CERN Accelerating science

Thesis
Report number	CERN-THESIS-2023-401
Title	The search for pair production of Higgs Bosons at the ATLAS experiment
Author(s)	Veliscek, Iza (Oxford U.)
Publication	247.
Thesis note	PhD : Oxford U. : 2023
Thesis supervisor(s)	Huffman, Brian Todd ; Frost, James Alexander
Note	Presented 14 Sep 2024
Subject category	Particle Physics - Experiment
Accelerator/Facility, Experiment	CERN LHC ; ATLAS
Abstract	The thesis presents the search for $HH\rightarrow b\bar{b}b\bar{b}$ that exploits the 126$\text{fb}^{-1}$ of proton-proton data collected by the ATLAS detector during the Large Hadron Collider Run 2 at $\sqrt{s} = 13\text{TeV}$. Done for the first time in an ATLAS diHiggs analysis, the independent optimisation of the gluon-gluon fusion (ggF) and vector boson fusion (VBF) production channels not only improves the sensitivity to anomalies in the trilinear Higgs self-coupling ($\kappa_\lambda$) but also to anomalies in the barely explored HHVV coupling ($\kappa_{2V}$). The combination of the ggF and VBF channels allows the exploration of the $\kappa_\lambda - \kappa_{2V}$ parameter space and extracting the $\pm 2\sigma$ confidence region with the $-2\ln\Delta L$ test statistic. The observed upper limit on the Standard Model (SM) HH signal strength is found to be 5.4 times the SM value. Lastly, the ggF channel is interpreted using Higgs and SM Effective Field Theory, the latter is the first such diHiggs ATLAS result. Taking a step further the combination of the three most sensitive diHiggs channels in the $b\bar{b}b\bar{b}$, $b\bar{b}\gamma\gamma$ and $b\bar{b}\tau^{-}\tau^{+}$ final states is performed, which is further joined with the single Higgs combination. The thesis mainly focuses on the technical adjustments needed in the non-resonant $HH\rightarrow b\bar{b}b\bar{b}$ analysis to allow for its inclusion in the single Higgs and diHiggs combination. The upper bound on the diHiggs SM signal strength extracted from the diHiggs combination is 2.4 times the SM prediction. The power of the single Higgs and diHiggs combination is that it allows the consideration of a generic $\kappa_{\lambda}$ model, where in addition to the trilinear Higgs self-coupling, the couplings of the Higgs to the top quark, bottom quark, tau lepton and vector bosons can vary. The observed (expected) 95\% confidence level is measured to be $-1.3<\kappa_{\lambda}<6.1$ $(-2.1<\kappa_{\lambda}< 7.6)$, when the generic $\kappa_{\lambda}$ model is fitted. The thesis concludes with the in-situ calibration of the mis-tag rate for the multijet background in the boosted $X\rightarrow b\bar{b}$ tagger. The calibration uses simulated dijet events enriched with $g\rightarrow b\bar{b}$ and the full Run 2 data to derive preliminary scale factors that quantify the difference in the mis-tag efficiency between data and simulations.

Corresponding record in: Inspire
Email contact: ivelisce@bnl.gov