Almut Maria Pingel
A thesis submitted in 2015 for the degree of Doctor of Philosophy and defended August 18, 2015.
The PhD School of Science
Faculty of Science
Niels Bohr Institute, Discovery Center, University of Copenhagen
Mogens Dam, NBI
Stefania Xella, NBI
Ian Bearden, NBI (Chair)
Ulrik Egede, Imperial College London, UK
Paula Eerola, University of Helsinki, Finland
Tau lepton identification and studies of associated Higgs boson production with the ATLAS detector
In this thesis two main topics are presented: the identification of hadronic tau decays and a search for Higgs bosons that decay to tau leptons and are produced in association with a Z boson.
In 2012, the ATLAS detector at the LHC recorded 20 fb1 of data at a centre-of-mass energy of 8 TeV. Many studies on Standard Model (SM) processes, as well as searches for physics beyond the SM were carried out, which all rely on a good particle identification. This thesis focuses on the identification of hadronically decaying tau leptons. Multivariate techniques are used to reject backgrounds that mimic the tau signature in the detector and are falsely reconstructed as tau leptons. The focus is on the optimisation of the algorithm using boosted decision trees (BDT). The method is prepared for the 2012 data taking period and made robust against run conditions with many simultaneous protonproton interactions per bunch crossing (pile-up). Furthermore, the dependence of the identification method on the tau momentum is studied, and the usage of additional algorithms that explore the pion content of the tau decay.
The second part concerns the Higgs boson which was discovered in 2012. Higgs boson production via a top quark-loop (gg fusion) as well as via vector boson fusion (VBF) has been observed, but no evidence for the production in association with a vector boson (VH) has been claimed yet. The thesis presents a study of this production mode, with a leptonically decaying Z boson in the final state, and the Higgs boson decaying to a pair of tau leptons. Both the hadronic decay of the tau lepton as well as the leptonic decays (_e, __) are considered. The background is estimated from simulation, and renormalised in side bands to match the data. The contribution is then transferred to the signal region using the fake factor method. Combining the result from the four final states _had _had, _e_had, ___had and _e__, a limit is set on the cross section of the process Z(! ee; __)H(! _had_had). No deviation from the Standard Model is found. An upper observed (expected) limit of 8:96 (6:18+8:94 4:45) times the SM cross section is set at 95% confidence level for a Higgs boson mass of 125 GeV.