A Test of Lepton Flavour Universality in the Decays of the Top Quark and Data driven validation of the Hard Scatter Vertex Selection Efficiency at ATLAS

Marshall, Emma and Borissov, Guennadi (2025) A Test of Lepton Flavour Universality in the Decays of the Top Quark and Data driven validation of the Hard Scatter Vertex Selection Efficiency at ATLAS. PhD thesis, Lancaster University.

Text (2025marshallphd)
2025marshallphd.pdf - Published Version
Available under License Creative Commons Attribution-NonCommercial-NoDerivs.
Download (16MB)

Abstract

The first part of the thesis covers the work done for the ‘qualification task’ at ATLAS, contributing to the maintenance of the experiment, measuring the efficiency of Hard Scatter identification in isolated lepton events: Z 0 → ee¯, Z 0 → µµ¯ and tt¯ → eµ . Code was developed to extract the HS efficiency, the efficiency at which the ‘true’ first interaction of an event is identified by ATLAS software , with an ‘estimation’ method and a ‘fitting’ method in Release 21 of AnalysisBase. It was found that the Z 0 decay processes have a large discrepancy in the percentage of misidentified Hard Scatter vertices between Monte Carlo simulation and data unlike in the tt¯ channel where Monte Carlo and data agreed. The percentage of misidentified Hard Scatter vertices was also found to be far greater in Z 0 decays than tt¯. The fitting method was then developed for Release 22, and found that the HS efficiency was much better than Release 21, with the HS efficiency for Z 0 → ℓℓ events comparable to the efficiency of tt¯ events in Release 21. The second part is the main physics analysis, aiming to test a fundamental axiom of the Standard Model, Lepton Flavour Universality, the hypothesis that the couplings of each lepton flavour to the electroweak gauge bosons are the same regardless of flavour. The aim of the analysis presented is to test lepton flavour universality through the measurement of the of the rate of decay of W bosons to τ-leptons and electrons, R(τ/e) = B(W → τντ )/B(W → eνe). Di-leptonic tt¯ events are based on 139 f b−1 of data recorded with the ATLAS detector in proton–proton collisions at √s = 13TeV. The analysis follows a ‘tag and probe’ method; one lepton is the tag to select events, the other a probe to determine if the lepton comes from a ‘prompt’ decay, W → eνe, or an intermediate τ lepton, W → τντ → eνeντντ . The result of the analysis is still blinded as it is still in the approval process by the ATLAS collaboration. The expected precision of the measurement of R(τ/e) is 0.0234.