Reconstruction of Low Energy Neutrino-Induced Neutrons in the MicroBooNE Detector

Identifying neutrons from neutrino interactions in liquid argon time projection chambers (LArTPCs) can enhance future oscillation measurements by recovering missing energy and improving interaction channel determination. Neutron reconstruction poses many challenges since neutron scatters on argon often leave only small, isolated charge signatures known as "blips" that fall below the threshold of traditional 3D tracking algorithms. A channel of interest for neutron identification is proton production via neutrino-induced neutron inelastic scatters. Efforts are underway to lower identification thresholds for low energy blip-like protons using data collected by MicroBooNE, an 85-tonne LArTPC detector which ran in the Booster Neutrino Beam at Fermilab from 2015 to 2021. A variety of metrics are being explored in MicroBooNE to reduce the considerable background contribution from cosmic rays and neutrino-induced electromagnetic activity. These metrics include blip energy density, spatial distributions, and the 3D direction of blips obtained using a custom machine learning algorithm trained on wire signals from low-energy protons. Here we present the status of this analysis and highlight its implications for the physics reach of current and future neutrino LArTPCs.