Searches for New Physics at MiniBooNE: Sterile Neutrinos and Mixing Freedom

Abstract

The MiniBooNE experiment was designed to perform a search for ν µ → ν e oscillations in a region of ∆m 2 and sin 2 2θ very different from that allowed by standard, threeneutrino oscillations, as determined by solar and atmospheric neutrino experiments. This search was motivated by the LSND experimental observation of an excess of ν e events in aν µ beam which was found compatible with two-neutrino oscillations at ∆m 2 ∼ 1 eV 2 and sin 2 2θ < 1%. If confirmed, such oscillation signature could be attributed to the existence of a light, mostly-sterile neutrino, containing small admixtures of weak neutrino eigenstates. In addition to a search for ν µ → ν e oscillations, MiniBooNE has also performed a search forν µ →ν e oscillations, which provides a test of the LSND two-neutrino oscillation interpretation that is independent of CP or CP T violation assumptions. This dissertation presents the MiniBooNE ν µ → ν e andν µ →ν e analyses and results, with emphasis on the latter. While the neutrino search excludes the two-neutrino oscillation interpretation of LSND at 98% C.L., the antineutrino search shows an excess of events which is in agreement with the twoneutrinoν µ →ν e oscillation interpretation of LSND, and excludes the no oscillations hypothesis at 96% C.L. Even though the neutrino and antineutrino oscillation results from MiniBooNE disagree under the single sterile neutrino oscillation hypothesis, a simple extension to the model to include additional sterile neutrino states and the possibility of CP violation allows for differences between neutrino and antineutrino oscillation signatures. In view of that, the viability of oscillation models with one or two sterile neutrinos is investigated in global fits to MiniBooNE and LSND data, with and without constraints from other oscillation experiments with similar sensitivities to those models. A general search for new physics scenarios which would lead to effective non-unitarity of the standard 3×3 neutrino mixing matrix, or mixing freedom, is also performed using neutrino and antineutrino data available from MiniBooNE.Thesis Supervisor: Janet M. Conrad Title: Professor of Physics Acknowledgments I owe this work to my advisor, Janet Conrad. Janet, you have been not only an unending source of inspiration for me these past five years, but also the reason I joined neutrino physics in the first place. Our very first meeting during the Columbia graduate student open house in 2005 was what introduced me to the excitement of neutrinos, while our interaction since then has made me appreciate the richness of this field. Thank you for the opportunity of becoming your advisee, and for your guidance and dedication over these past five years.I can't express how grateful I am to all the people I've worked most closely with during these past five years: I must begin with Michel Sorel, who has been like an older brother to me since the summer of 2005, introducing me to his work on sterile neutrinos and supporting me as I followed in his footsteps; Mike Shaevitz, for looking o...

Assessing the role of nuclear effects in the interpretation of the MiniBooNE low-energy anomaly

“…The correlation is important because the (−) ν µ events, which are measured with high statistics, provide a normalization for the (−) ν e event predictions [101,102]. In principle one should add the multinucleon interactions also to the analysis of the (−) ν µ data.…”

“…However, we could not do it because there is no publicly available sample of Monte Carlo (−) ν µ events corresponding to the final MiniBooNE results in neutrino [45] and antineutrino [46] mode. Moreover, the MiniBooNE Monte Carlo has been tuned in order to fit the high-statistics (−) ν µ events [101,102] through the choices of an overall normalization factor 1.28, the value of the axial mass m A = 1.23 GeV and a nuclear Pauliblocking factor κ = 1.022. Therefore, our best approach is to neglect possible corrections to the analysis of modification of the nucleon axial mass parameter in order to fit the measured ν µ cross sections simulating the multinucleon influence naturally leads to the following question: is the multinucleon effect already incorporated in the Monte Carlo ν e full transmutation events?…”

Effects ofCPviolation from neutral heavy fermions on neutrino oscillations, and the LSND/MiniBooNE anomalies

“…I conclude that the inclusion of CP violation and nonunitarity in the 3+1 dimensional mixing matrix allows for mixing of the 3 active neutrinos with a sub eV mass sterile neutrino to fit all the short baseline electron flavor appearance data without necessarily conflicting with muon and electron neutrino disappearance data. Previous attempts [42][43][44] to fit the short baseline data with oscillations among 3 active plus 2 sterile neutrinos were not able to obtain as good a fit because only mass squared differences of less than 1000 eV 2 were considered, for which the values of U e5 and U µ5 are more constrained.…”

“…The MiniBooNE data on neutrinos disfavors electron neutrino appearance in a muon neutrino beam at the values of L/E explored by LSND, but favors an excess of electron neutrinos at higher values of L/E [40]. With at least 2 additional sterile neutrinos, CP violation in oscillations can reconcile the LSND and MiniBooNE anti neutrino results with MiniBooNE neutrino results [42][43][44]. However reconciling MiniBooNE and LSND with constraints on muon or electron neutrino disappearance from a variety of other short baseline experiments [45][46][47][48][49] is more difficult.…”

On the decaying-sterile-neutrino solution to the electron (anti)neutrino appearance anomalies

“…To derive M ij , we followed the step-by-step description available in ref. [54]. We are going to define the fractional systematics-only covariance matrix as M frac kl .…”