Development and optimization of novel large area silicon pad sensors for the CMS High Granularity Calorimeter

The Large Hadron Collider (LHC) is currently (2024) the largest (27 km in circumference) storage ring and has the highest collision energy (13.6 TeV). While its contributions to high-energy physics remain substantial, more luminosity is needed to obtain sufficient experimental data for new physical discoveries in a reasonable time. To secure the role of the LHC for the future, CERN is planning an upgrade to increase the luminosity by a factor of 10, called "High-Luminosity LHC" (HL-LHC). This upgrade will improve the machine's ability to detect rare processes and increase statistical precision. However, the experiments require better spatial and temporal resolution and improved beam hardness at higher luminosity.$\\$ This work focuses on the silicon detectors for the Compact Muon Solenoid (CMS) experiment. CMS will also upgrade the sensors during the HL-LHC upgrade due to the higher requirements mentioned above. It is called the "CMS Phase-2 Upgrade". The upgrade to the new endcap calorimeters is called "High Granularity Calorimeter" (HGCAL). CMS plans the HGCAL as a sandwich calorimeter consisting of silicon sensors for the electromagnetic sub-detector and for parts of the hadronic sub-detector. A further motivation for the HGCAL is particle flow analysis, which allows the identification of individual particles in jets. CMS must meet these requirements at a reasonable cost, so large hexagonal p-type sensors manufactured in an 8-inch process are used. Large particle physics experiments did not employ 8-inch sensors yet, so they represent a first-of-a-kind (FOAK) novelty. Accordingly, this research is necessary to verify the suitability of this technique for high-energy particle physics experiments. These 8-inch sensors were designed by the author based on preliminary work and improved in iterative prototyping cycles. For the irradiation studies, the author designed and used Ministrip sensors, which are much more flexible to use, contact, and handle due to their small dimensions.$\\$ The author further improved an existing test station and carried out initial inspections for mechanical damage in preparation for the studies and to qualify the test system. The electrical properties, such as leakage current, capacitance, and high-voltage stability, were tested and compared with the requirements of the CMS experiment. Analysis methods such as spreading resistance profiling (SRP), secondary ion mass spectrometry (SIMS), scanning electron microscopy (SEM), and confocal laser microscopy complement the electrical methods mentioned above. The 11 tested sensors showed a yield of around 36 %. An essential point of this work is the evaluation of radiation hardness in terms of electrical characterization, such as leakage current, pad capacitance, interpad isolation through ohmic resistance, and interpad capacitance. The neutron fluence of the irradiated sensors and test structures ranges up to 10$^{16}$ n$_{\text{eq}}$/cm$^{2}$, so that low temperatures from -30 to -20 °C) and the blowing of dry air are required for electrical measurements on irradiated sensors. The core element of the irradiation studies is the test series of annealing, or more precisely, the behavior of the sensors after defined heating. Regarding radiation hardness, this study has shown that the new 8-inch sensors offer comparable results to silicon sensor technologies already established and comprehensively described in the literature, such as 6-inch and n-type sensors. At one point, when reducing the damage function $\alpha$ through annealing, the five Ministrip sensors examined showed significantly better annealing behavior than already established silicon sensor technologies in n-type and on 6-inch wafers. The author suggests future examinations to verify these results. In that case, these new sensors should be considered well qualified for the Phase-2 Upgrade of the HGCAL, and CMS will use the final sensor designs for series production, and the sensors eventually implemented in the HGCAL. In total, around 30,000 silicon sensors with a total area of around 620 m$^{2}$ will be installed in HGCAL.