Phd in ACCELERATOR PHYSICS

Thesis title: Machine-Detector Interface Optimization for a √s = 3 TeV Muon Collider

This thesis addresses the optimization of the Machine-Detector Interface (MDI) for a Muon Collider operating at a center-of-mass energy of 3 TeV. The central focus is the design of the shielding nozzle, a critical component for mitigating Beam-Induced Backgrounds (BIB). The BIB arises from the interactions of secondary particles—mainly photons, neutrons, and electrons—produced by electromagnetic showers initiated by electrons and positrons originating from muon decays along the beamline. A detailed characterization of this background was carried out through high-statistics FLUKA simulations, followed by an extensive study of how key geometrical parameters of the shielding influence the background flux. Due to the significant computational cost of full simulations, a Machine Learning–based strategy was developed to guide the design process. A surrogate model was trained to reproduce the simulation output, enabling rapid evaluation of new configurations. A custom metric was introduced to balance background suppression with geometric acceptance, and a deep learning model was employed to identify the optimal nozzle design. The resulting geometry preserves strong shielding performance while reducing material volume, offering a promising solution for the MDI of future muon colliders.