Back to search

Boston University

Enhancing the neutrinoless double beta decay sensitivity of SNO+ with deep learning

Abstract

dc:description.abstract

Measurements of neutrino oscillations have confirmed that neutrinos are massive, but other questions remain: What are the absolute neutrino masses? Which neutrino is the lightest? Answers to these questions don't only provide all of the remaining unknown quantities in The Standard Model. They also have consequences for our understanding of how fundamental particles acquire mass. One way to investigate neutrino mass is to determine whether or not the neutrino is its own anti-particle, or Majorana in nature. This can be done experimentally by measuring neutrinoless double-beta decay. Once the SNO+ experiment is loaded with natural tellurium, containing 34% ^{130}Te, it will begin to look for this hypothetical process. This dissertation describes a log-likelihood approach to quantifying current background levels in the SNO+ detector, as well as a toy Monte-Carlo study to calculate the experiment's projected sensitivity to neutrinoless double beta decay after 5.25×10-2 ton-years of exposure. This work also describes the additional background rejection achieved by KamNet, A novel deep neural network developed by the KamLAND collaboration. At 90% confidence, the deployment of this tool increases the projected neutrinoless double beta decay half-life limit of SNO+ by 98.5%, from T1/20nu> 3.41×1025 years to T1/20 nu> 6.67× 1025 years. Additionally, Asimov datasets were used to project the decay half-life limit for different tellurium exposures. At a benchmark of three years of livetime, KamNet is projected to achieve a limit of T1/20nu > 1.63 × 1026 years. This limit represents a savings of two years for a traditional analysis of the same tellurium loading. And, although 1.5% tellurium loading is projected to achieve this sensitivity in just under two years, it does so at an extra cost of roughly 2.6M.

Degree

thesis:*
Name thesis:degree_name
Doctor of Philosophy
Level thesis:degree_level
doctoral
Discipline thesis:degree_discipline
Physics
Grantor
Boston University
Year dc:date.issued
2025

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Gooding, Diana M.
Advisor dc:contributor.advisor
  • Grant, Christopher P.

Subjects

dc:subject × 1

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Handle dc:identifier.uri
https://hdl.handle.net/2144/53291

Chain of custody

source
Harvested from
Boston University
Base URL
open.bu.edu/oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
related terms
citation

Gooding, Diana M.. Enhancing the neutrinoless double beta decay sensitivity of SNO+ with deep learning. doctoral thesis, Boston University, 2025. https://hdl.handle.net/2144/53291