Spin Correlations, Polarisation and Charge Asymmetries in Top Quark Pairs James Howarth on behalf of the ATLAS Collaboration The University of Manchester 1
Jan 21, 2016
1
Spin Correlations, Polarisation and
Charge Asymmetries in Top Quark Pairs
James Howarthon behalf of the ATLAS Collaboration
The University of Manchester
2
Top Quark Properties
t
t
l+
ν
W+
W-
d
u
b
b
PropertiesMass, Mass difference,
Width, Charge, Spin, Polarisation,
Spin correlation, Asymmetries
ProductionCross section, generator studies, rare couplings
DecayW Polarisation, Colour Flow,
Branching Ratios, |Vtb|
The Top Quark
• Very heavy (Mtop =173 GeV)
• Narrow width short lifetime (10-25s)• Decays before hadronisation• “Bare” quark (almost)
3
Dilepton Events
Dilepton Events:• 2 Leptons (opposite sign, isolated, trigger match)• At least 2 Jets• ET > 60 GeV (ee/μμ)
• HT > 130 GeV (eμ) HT = Scalar sum of all lepton & jet pT
• mll > 15 GeV and veto 81 GeV < mlll < 101 GeV (ee/μμ)
• Optional b-tagging (70% efficiency working point)
l+
ν
l-
νl
b
bt
t
W+
W-
miss
4
Single Lepton
u
d
l-
νl
b
bt
t
W+
W-
Single Lepton Events:• 1 Lepton (isolated, trig match)• At least 4 Jets with at least 1 b-tag
(e + jets)• ET > 20 GeV • ET + mT > 30 GeV
(μ + jets)• ET > 35 (30) GeV • mT > 25 (30) GeV
mT =
miss miss
miss W W
W
5
Spin Correlation and
Polarisation
6
Introduction – Spin Variables
• Polarisation (P) and Spin Correlation (A) closely related.
Greg Mahlon, 3rd International Workshop on Top Quark Physics, Bruges, May 2010
7
Introduction – Spin Variables
• Polarisation (P) and Spin Correlation (A) closely related.
cos(θi/j)
• Angle between a decay particle from the top and some chosen spin quantisation axis in top rest frame.
• Top Helicity is a well known example.
8
Introduction – Spin Variables
• Polarisation (P) and Spin Correlation (A) closely related.
Spin Correlation
Spin Correlation
&
9
Introduction – Spin Variables
• Polarisation (P) and Spin Correlation (A) closely related.
• Looking at the double differential distribution requires full tt reconstruction.
• ΔΦ between two spin analysers in lab frame is also sensitive to spin correlations.
• Does not require tt reconstruction and sensitive to gluon gluon fusion.
• Precision test of the SM.
• Ideal variable for LHC!
10
Introduction – Spin Variables
• Polarisation (P) and Spin Correlation (A) closely related.
Polarisation
11
Introduction – Spin Variables
• Polarisation (P) and Spin Correlation (A) closely related.
lepton d u
αi/j (LO) 1.00 1.00 -0.31
αi/j (NLO) 1.00 0.93 -0.31
• Leptons and down type jets are the best analysers• Hard to distinguish u type jets from d type jets
F. Hubaut, E. Monnier, P. Pralavorio, K. Smolek, V. Simak, Eur.Phys.J. C44S2 (2005) 13-33
12
Results
13
Observation of Spin Correlation
AHELICITY : 0.40 ± 0.04 (stat) +0.08/-0.07 (syst)SM Prediction: 0.31
2.1 fb-1 7 TeV
• Perform likelihood fit with templates A=SM and A=0 and extract fSM.
• fSM = 1.30 ± 0.14 (stat.) +0.27/-0.22 (syst.)
• Extrapolate AHELICITY = fSM * PredHELICITY
• Exclude the A = 0 case with 5.1
First Observation of Spin Correlation in tt!
Consistent with hypothesis of spin ½
Phys. Rev. Lett: Volume 108 Issue 21(Prediciton) W. Bernreuther and Z. G. Si, Nucl. Phys. B 837, 90(2010)
14
Top Quark Polarisation4.7 fb-1 7 TeV
αP : -0.060 ± 0.018 (stat) +0.046/-0.064 (syst)
• Template fit similar to Spin Correlation, hypotheses αP = +1 and αP = -1• Dominant Systematic comes from jet reconstruction, jet resolution and signal modeling.
• All results consistent with SM.
ATLAS-CONF-2012-133
Even
ts
15
Charge Asymmetry
16
Introduction – Charge Asymmetry
• |Δyl|=|yl+|-|yl-|: Doesn’t Require Reconstruction• |Δyt|=|yt+|-|yt-|: Easier to compare
AC
Asymmetry measurements at the LHC and Tevatron:
yAFB
• Small Asymmetry expected at Tevatron (~5%)• Large deviation seen (~15%)• LHC tt is pp collider and charge symmetric.• Asymmetry observable but expected to be very small
17
Single Lepton Charge Asymmetry1.04 fb-1 7 TeV
Eur. Phys. J. C (2012) 72:2039: http://arxiv.org/abs/1203.4211
• Events reconstructed using Kinematic Likelihood Fitter (KL)• Rapidity distributions unfolded to parton level.
• Inclusive and differential measurements in mtt
MC@NLO (inclusive) Prediction AC = 0.006 ± 0.002
AC = -0.019 ± 0.028 (stat.) ± 0.024 (syst) inclusive
AC = -0.052 ± 0.070 (stat.) ± 0.054 (syst) mtt > 450 GeV
AC = -0.008 ± 0.035 (stat.) ± 0.032 (syst) mtt < 450 GeV
high
low
18
Single Lepton Charge Asymmetry
Eur. Phys. J. C (2012) 72:2039: http://arxiv.org/abs/1203.4211
INCLUSIVE mtt > 450 GeVNote: Does not include most recent results from all colliders
AC = -0.019 ± 0.028 (stat.) ± 0.024 (syst) inclusive
AC = -0.052 ± 0.070 (stat.) ± 0.054 (syst) mtt > 450 GeV
AC = -0.008 ± 0.035 (stat.) ± 0.032 (syst) mtt < 450 GeV
high
lowD
0
19
Dilepton Charge Asymmetry4.7 fb-1 7 TeV
• Asymmetry measured at parton level
• Background subtracted data corrected for detector acceptance.
• Measured both lepton and top asymmetry.
• Results consistent with SM (predictions from MC@NLO).
COMB WITH L+JETS RESULT
ATLAS-CONF-2012-57
20
Current ATLAS Asymmetry Results
• Current Status of the 7 TeV Asymmetry measurements at ATLAS• Everything consistent with SM
AC
-0.1 -0.08 -0.06 -0.04 -0.02 0 0.02 0.04 0.06 0.08 0.1
ATLAS (combined)
ATLAS (dilep)
ATLAS (single lepton)
prediction
21
tt + HF NEW!
• tt + HF (HF = bb, cc, b, c)
• Flavour coming from gluon splitting or proton• main irreducible background to ttH• Many BSM models also predict additional HF in tt events.
At least 3 b-tagged jets
At least 2 b-tagged jets
• Many experimental uncertainties cancel in ratio• Cross sections measured in fiducial region (with a factor to correct for detector acceptance)• Correction factor is 0.093 ± 0.004(stat.) in HF, 0.129 ± 0.001(stat.) in tt + j
Heavy Flavor Ratio
22
tt + HF 4.7 fb-1 7 TeV NEW!
RHF = [7.1 ± 1.3 (stat.) +5.3/-2.0 (syst.)] %
RHF (LO) = [3.4 ± 1.1 (syst.)] %RHF (NLO) = [5.2 ± 1.7 (syst.)] %
• perform a binned maximum likelihood fit to extract σfid (tt + HF)
• σfid (tt + j) taken from observed yield in data.
• Use the displaced vertex mass in bins of b-tag purity (improves LF-c jet discrimination).
To be submitted to Physical Review D
• Dominant systematic is flavor composition
23
Summary and Outlook
• LHC delivering precision measurements.
• First observation of non vanishing Spin correlations, measurements on Polarisation and Charge Asymmetry.
• 7 TeV Analyses for ATLAS Top Properties almost completed (one or two more to look forward to)
• Start to see 8 TeV properties results soon!
• Charge Asymmetry results so far statistics limited.
• Long Shutdown 1 will be a very exciting time for top properties and plenty to look forward to with 13 TeV!
24
Backup
25
References
“Spin Correlations: TEVATRON vs. LHC”, Greg Mahlon, 3rd International Workshop on Top Quark Physics, Bruges, May 2010https://agenda.irmp.ucl.ac.be/conferenceTimeTable.py?confId=538#20100530
“ttbar asymmetries in the SM”, Germán Rodrigo, LHC Workshop “From Charge Asymmetry to the Boosted Regime” 02-04 May 2012https://indico.cern.ch/conferenceOtherViews.py?view=standard&confId=175916
“Top Charge Asymmetry at ATLAS”, Kerim Suruliz, LHC Workshop “From Charge Asymmetry to the Boosted Regime” 02-04 May 2012https://indico.cern.ch/conferenceOtherViews.py?view=standard&confId=175916
Figures
Papers“ATLAS sensitivity to top quark and W boson polarization in tt events” ̄� , F. Hubaut, E. Monnier, P. Pralavorio, K. Smolek, V. Simak, Eur.Phys.J. C44S2 (2005) 13-33http://arxiv.org/abs/hep-ex/0508061
“Distributions and correlations for top quark pair production and decay at the Tevatron and LHC”W. Bernreuther and Z. G. Si, Nucl. Phys. B 837, 90(2010), and private communication.
26
Systematics: Dilep Charge Asymm
27
tt + HF – b tag purityHigh Purity Medium Purity
Low Purity
28
Asymmetries Cartoon
http://cds.cern.ch/record/1493476/files/TOP-12-010-pas.pdfCMS Dilepton
CMS l+jets http://cds.cern.ch/record/1458955/files/plb.717.129.pdf
http://cds.cern.ch/record/1453785/files/ATLAS-CONF-2012-057.pdf
ATLAS Dilep + Combination
ATLAS l + jets http://link.springer.com/article/10.1140%2Fepjc%2Fs10052-012-2039-5
29
Introduction – Spin Variables
lepton d u
αi/j (LO) 1.00 1.00 -0.31
αi/j (NLO) 1.00 0.97 -0.32
• Polarisation (P) and Spin Correlation (A) closely related.
Double Differential DistributionP = 0, A = SM
[a]
30
Physics Objects at ATLAS
JETS:• Anti-kT jets (R=0.4)• Corrected to EM Scale• |η| < 2.5, ET > 25 GeV• 75% tracks point to vertex• B-Tagged using Neural Net based tagger, eff 70%.
MUONS:• Muon Spectrometer track combined with Inner Detector track• Isolated from other objects•|η| < 2.5• PT > 20 GeV
ELECTRONS:• Central energy deposit combined with Inner Detector track• Isolated from other objects• |ηcluster| < 2.47, • 1.37 < |ηcluster | < 1.52• ET > 25 GeV
MISSING TRANSVERSE ENERGY:• Sum of transverse energies of all reconstructed physics objects.• Accounts for unassociated calorimeter cells. ?
31
Event Model - Backgrounds
“lepton” Fakes (mismeasured leptons)• Data Driven• Matrix Method
Single Top:• MC@NLO + HERWIG & JIMMY• Dilepton Wt channel only.
Drell-Yan: ee, μμ, ττ• ALPGEN + HERWIG & JIMMY• Z + (0-5 jets inclusive)
W + jets:• ALPGEN + HERWIG & JIMMY• W + (0-5 jets inclusive)
DiBoson:• HERWIG (ALPGEN) & JIMMY• WW, WZ and ZZ