|
|
Search for the top → τνq
channel in top dilepton
|
Authors
S.Tourneur, A.Savoy-Navarro
CNRS/IN2P3 & University of
Read the public note
email the authors
Public presentations and papers
Read the public note
Presentation at the international conference of
the American Physical Society, at
Summary
We are looking for a first
observation of top pair production in the τ+l (l
= e or μ) channel where the tau decays semihadronically. This channel is interesting as a signal
yet to be observed, but also as a background for new physics. The main
backgrounds are events with wrongly identified taus (W+jets, ttbar, QCD) and Z→ττ+jets.
Using
350 pb-1 of collected data, we observe 5 events against
2.7±0.4 background events expected. The probability
for the background alone to fluctuate to 5 or more events is 15%. This gives a
1 sigma evidence for the top tau dilepton
signal.
The 5
events observed are in perfect agreement with the standard model prediction for
signal+background, so that no indication for new
physics is observed in this top pair lepton+tau
channel.
350
pb-1 goal and strategy
|
|
With 350 pb-1 of
data collected at CDF from 2001 to September 2004, the physics goal is still the mere observation of
the signal.
Fot this, we need to compute the standard model
expectations for the backgrounds and compare to the observation. For this
“search-like” analysis, the quantity to be optimized is S/√B.
Event selection:
·
1 central
isolated lepton (e or μ) with a transverse energy greater than 20 GeV
·
1 central isolated tau decaying into hadrons
with a transverse energy greater than 15 GeV. The identification looks for a narrow cone
containing the 1 or 3 charged hadrons of the tau
decay.
·
2 jets : the leading one has Et > 25 GeV
and the second one Et > 15 GeV
·
Missing
transverse energy (>20 GeV) to account for the 3
neutrinos in the signal
·
High
activity (Ht > 205 GeV) to account for the high
top mass
·
Passes a
veto against Z→ττ+2 jets
Results
|
|
CDF Run II preliminary
|
electron
+ tau (360 pb-1)
|
muon
+ tau (340 pb-1)
|
jet → τ fakes
|
0.91 ± 0.29
|
0.92 ± 0.29
|
e → τ fakes
|
0.10 ± 0.025
|
0.05 ± 0.012
|
Z → ττ + 2 jets
|
0.39 ± 0.13
|
0.32 ± 0.10
|
WW + 2 jets
|
0.034 ± 0.011
|
0.027 ± 0.008
|
Total background
|
1.43 ± 0.31
|
1.32 ± 0.30
|
SIGNAL (ttbar →
l+τ)
|
1.32 ± 0.05stat only
|
0.92 ± 0.05stat only
|
|
Nb observed
events |
2 |
3 |
Background prediction method
The Z→ττ + 2 jets
background is estimated using Monte-Carlo (Pythia, Alpgen) corrected to agree with data in Z→μμ
+ 2 jets.
The most
important and trickiest background comes from events where the tau is misidentified. W→lν+
3 jets and ttbar→lνbjjb can end this way
into the signal region. The probability for a generic central jet to
be misidentified as a tau is determined to vary
between 0.3 and 1% depending on the jet Et and activity in the event. This
background is determined this way:
1. apply full signal
selection to the data sample but the tau
identification : 1 lepton, missing Et, >=2 jets, Ht>205 GeV, Z veto
2. Ask for a central
loose tau instead of the final tau
identification : the tau candidate is not required to
be isolated and to have a mass < 1.8 GeV
3. weight the event by
the probability for the loose tau(s) to be
isolated and to pass the mass cut*
*The
probability for a loose tau to be isolated and to
pass the mass cut (fake rate) is computed from multijet
events and lies between 3 and 10%. This is parametrized
with the jet Et, the sum of transverse energies deposited
in the event, and the number of jets in the event. This fake rate is checked to
make good predictions in a W+3 jets sample, with a conservative uncertainty of
30%.
Control regions
Control regions are defined close
enough to the final signal region to constitute a valuable check, but far
enough to keep enough data in for statistical relevance.
We defined 4 control regions:
(to be compared
with signal region : 1 identified electron or muon +
missing Et > 20 GeV + >= 2 jets, Ht > 205 GeV, Z veto)
·
1
identified electron + missing Et > 20 GeV + >= 0 jet
·
1
identified electron + missing Et > 20 GeV + >= 1 jet
·
1
identified muon + missing Et
> 20 GeV + >= 0 jet
·
1
identified muon + missing Et
> 20 GeV + >= 1 jet
For each
of them several distributions are checked:
Electrons:
>=0 jet:
>=1 jet:
>=0 jet:
>= 1 jet:
>=0 jet:
>=1 jet:
>=1 jet:
Muons:
>= 0 jet:
>=1 jet:
>=0 jet:
>=1 jet:
>=0 jet:
>= 1 jet:
>=1 jet:
Kinematical characteristics of the 5 observed events
CDF RunII preliminary |
Event 1 (electron) |
Event 2 (electron) |
Event 3 (muon) |
Event 4 (muon) |
Event 5 (muon) |
Missing Et (GeV) |
71.9
|
53.5
|
99.8
|
29.2
|
35.9
|
Missing Et Φ (rad) |
-1.52
|
-1.42
|
1.46
|
0.89
|
0.76
|
Tau Et (GeV) |
19.7
|
39.6
|
43.7
|
44.4
|
30.9
|
Tau pseudorapidity |
-0.43
|
-0.95
|
0.02
|
0.62
|
-0.35
|
Tau Φ (rad) |
3.80
|
3.75
|
2.05
|
3.10
|
5.60
|
Tau z0 (cm) |
-51.1
|
3.7
|
-22.8
|
7.9
|
41.3
|
Tau N tracks |
3
|
1
|
3
|
3
|
3
|
Lepton Et (GeV) |
67.7
|
41.9
|
41.6
|
29.4
|
53.1
|
Lepton pseudorapidity |
-0.79
|
-0.27
|
-0.80
|
-0.23
|
0.07
|
Lepton Φ (rad) |
1.71
|
1.06
|
5.99
|
1.20
|
2.93
|
Lepton z0 (cm) |
-51.3
|
-3.35
|
-23.0
|
9.4
|
41.3
|
Number of jets |
2
|
3
|
2
|
4
|
2
|
1st jet Et (GeV) |
35.3
|
69.3
|
61.9
|
101.8
|
48.6
|
1st jet pseudorapidity |
-0.74
|
-0.25
|
-0.61
|
1.93
|
0.38
|
1st jet Φ (rad) |
1.25
|
2.89
|
0.82
|
5.64
|
1.06
|
2nd jet Et (GeV) |
33.2
|
39.8
|
55.0
|
96.2
|
47.7
|
2nd jet pseudorapidity |
0.52
|
-1.49
|
-1.20
|
0.41
|
0.95
|
2nd jet Φ (rad) |
5.00
|
0.22
|
2.65
|
2.29
|
4.44
|
Ht (GeV) |
228
|
244
|
302
|
301
|
216
|
6 distributions of the 5
events compared with Standard Model expectations:
Ht :
= lepton Et + tau Et + 1st jet Et + 2nd
jet Et + missing Et
(strongly
correlated with the high top mass)
Missing Transverse Energy:
Lepton (e or μ)
transverse energy:
Tau transverse energy:
Leading jet transverse energy:
2nd jet transverse
energy:
Last modified 2006-06-10 : added link to
a preliminary public note, commenting text, updated the histograms for
predictions of the final candidate events, added number of tracks in the 5 tau candidates