Research of Supersymmetry with the ATLAS detector and development of the High Granularity Timing Detector

The Standard Model of particle physics is an extremely successful theoretical framework, describing the elementary particles and their interactions. With the discovery of the Higgs boson by the ATLAS and CMS experiments in 2012, the Standard Model is now complete. However, open questions remain unanswered, calling for a larger theoretical model that encapsulates the Standard Model, while providing mechanisms for the unexplained phenomena. Supersymmetry offers such a framework by introducing a new symmetry between bosons and fermions. It provides potential solutions to the hierarchy problem for the Higgs boson mass and also offers a candidate to explain the dark matter of the universe. The first part of this thesis is the search for supersymmetry with the ATLAS detector at LHC, using the full dataset of Run 2, amounting to an integrated luminosity of 139 fb$^{-1}$. The focus is on the search for squarks and gluinos, the "super-partners" of quarks and gluons, in models where R-parity is conserved and in final states with jets and large missing transverse momentum. My main contribution to this analysis was the development and optimization of a novel technique named Multi-Bin fit to enhance the signal to background separation and extend the exclusion reach of the search. The expected gain in the excluded cross section from using a Multi-Bin fit configuration, opposed to the traditional "cut & count" approach, was estimated to be 40 - 70 % in the studied models. In addition, I worked on the statistical inference of the search, ranging from the evaluation of various systematics to the interpretation of the results in various simplified supersymmetric models. No excess above the Standard Model prediction was found and therefore squarks and gluinos with masses up to 1.85 TeV and 2.34 TeV were excluded, respectively. This result is a significant improvement over the previous round of the analysis and one of the strongest constraints on squark and gluino masses today. The high-luminosity data acquisition phase (HL-LHC) will see an increase of the collision rate by a factor of 5 to 7. In order to mitigate the increase of pile-up, ATLAS will install a new highly granular silicon detector with a very good time resolution that would be located at the forward region, the High Granularity Timing Detector (HGTD). The goal of this detector is to provide a time resolution better than 50 ps per track. The second part of this thesis focuses on two main aspects in the development of HGTD. On one hand, I performed simulation studies to evaluate the occupancy and read-out requirements of the detector under various geometries. The occupancy of the detector must remain below 10%, in order to correctly assign energy deposits to tracks crossing the detector. It was found that this requirement was met with a sensor size of 1.3 $\times$ 1.3 mm$^2$, which is now the baseline for the future detector. Additionally, the organization of the on-detector read-out system was optimised, in order to maximise the available space and minimise the necessary components. The performance of any silicon detector is strongly linked to the design of the front-end electronic circuit. As part of my work in HGTD, I also participated in the characterization of two front-end electronic prototypes, ALTIROC0 and ALTIROC1, both in laboratory with a calibration system and in testbeam with highly energetic electrons and protons. The temporal resolution was found to be better than 55 ps in all tested devices, with a best achieved performance of 34 ps.