Experimental neutrino physics and astrophysics with the IMB-3 detector.

Abstract

Description of the universe on the smallest (elementary particle physics) and largest (cosmology) scales has become dependent on the properties of the most weakly interacting fundamental particle known, the neutrino. The IMB experiment, designed to study nucleon decay, is also the world's largest detector of neutrinos. The experiment uses 6800 tons (3300 tons fiducial) of water as both target and detecting medium. Relativistic charges particles traversing the water radiate Cerenkov light. The distinctive ring patterns are imaged by 2048 light collectors (each a photo-multiplier tube coupled with a wavelength-shifting plate) distributed over the surfaces of the tank. This dissertation describes the IMB-3 detector, a four-fold increase in sensitivity over the original apparatus. Neutrino interactions of both atmospheric and extragalactic origin were collected during a 3.4 kiloton-year exposure. A consequence of non-zero neutrino mass could be oscillation of neutrino flavor. The energies and long flight distances of atmospheric neutrinos offer a unique opportunity to explore this possibility. To study the composition of the atmospheric neutrinos, single-ring events are classified as showering or non-showering using the geometry of the Cerenkov pattern. A simulation of neutrino interactions and a model of atmospheric neutrino production are used to predict the composition of the sample. The showering/non-showering character of an event is strongly correlated with the flavor of its neutrino parent. In the lepton momentum range p $<$ 1500 MeV/c, non-showering events comprise 41 $\pm$ 3(stat.) $\pm$ 2(syst.)% of the total. The fraction expected is 51 $\pm$ 5(syst.)%. Although this is evidence for an anomaly in the composition of atmospheric neutrinos, the 2$\sigma$ deviation is not sufficient to require neutrino oscillations. Eight interactions recorded over a six second interval on February 23, 1987 are coincident with the discovery of Supernova 1987a. These data, together with simultaneous signals in a least one other detector, confirm transport of energy by neutrinos, a lynchpin of stellar collapse models. Prompt arrival of neutrinos from $\sim$50 kiloparsec distant places many constraints on the properties of neutrinos which rival those from terrestrial experimentation. Some solutions to the missing mass or dark matter problem result in high-energy neutrino production within the Sun. A model of dark matter capture and annihilation in the Sun predicts the resulting neutrino fluxes at Earth. No evidence of the phenomenon is observed, but for canonical values of dark matter density and velocity in the solar system, greater exposure will be required to verify or exclude the expected signal.