tb final state cross-sections are modified by
  an H

Preliminary
Preliminary
Nb of events (30fb-1)
Contours
5s
3s
2s
34
s-channel with 30 fb-1 Why is it so
interesting ?

• Charged Higgs single-top
• Production mode in 2 HDM
• 5 higgs 3 neutral (A,h,H) 2 charged (H)
• Mass spectrum predicted in MSSM
• (Htb) couplings depends on mH and tan ß
• ? tb final state cross-sections are modified by
  an H

ATL-PHYS-2001-017
Preliminary
tanß30, mH300 GeV
L30 fb-1
complementary to direct gb? H? tb search
Contours
5s
3s
2s
35
Conclusion perspectives

• Top pair production
• LHC as a top factory
• more than 300k events a year
• ? measurements will be systematics limited
• Top mass measurements
• Total uncertainty of 1 GeV/c2 seems achievable
• Main systematics
• ? 1 level calibration of (b)-jet energy scale
• Together with mW 15 MeV/c2
• ? constraint mH /mh
• Cross-section measurements
• Systematics limited measurements
• ? luminosity determination ...
• Cross-check of direct mt measurements
• Test of QCD at 6 level
• Single-top production
• Cross-section measurements
• Specific tests of EW production
• t-channel, Wt channel

36
Thanks to
Bentvelsen, Van Vulpen et al.
Correard, Devivie, Rozanov, Vacavant et al.
D. Pallin, P. Roy et al.
37
BACKUP SLIDES
38
Fast simulation vs Full Simulation
39
leptonjets systematics
40
b-tagging 3D impact parameter
courtesy A. Rozanov
s(a0) 11 µm
s(z0) 120 µm
41
Conclusion

• Physics
• Precision measurements are only possible _at_ LHC
• s-channel
• tt , WQQ, Wjets, Wg major backgrounds
• Statistical Precision few with 30 fb-1
• W-g channel
• Higher signal cross-section (cont. by tt
  Wjets)
• Expect statistical precision of 1-2
• Wt channel
• top-pair is the major backgd
• Expect statistical precision of few
• Systematic uncertainty
• JES should be a dominant source of error
• b-tagging knowledge model. (eff, rejection) is
  crucial
• Background knowledge is important
• ? Use of data is mandatory (ttbar, WQQ, Wjets))
  
• FullSim
• Results seem compatible with AtlFast, but

42
Light jet energy scale sharing between two jets
43
Top Quark Mass measurement Top Quark
reconstruction

• Top Quark reconstruction
• Association of hadronic W and b-jet
• Combination leading to the highest pTtop
• or that maximizes ?R(l,b) / minimizes ?R(b,W?jj)
  
• ? right (jjb) combination in 80 cases

right combination
wrong combination
44
Single Top cross-section Production _at_ LHC

• Production at the LHC
• All 3 contributing mechanisms in SM
• Theoretical prediction
• NLO/NLL available for W and W-g only
• ? affect significantely s as well as pT(jet), HT
  etc

t-channel
S-channel
Wt channel
hep-ph/0408049
Channel s(pb) Uncertainties Uncertainties Uncertainties
Channel s(pb) PDF µ-scale ?mtop
W-g 246.6 8.7 4 3 1
Wt 60 20? 10 ? 1
W 10.6 0.7 4 2 3