Searching for Heavy Higgs Bosons Using Events with Leptons and Missing Transverse Momentum with the ATLAS Detector

An additional heavy Higgs boson is predicted by many models, such as the two-Higgs-doublet model (2HDM). This thesis presents a data-driven method to estimate the dominant $ZZ$ background in the search for the decay of such a heavy scalar particle to a pair of $Z$ bosons in the $ZZ$ $\rightarrow$ $\ell^{+} \ell^{-} \nu \bar{\nu}$ ($l$ = electrons, muons, $\nu$ = neutrinos) final state. Currently, the $ZZ$ $\rightarrow$ $\ell^{+} \ell^{-} \nu \bar{\nu}$ background is estimated from simulated Monte Carlo samples, and it contributes to the largest systematic uncertainties in the analysis. The idea of the $Z\gamma$ method is to make use of the similarity of the $ZZ$ and $Z\gamma$ processes, especially in the regions where the mass difference between the $Z$ and the photon does not matter. The $ZZ$/$Z\gamma$ cross-section ratio is calculated as a function of the $\nu\nu$ (or photon) transverse momentum by the MATRIX and SHERPA generators. The ratio is then applied to the $Z\gamma$ events in data, together with a correction of the photon reconstruction efficiency to mimic the production of the $ZZ$ background. The systematic uncertainties on the estimate and comparisons of the results of the method with the Monte-Carlo-based estimate are presented as well. In such searches, the missing transverse momentum ($E_{\text{T}}^{\text{miss}}$) reconstruction is crucial. Studies on the electron-jet overlap removal in the $E_{\text{T}}^{\text{miss}}$ algorithms reducing fake $E_{\text{T}}^{\text{miss}}$ are shown. The main experimental results presented use the full ATLAS Run-2 data sample (2015$-$2018) in proton-proton collisions at $\sqrt{s}$ = 13 TeV centre-of-mass energy, corresponding to 139 fb$^{-1}$. No significant excess over the Standard Model prediction is observed. Therefore, exclusion limits on the production cross-section times the branching ratio of a heavy Higgs boson decaying to $ZZ$ are set at a 95\% confidence level.