Transcript Top: latest results from Tevatron – cross-section and mass Mircea N. Coca

Top: latest results from
Tevatron –
cross-section and mass
Mircea N. Coca
University of Rochester, NY- CDF
For the CDF and D0
collaborations
FPCP 2003, Paris, June
Outline
Tevatron Status
The upgrades of the CDF and D0 detectors
Top Production and Decay
Top Physics Program for Run II
First Cross-Section Measurements at
1.96 TeV, in the Dilepton and Lepton+jets
channels
Top Mass Measurements in CDF (Run II)
and D0 (Run I)
Top Physics Prospects
June 6th, 2003
Mircea Coca, U of Rochester - CDF
2
Tevatron Upgrades/Status
Integrated Luminosity
Run II upgrades
ECM increase from 1.8 → 1.96
TeV→larger cross sections
Higher luminosity



Run I peak:2.4x1031 cm-2 s-1
Run II goal:3–4x1032cm-2 s-1
Run II peak:4.7x1031 cm-2 s-1
Winter ‘03
Commissioning
Analysis-quality data
accumulated by Jan ‘03
CDF: 72.0 pb-1
( 57.5 pb-1 with silicon)
D0: 30 - 50 pb-1
Peak Luminosity
Immediate goal for
accelerator:
Deliver 225 pb-1 in FY 2003
Run IIa goal: 2 fb-1
June 6th, 2003
Mircea Coca, U of Rochester - CDF
3
CDF and D0 Detectors Upgrades
D0
CDF
Tracking:
Expanded silicon coverage
New drift chamber (COT)
Extended lepton-ID: ||>1.0 →2.0
End Plug calorimeter
Expanded muon coverage
June 6th, 2003
New Inner tracking
silicon tracker, fiber tracker
2T superconducting solenoid
Upgraded  system for
better  -ID
Mircea Coca, U of Rochester - CDF
4
Top Production and Decay
-
In proton-antiproton
collisions, at 1.96 TeV,
top quarks are primarily
produced in pairs
single top production:
W Decay Mode
smaller rate (s =1.5 pb)
large backgrounds
not observed yet
-
q
q
stt increased by 30% with
the CM energy increase
from 1.8 →1.96 TeV
Br(tW+b) ~100% in SM
Based on the W decay
modes →3 experimental
signatures:
June 6th, 2003
~15%
~85 %
t
g
t
W-
ln
jj
ln
jj
b
b
b
b
b
W+
ln
ln
b
b
b
jj
b
jj
b
(1)
(2)
(3)
(1) Dilepton Very small backgrounds, but
very small rate
(2) Lepton+Jets Manageable
backgrounds and good rate
(3) All Jets Large QCD Background
Mircea Coca, U of Rochester - CDF
5
Top Physics in Run II
( Fermilab 1995) → crude look
at top’s properties
Run II: precision mode→
we hope to answer
fundamental questions:
Why is the top so
heavy?
Is the third
generation special?
Is top involved
in EWSB?
Is the top the liaison
to new physics?
W helicity
Top Mass
Run I: discovery mode
Top Width
Production
cross-section
Resonance
production
l+
Top Spin
W+
p
n
t
b
Production
kinematics
Top Spin
Polarization
CP violation
Top Charge
X
_
p
_
b
_
t
q
Rare/non SM Decays
(eg: t→Zc/gc, t→H+b)
W_
q’
Branching Ratios
|Vtb|
June 6th, 2003
Anomalous
Couplings
Mircea Coca, U of Rochester - CDF
6
Production Cross-Sections
s tt measurement:
Nobs -Nbck
benchmark measurements
s tt =
test of perturbative QCD
probe for physics beyond SM
 non-SM production, X  tt
 non-SM decay, t  Xb
 SUSY models with a tt-like signal
Higgs production (WH,ZH) is a background
and the opposite
Acceptance
Run I:
dstt/stt ~26 %
Run IIa (2fb-1): dstt/stt ~ 7 %
A.L
Luminosity
Theoretical cross-section:
At NLO @ s=1.96 TeV for Mtop = 175 GeV:
hep-ph/0303085
s tt = 6.7 +-00..71
pb
88
(Mangano et al)
To estimate signal contribution we use 7 pb
June 6th, 2003
Mircea Coca, U of Rochester - CDF
7
stt in the Dilepton Decay Mode
Backgrounds
Event Selection
2 high-ET, isolated leptons (e, )
WW/WZ, Z/g*→ tt determined
from Monte Carlo (MC)
Z/g*→ee,  from data+MC
W+jets, QCD Heavy Flavor
from data
jet
t to be included for the future
l
large missing energy ET
CDF: Veto Z-mass window events for
ee, 
at least 2 jets with large ET
large transverse energy flow
HT =(ETleptons ,ETjets)
June 6th, 2003
n
b
D0: Raised ET cut in Z window
p
p
b
jet
Mircea Coca, U of Rochester - CDF
n
l
8
ET
Dilepton Channel (ee, e, ) stt
Source
ee (events)
L (pb-1)
48.0
 ( events)
42.6
e (events)
33.0
Background
1.00  0.49
0.60  0.01
0.07  0.01
Signal
0.25  0.02
0.30  0.04
0.50  0.01
Run II data
Run II
Preliminary:
4
2
1
21.0
+14.1
+3.0
s tt = 29.9 +-15
(
stat
)
(
sys
)
.7
- 6.1
-3.0 (lum ) pb
June 6th, 2003
Mircea Coca, U of Rochester - CDF
9
e+2 jets Top Candidate
ET(e+) = 20.3 GeV
pT(-) = 58.1 GeV/c
ETjet(1) = 141.0 GeV
ETjet(2) = 55.2 GeV
ET
= 91
GeV
HT (e) = 216 GeV
Transverse View
ET
jet
Longitudinal View
e+
jet
June 6th, 2003
Mircea Coca, U of Rochester - CDF
10
Dilepton Channel stt
Data
sample
luminosity:
72 pb-1
Source
ee (events)
 ( events)
e (events)
Background
0.10  0.06
0.09  0.05
0.10  0.04
Signal
0.47  0.05
0.59  0.07
1.44  0.16
Run II data
1
3
s tt = 13.2  5.9(stat)  1.5(syst)  0.8(lum) pb
Events
DF(leptons)
Run II
Preliminary:
1
Missing energy ET (GeV)
June 6th, 2003
Missing energy ET (GeV)
Mircea Coca, U of Rochester - CDF
11
Kinematics of Dilepton Candidates
Run 1
L = 109 pb-1
1.8 TeV
9 events
Run 2 Preliminary
L= 72 pb-1
1.96TeV
5 events
Events with very large missing ET in Run 1
June 6th, 2003
Mircea Coca, U of Rochester - CDF
12
Lepton+Jets stt
Event Pre-Selection
A high PT isolated, charged lepton (e,
), large missing ET ( n undetected)
Large jet multiplicity ( ≥ 3 )
Cosmic ray, electron conversion
removal, dilepton veto, Z boson veto.
Further selections to reduce the
background
jet
l
ET
n
b
topological:
b jets with Soft Lepton Tag (SLT)
≥ 3 jets, ≥ 1 SLT tag (DØ)
b jets with displaced vertex (SECVTX)
≥ 3 jets, ≥ 1 b tag (CDF)
June 6th, 2003
p
b
p
≥ 4 jets (DØ)
B-Jet with
SECVTX
jet
l
jet
jet
B-Jet with SLT
Mircea Coca, U of Rochester - CDF
13
Lepton+Jets Topological stt
Backgrounds
Event Pre-Selection
Preselect a sample
enriched in W events
an EM object or  with large
PT and large missing energy
Veto soft ’s in sample,
veto dilepton events
DØ Run II Preliminary
a = 0.1450.02
June 6th, 2003
QCD multi-jets evaluated from
data vs.Njets
e+jets: due to fake jets (po and g)
+jets: due to heavy flavor decays
W multi-jets background in the
4 jet bin estimated using data
by Berends scaling law before
topological cuts
a
s (W + (n + 1) jets )
Mircea Coca, U of Rochester - CDF
s (W + n jets )
14
Results for Topological Analysis
QCD background estimation
Topological Selection
≥ 4 jets (|| < 2.5() or
|| < 2.0(e), pT >15 GeV)
Aplanarity >0.065
HT(ETjets) >180 GeV (e)
HT(ETjets+pTW)>220GeV ()
QCD
Source
L (pb-1)
Background
Signal
Run II data
June 6th, 2003
Mircea Coca, U of Rochester - CDF
+jets
e+jets
49.5
40.0
2.7  0.6
2.7  1.1
1.8
2.4
4
4
15
Lepton+Jets stt with an SLT tag
Event Selection
preselection as for
topological stt
≥ 3 jets
softer topological cuts:
HT(ETjets)> 110 GeV
Aplanarity > 0.04
soft  inside a jet
(b→, b→c→)
Backgrounds
QCD and W+jets determined
from data
Source
L (pb-1)
Background
e+jets
+jets
50
0.2  0.1
Expected
Signal
Run II data
40
0.7  0.4
0.5
0.8
2
0
Lepton+jets channels (SLT + Topological) combined s
Run II
Preliminary:
s tt = 5.8+-43..34 ( stat ) +-42..16 ( sys ) +-00..66 (lum ) pb
June 6th, 2003
Mircea Coca, U of Rochester - CDF
16
Lepton+jets stt with a SECVTX-tag
Secondary
Vertex
Event Selection
preselect a sample
enriched in W events as
already mentioned
≥ 3 jets with ET>15 GeV
≥ 1 jet with secondary
vertex tag (SECVTX)
d0
+Lxy
Jet axis
Prompt
Tracks
Primary
Vertex
A jet is tagged as b jet if it
Probability of tagging a tt event:
has at least 2 good tracks
and the displacement Lxy
satisfies Lxy/sxy >3 (typical
(event tag) = 45  1  5 %
sxy~150 m, while Lxy~3 mm)
June 6th, 2003
Mircea Coca, U of Rochester - CDF
17
Backgrounds Estimation
Backgrounds
Jet Multiplicity for the background events
and the data
Mistags:
from # tagged jets with
Lxy<0 in inclusive jet data
W+heavy flavor:
from W+jets data, b tag rate
and flavor composition
Non W:
from data
WW, WZ, Z→tt, single top:
from Monte Carlo simulation
1 and 2 jet bins are used as a
control sample, the top events
are in >= 3 jet bins
15 Candidates in ~ 57.5 pb-1
June 6th, 2003
control
Mircea Coca, U of Rochester - CDF
signal
18
Lepton+jets stt - SECVTX-tagging
Data
sample
luminosity:
57.5 pb-1
Run II
Preliminary:
Source
W+1jet
W+2 jets
W+3jets
W+4jets
Expected
Bkgr+Signal
34.0 ± 5.0
18.7 ± 2.4
7.4 ± 1.4
7.6 ± 2.0
26
7
8
Run II data
31
σ(t t ) = 5.3  1.9(stat)  0.8(sys)  0.3(lum) pb
June 6th, 2003
Mircea Coca, U of Rochester - CDF
19
*
A “golden” lepton+jets candidate
tt l+jet candidate: Nov 02 2002
run: 153693 event: 799494
 + 4 jets, with 2 SECVTX b-tags
Lego view
Jet3
SECVTX tag
Event primary
vertex
Jet2
Jet1
Jet4
µ
SECVTX tag
June 6th, 2003
Mircea Coca, U of Rochester - CDF
20
Top Cross-Sections Summary
D0 Preliminary
D0: All channels combined
Run II Preliminary:
s tt = 8.5
+4.5
-3.6
( stat )
June 6th, 2003
+6.3
-3.5
( sys )
+0.8
- 0.8
(lum ) pb
Mircea Coca, U of Rochester - CDF
21
Top Mass: Lepton+jets
Event Selection
Reconstruction Method
Select ≥4 jet events, similar to
stt analysis, except no
requirement for a jet to be btagged
l
W+
X
PDG: MW, GW, Gt
5 vertices:
20 constraints
t
t
W-
jet
jet
b-jet
June 6th, 2003
12 different jet-partons assignments
Every combination has two solutions
for the n longitudinal momentum
Impose Mt=Mt , M(j,j)=M(l,n)=MW
b-jet
n
Each event→up to 24 solutions
consistent with a top decay:
2-C fit applied, chose the event
top mass corresponding to the
lowest 2 (iff 2 < 10)
Parameterized templates of top
masses (150, 200) GeV and
bkgd
Continuous likelihood to extract top
mass and statistical uncertainty
Mircea Coca, U of Rochester - CDF
22
Top Mass Measurement
33 candidates after event
selection
8 events with a b tagged
14.4
Mtop = 171.2-+12.5
(stat )  9.9(sys)Ge V/c 2
Systematic uncertainty summary
Work to improve understanding of detector
June 6th, 2003
CDF Run 1 combined:
Mtop = 176.1 ± 6.5 GeV/c2
Mircea Coca, U of Rochester - CDF
23
Top Mass using b-tagging
Identifying a b-jet has a
great impact:
Smaller combinatorics →
improves the mass resolution
by ~10 %
Reduction in background→
S/B = 3, increase by 300%
Allow to loosen the 4th jet
selection cuts ( 40% more
events)
In 57.5 pb -1 there are 11
candidates with at least
one jet tagged as a b-jet
Mtop with b-tagging is
coming…
June 6th, 2003
Mircea Coca, U of Rochester - CDF
24
Run I Mass: lepton+4 jets events
Similar with Kondo’s method, uses full set of event observables
Define a signal event probability P - ( x , M )
tt
i
top
i-th event
observables
Define a background probability
Pbkg ( xi )
Build an event probability P( xi ;a ) = c1P - ( xi , M top ) + c2 Pbkg ( xi )
tt
where a = (Mt,c1,c2)
Build a likelihood L(a), minimize –lnL(a) to get c1, c2 and Mt
Measured
Transfer fn: resolutions,
reconstruction effects
to be estimated
1 n
P ( x; a) = Acc( x )   d s( y; a) dq1 dq 2 f (q1 ) f (q 2 ) W( x, y)
s
Acceptance
Matrix Element
PDF’s
LO ME used, 4 jets required exclusively, additional cut on background
probability (to improve the sample purity)
June 6th, 2003
Mircea Coca, U of Rochester - CDF
25
Run I: Preliminary result
D0 Run I Statistics [PRD 58(1998), 052001]
Events 91→ 71 with exactly 4 jets → 22 after probability cut
-log(likelihood) vs Mt
likelihood vs Mt
m top = 180.1  3.6 (stat)  4.0(syst) GeV/c 2
Run I D0 lepton+jets:
173.3± 5.6(stat) ± 5.5 (syst) GeV/c2
June 6th, 2003
Stat : 5.6 GeV from PRD 2001
improvement on the statistical
uncertainty (~2.4 stats)
Mircea Coca, U of Rochester - CDF
26
Summary & Conclusions
Top physics is extremely rich and has a great
potential
Many top analyses are in progress
we re-established the benchmark top quark measurements
we are getting close to Run I precision
Improvements are underway
Better detector understanding
Increase the tagging efficiencies of b jets
Include forward leptons
We are enthusiastic about the top physics prospects at the
Tevatron until first LHC results
Expect results from larger samples soon
Many measurements will supersede those of Run I
Test the Standard Model to even greater
precision
June 6th, 2003
Mircea Coca, U of Rochester - CDF
27
Top Physics Prospects for 2 fb-1
Measurement
Est. Uncertainty
Mt
2-3 GeV/c2
dstt
7%
d[sll/sl+j]
12%
d[B(t→Wb)/B(t→WX)]
2.8%
d[B(t→Wb)/B(t→Xb)]
9%
d[B(t→Wlong)]
5.5%
d[B(t→WV+A)]
2.7%
d[s*B(Z’→t t)]
~90 fb
dstbX+btX
dG(t→Wb)
dVtb
June 6th, 2003
24%
26%
13%
Tests
Indirect MH
QCD Couplings
Non-SM Decays
‘’
‘’
Non-SM Coup.
W helicity
Exotics
Observe single top
Mircea Coca, U of Rochester - CDF
CKM Matrix
28
End of talk : Backup Slides
June 6th, 2003
Mircea Coca, U of Rochester - CDF
29
Top Mass Templates
Reconstructed top masses
from data are compared to
parameterized templates
of top and background
Monte Carlo for masses
(150, 200) GeV
Use a continuous
likelihood method to
extract top mass and
statistical uncertainty
The bump in the
background shape around
130 GeV is due to the
kinematic selection of the
events
June 6th, 2003
Mircea Coca, U of Rochester - CDF
signal
background
30
Top Dilepton Kinematics
June 6th, 2003
Mircea Coca, U of Rochester - CDF
31
Constraint MHiggs with a Mtop and MW
DØ / CDF
Run 2a
Goal
June 6th, 2003
Mircea Coca, U of Rochester - CDF
32
Direct Higgs Search
June 6th, 2003
Mircea Coca, U of Rochester - CDF
33
Single Top
2.44 ± 0.12 pb ?
0.88 ± 0.12 pb ?
s<0.1 pb
Steltzer, et al. ‘98
Smith/Willenbrock ‘96
Tait ‘99
June 6th, 2003
Mircea Coca, U of Rochester - CDF
34