🛠️ This is a sandbox environment
Published January 28, 2019 | Version v1

Search for charged Higgs boson in the τ+jets final state with proton-proton collision data recorded with the ATLAS experiment

Authors/Creators

  • 1. Cracow INP

Contributors

  • 1. Cracow INP

Description

The thesis presents a search for charged Higgs bosons, predicted in many the- ories beyond the Standard Model, decaying via H ± → τ ± ν in the mass range be- tween 90 and 2000 GeV. It is based on 36 fb −1 of data collected with the ATLAS √ detector at center-of-mass energy s = 13 TeV in the years 2015 and 2016. As an improvement to the former searches, a multivariate analysis is applied and used to separate the H ± signal from the Standard Model background processes. The Boosted Decision Trees, BDT, output score is used as the discriminating variable in the final statistical analysis. Furthermore, since at low H ± masses i.e. between the W -boson and top-quark masses, the kinematics of the t → bH ± and t → bW ± decay products can be very similar, variable sensitive to the the τ lepton polarisation is additionally used to improve discrimination power. The data are found to be in agreement with the background-only hypothesis. Model-independent limits are set on the production cross-section times branching fraction σ (pp → [b]tH ± ) × BR(H ± → τ ± ν) between 4.2 pb and 0.0025 pb for the charged Higgs boson mass range of 90–2000 GeV, corresponding to upper limits between 0.25% and 0.031% for the branching fraction BR(t → bH ± ) × BR(H ± → τ ± ν) in the mass range 90– 160 GeV. In the context of the hMSSM scenario, all tan β values are excluded for m H ± below 160 GeV, whereas the H ± mass range up to 1100 GeV is excluded at tan β = 60. The presented thesis constitutes also the first complete implementation of the embedding method in the ATLAS Run-2 analysis. The technique makes use of the lepton universality of the W boson decay by constructing hybrid samples with sim- ulated hadronically decaying τ leptons replacing muons from W ± → μ ± ν events in data. The main advantage of the embedding method is that one relies on the sim- ulations only for the well understood electroweak decays of W boson and τ lepton and all other event properties are taken directly from data. It has been proven on 2015 data that the technique presents a significant improvement in the precision of background estimation with true hadronically decaying τ leptons. Compared to the fully simulated events, the uncertainty on the yield of the dominant t t ̄ background containing true τ leptons reduces from 26% to the mere 11%.

Files

CERN-THESIS-2018-340.pdf

Files (22.8 MB)

Name Size Download all
md5:67c413bfffe40542a0a1052f2dc7b994
22.8 MB Preview Download

Additional details

Identifiers

CDS
2655308
CDS Report Number
CERN-THESIS-2018-340

CERN

Department
PH - Physics Department
Programme
No program participation
Accelerator
Not applicable
Experiment
ATLAS

Linked records