1 Measurements at 8 TeV of TTbar events with additional particles in the final state Boris Mangano (ETH Zürich) on behalf of the CMS collaboration
Dec 18, 2015
page 1
Measurements at 8 TeV of TTbar events
with additional particles in the final state
Boris Mangano (ETH Zürich)
on behalf of the CMS collaboration
LHCP 2014Boris Mangano page 2
Outline
TTbar plus what?
- TTbar+jets
- TTbar+BBbar
- TTbar+V:
o TTbar+gamma
o TTbar+Wo TTbar+Z
8TeV results released during the last 12 months
Were these processes seen also at 8TeV ?
new results
PAS TOP-12-041
PAS TOP-13-010
PAS TOP-13-011
8 TeV
TOP-12-018
PAS TOP-12-024
TOP-12-014
7 TeV
LHCP 2014Boris Mangano page 3
Motivation
TTbar plus what?
- TTbar+jets
- TTbar+BBbar
- TTbar+V:
o TTbar+gamma
o TTbar+Wo TTbar+Z
About ½ of TTbar events contain extra jets with |h| < 2.4, pT>30 GeV. Possibility to test higher order QCD calculations.
Irreducible non-resonant background for ttH (H->bb) analysis.
Potential to verify if couplings to the top quark are compatible with SM or altered by BSM effects.
LHCP 2014Boris Mangano page 4
tt+jets @ 8TeV (19.6 /fb)PAS TOP-12-041Event selection:
- Dilepton OS events (el and mu)- For same-flavour pairs, Z-veto
and require MET- N-jets >=2- N-bjets >=1
Unfolding:- cross-section in N-jets bins is
presented at particle level
Extra jets definition:- Kinematic reconstruction
method to identify b-jets from ttbar decay.
- Other extra jets are then analysed
Main source of uncertainty:Up to 20% uncertainty from jet-energy-scale and signal model uncertainties.
LHCP 2014Boris Mangano page 5
The whole is greater than the sum of its parts (Aristotle)
MC@NLO + Herwig does not generate sufficiently large jet multiplicities.
tt+jets @ 8TeV (19.6 /fb)
LHCP 2014Boris Mangano page 6
The whole is greater than the sum of its parts (Aristotle)
Leading extra jet pT (GeV)
tt+jets @ 8TeV (19.6 /fb)
Second Leading extra jet pT (GeV)
These distributions are at reconstruction level
LHCP 2014Boris Mangano page 7
PAS TOP-13-010Event selection:- Dilepton OS events (el and mu)- For same-flavour pairs, Z-veto and
require MET- N-jets >=4- N-bjets (tight) >=2
tt+bb@ 8TeV (19.6 /fb)
Main source of uncertainty:20-25% total uncertainty. Mostly from uncertainties in b-tagging discriminator templates.
Normalization of simulated samples:- Performed by means of a fit to the
measured b-tagging algorithm discriminator.
- Templates for different tt+X samples (including tt+bb signal) are taken from simulation.
LHCP 2014Boris Mangano page 8
The whole is greater than the sum of its parts (Aristotle)
First additional jet Second additional jet
Simultaneous template fit of two data distributions
tt+bb@ 8TeV (19.6 /fb)
LHCP 2014Boris Mangano page 9
The whole is greater than the sum of its parts (Aristotle)
Factor 0.38 to account for difference in selection efficiency between ttbb and ttjj (from simulation)
Cross-section ratio at particle level:
tt+bb@ 8TeV (19.6 /fb)
using again selection efficiencies from simulation
LHCP 2014Boris Mangano page 10
PAS TOP-13-011Event selection:
- one isolated central m with pT > 26 GeV - Veto on extra leptons- N-jets >=4, with pT > 55, 45, 35, 20 GeV- N-bjets >=1- One central photon ET > 25 GeV- DR (photon,muon) > 0.7
tt+gamma @ 8TeV (19.7 /fb)
Main source of uncertainty:20-25% total uncertainty from modelling of the background template.
Signal extraction:
- Template fit of charged-hadron-isolation discriminant distribution in data.
- Templates for signal and misidentified photons from simulations.
LHCP 2014Boris Mangano page 11
tt+gamma @ 8TeV (19.7 /fb)
N-1 plots for some of the observables used to select the signal photons.
LHCP 2014Boris Mangano page 12
The whole is greater than the sum of its parts (Aristotle)
tt+gamma @ 8TeV (19.7 /fb)
Fit of distribution of thecharged-hadron-isolation discriminant
Inclusive ttbar cross section (TOP-12-017)
LHCP 2014Boris Mangano page 13
ttW and ttZ @8TeV (19.5 /fb)
- Both ttW and ttZ processes are very rare and the measurement of their cross sections can be used to test the internal consistency of SM.
TOP-12-036
pre-print to be released in the next days
- ttZ production allows the probe of tZ coupling.
LHCP 2014Boris Mangano page 14
ttW in same-sign dilepton
Signal region definition:- Tight prompt-lepton selection- same-sign dilepton (el or mu)- third lepton vetoed only if it makes a
Z-candidate- N-jets >= 3- N-bjets >=1- lepton pT>40 GeV and HT > 155 GeV
Analysis type:counting experiment
Against misidentified prompt leptons and ttbar background
Against WZ and other diboson processes. Remove overlap with ttZ analysis.
Final optimization against ttbar background
LHCP 2014Boris Mangano page 15
ttW in same-sign dileptonDominant background:
- ttbar with misidentified leptons is estimated from data. Uncertainty is about 50%.
Other backgrounds:
- ttZ, WZ and other rarer processes are estimated from simulations.
- Z+jets with mis-measured lepton charge estimated from data
Six signal regions to exploit larger S/B in plus-plus channels.
LHCP 2014Boris Mangano page 16
ttZ in trilepton
Analysis type:counting experiment
Signal region definition:- Three tight leptons (el or mu)- Two same-flavour opposite-sign
making a Z-candidate- N-jets >= 4- N-bjets (loose) >=2 & N-bjet
(medium) >=1
Against misidentified leptons and ttbar background
Against WZ and other diboson processes. Remove overlap with ttW analysis.
LHCP 2014Boris Mangano page 17
ttZ in trileptonDominant background:
- non-top-quark background processes with b-jets from ISR. Uncertainty is about 50%.
Other backgrounds:
- TTbar with misidentified leptons is estimated from data.
- Rarer, and irreducible, SM processes estimated from simulation
Only one signal region is used for the statistical interpretation of the data
LHCP 2014Boris Mangano page 18
ttZ in four-lepton
Analysis type:counting experiment
Signal regions definition:- Four tight leptons (el or mu)- Two same-flavour opposite-sign
making a Z-candidate- Veto on second Z-candidate- N-bjets =1- N-bjets >1
Against WZ and lepton misidentification
Two non-overlapping signal regions
Against ZZ background
LHCP 2014Boris Mangano page 19
ttZ in four-leptonBackgrounds:
- ZZ(*) in four leptons is estimated from MC
- Z+X events with misidentified leptons
Limits and benefits of this channel:
- statistically limited
- Large S/B ratio
LHCP 2014Boris Mangano page 20
ttV (V = W or Z)
Results in agreement with SM predictions:
cross section ttW = 206 ± 23 fbcross section ttZ = 197 ± 25 fb
Numbers from MG5_aMC@NLO. They are compatible with first NLO calculations from MCFM and PowHel.
LHCP 2014Boris Mangano page 21
Summary
TTbar plus what?
- TTbar+jets
- TTbar+BBbar
- TTbar+V:
o TTbar+gamma
o TTbar+Wo TTbar+Z
8TeV results released during the last 12 months
PAS TOP-12-041
PAS TOP-13-010
PAS TOP-13-011
TOP-12-036
pre-print to be released in the next days
New Results
Latsis Symposium 2013 page 22Boris Mangano
Backup slides
LHCP 2014Boris Mangano page 23
ttV (V = W or Z)
LHCP 2014Boris Mangano page 24
The whole is greater than the sum of its parts (Aristotle)
tt+jets