Neutron Calibration and Characterisation of the DEAP-3600 Experiment Using a 74MBq AmBe Neutron Source

Franco La Zia

Research output: ThesisDoctoral Thesis

81 Downloads (Pure)


The existence of dark matter was first proposed by Fritz Zwicky in the 1930’s to explain the motions of galaxies in the Coma cluster. Since then the evidence for dark matter has mounted, however its nature has remained illusive. One of the leading dark matter
candidates is the Weakly Interacting Massive Particle (WIMP). The DEAP-3600 experiment is a single-phase liquid argon detector located 2 km underground at SNOLAB, Sudbury, Ontario, Canada which aims to measure WIMP interactions with a sensitivity of 10−46 cm2 for a 100 GeV/c2 WIMP mass. To detect WIMPs liquid argon detectors look for elastic scatters, using pulse shape discrimination (PSD) to distinguish between nuclear (WIMP-like) and electronic (background-like) recoils. Characterisation of PSD is therefore of paramount importance to achieve a low energy WIMP search threshold resulting in increased sensitivity. DEAP-3600 makes use of a 74 MBq AmBe neutron source to populate the detector with nuclear recoils, allowing for calibration of the detector to WIMP-like nuclear recoils. The goal of this thesis is to characterise DEAP-3600 response to nuclear recoils and to determine the WIMP nuclear recoil acceptance using the AmBe source. This is non-trivial due both to the higher cross-section of neutron scatters on liquid argon compared with WIMP interactions as well as the geometry of the detector relative to the source deployment position, as the neutrons will multiple scatters as they traverse the detector. This work presents a method to isolate single scatters (WIMP-like) in the AmBe data by using machine learning algorithms trained on AmBe Monte Carlo simulations. Due to the complexity of neutron interactions a method for simulation to
achieve the statistics required for this analysis was developed. The result of this thesis is a determination of the nuclear recoil acceptance in DEAP-3600.
Original languageEnglish
Awarding Institution
  • Royal Holloway, University of London
  • Walding, Joseph, Supervisor
Thesis sponsors
Award date1 May 2020
Publication statusUnpublished - 2020


  • Dark Matter
  • Neutron Calibration
  • Monte Carlo
  • Machine Learning

Cite this