1 Introduction

ATL-PHYS-PROC-2018- July 20, 2019

Top-quark pair production in association with a or boson with the ATLAS experiment

Sebastian Heer111Supported by European Research Council grant ERC-CoG-617185., on behalf of the ATLAS Collaboration\@footnotetextCopyright 2018 CERN for the benefit of the ATLAS Collaboration. CC-BY-4.0 license.

Physikalisches Institut

Universität Bonn, 53115 Bonn, Germany

The cross section of the  and  processes are measured in a simultaneous fit using 36.1 of proton-proton collisions at a centre of mass energy of recorded by the ATLAS detector at the LHC. The result is found to be and and compatible with the Standard Model.


International Workshop on Top Quark Physics

Bad Neuenahr, Germany, September 16–21, 2018

1 Introduction

The  and  processes provide direct access to the weak couplings of the top quark. In addition, the two processes are often found to be a relevant background in searches involving final states with multiple leptons and -quarks. Measurements of and can be used to set constraints on the weak couplings of the quarks, as well as impose limits on possible beyond the Standard Model (BSM) effects. Previous measurements by the ATLAS and CMS experiments [1, 2] indicate agreement with the Standard Model [3, 4].

The dataset of 36.1 of proton-proton collisions at a centre of mass energy of , collected in 2015 and 2016 by the ATLAS detector is analysed. Final states with two, three or four isolated prompt leptons are considered in the analysis. More details about the analysis can be found in Reference [5].

2 Analysis Overview

Separate analysis channels are defined to target the  and  processes. Each channel is divided into multiple regions in order to enhance the sensitivity to the signal. Table 1 lists the analysis channels and the targeted decay modes of the  and  processes.

Process decay Boson decay Channel
2 SS
Table 1: List of  and  decay modes and analysis channels targeting them. The symbols and denote a -quark or antiquark and neutrino or antineutrino, respectively [5].

The 2-OS channel targets the  process and is affected by large backgrounds from +jets or production. In order to separate the signal from background in the most efficient way, Boosted Decision Trees (BDTs) are used. Three BDTs are constructed and trained in regions requiring at least 5 jets. The number of input variables varies between 14 and 17. The most important input variables are found to be the pseudorapidity of the dilepton system and the transverse momenta of all jets divided by the sum of their energies. Each signal region is divided into 19 equal sized bins of the BDT discriminator. The normalization factors of the +jets background with one or two heavy flavour jets, labeled as Z+1HF and Z+2HF, are determined in the fit to data.

The 2-SS channel targets the  process and uses a total of 12 signal and control regions, classified by information about the lepton flavour, total charge and -jet multiplicity. The dominant background arises from fake leptons, which are estimated by the matrix method [6] and charge-flip leptons. The control regions are used in a fit to measure fake and real lepton efficiencies, as well as in the final fit to determine the signal cross sections, anticorrelating the fake lepton background with the signal process.

The regions in the 3 channel are divided into two groups, depending on whether a pair of OSSF leptons whose invariant mass is within of the boson mass is present. The signal regions are further categorized based on jet and -jet multiplicities. Four signal regions for each signal process are defined. The dominant backgrounds for the regions targeting  arise from diboson production and the production of single top quarks in association with a boson. A control region is used to determine the normalization of the background in the fit to data. The dominant background for the regions targeting  arises from fake leptons, which is estimated by the matrix method.

The 4 channel targets the  process and uses four signal regions depending on the -jet multiplicity and the flavour of the leptons from the system. In addition, a control region is used to determine the normalization of the background in the fit to data. The dominant backgrounds arise from the diboson production, the production of a single top quark in association with a and a boson, as well as fake leptons. The fake lepton background is estimated from simulation, applying scale factors obtained from a fit in control regions enriched by fake leptons.

3 Results

The  and  cross sections are simultaneously extracted using a binned maximum-likelihood fit to the numbers of events in the dilepton, trilepton and tetralepton signal and control regions. In total, the fit includes 11 signal plus 2 control regions from the , and 16 signal plus 12 control regions from the  analysis. In all regions, good agreement between observed values and the expectation is observed. The normalization corrections for the and backgrounds with respect to the predictions are obtained from the fit and found to be compatible with unity: for the background and for the background. The normalizations of the +1HF and +2HF backgrounds are mainly constrained in the low BDT output bins of the opposite-sign dilepton channel signal regions, where the signal contamination is low. Their values are found to be and , respectively. The measured values of the cross sections from the combined fit are and , assuming the SM value for the other cross section, respectively. Figure 1 shows a comparison of the fit results with theoretical expectations, pb and pb [7], demonstrating good agreement between the measured and predicted cross sections. For the  process, both the observed and the expected significances are found to be much larger than 5 standard deviations. For the  process, an excess of events over the expected Standard Model background-only hypothesis is found with an observed (expected) significance of () standard deviations. Table 2 shows the uncertainties in the measured  and  cross sections, grouped in categories, along with the total uncertainties. For both processes, the precision of the measurement is affected by statistical and systematic uncertainties in similar proportions. For the  determination, the dominant systematic uncertainty sources are the modelling of the backgrounds and of the signal. For the  determination, the dominant systematic uncertainty sources are the modelling of the signal and the limited amount of data available in the control regions and simulated samples.

Figure 1: The result of the simultaneous fit to the  and  cross sections along with the 68% and 95% confidence level (CL) contours [5]. The cross shows the theoretical uncertainties in the SM predictions, and include renormalization and factorization scale uncertainties, as well as PDF uncertainties including variations [7].
Luminosity 2.9 4.5
Simulated sample statistics 2.0 5.3
Data-driven background statistics 2.5 6.3
JES/JER 1.9 4.1
Flavor tagging 4.2 3.7
Other object-related 3.7 2.5
Data-driven background normalization 3.2 3.9
Modeling of backgrounds from simulation 5.3 2.6
Background cross sections 2.3 4.9
Fake leptons and charge misID 1.8 5.7
modelling 4.9 0.7
modelling 0.3 8.5
Total systematic 10 16
Statistical 8.4 15
Total 13 22
Table 2: List of relative uncertainties in the measured cross sections of the  and  processes from the fit, grouped in categories. All uncertainties are symmetrized. The quadratic sum may not be equal to the total due to correlations between uncertainties introduced by the fit [5].


The work of the author is currently funded by the European Research Council under the European Union’s Seventh Framework Programme ERC Grant Agreement n. 617185.


  • [1] ATLAS Collaboration, JINST 3 (2008) S08003.
  • [2] CMS Collaboration, JINST 3 (2008) S08004.
  • [3] ATLAS Collaboration, Eur. Phys. J. C 77 (2017), 40, arXiv:1609.01599 [hep-ex].
  • [4] CMS Collaboration, JHEP 08 (2018) 011, arXiv:1711.02547 [hep-ex].
  • [5] ATLAS Collaboration, ATLAS-CONF-2018-047,
  • [6] ATLAS Collaboration, Eur. Phys. J. C 71 (2011) 1577,
    arXiv:1012.1792 [hep-ex].
  • [7] J. Alwall et al., JHEP 07 (2014) 079, arXiv:1405.0301 [hep-ex].
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description