Presentation

• Yongyu PAN

Room: Amphi Charpak **Abstract:** Cosmological evidence confirms dark matter's (DM) existence, yet its microscopic nature remains a major mystery. This thesis searches for light dark matter (LDM) using the XENONnT detector. Standard searches require both scintillation (S1) and ionization (S2) signals, but low-mass DM often falls below the S1 threshold. Consequently, this work develops a dedicated ionization-only (S2-only) analysis. A detailed S2 pulse shape model is constructed to characterize electron drift and electric-field distortions, improving the understanding of ionization signals. The thesis's core achievement is the S2-only analysis framework. Since dropping S1 introduces instrumental backgrounds, this thesis designs robust event-selection strategies using machine learning and waveform analysis to build and validate the first complete background model for S2-only searches. Analyzing a 7.8 tonne-year exposure, this framework is applied to various LDM candidates. It successfully sets new constraints on few-GeV WIMPs, sub-GeV DM, axion-like particles, and dark photons.

Jury closed session and deliberation Room: Amphi Charpak