Entanglement and correlations in fast collective neutrino flavor oscillations
Alessandro Roggero, Ermal Rrapaj, Zewei Xiong
Abstract
Collective neutrino oscillations play a crucial role in transporting lepton flavor in astrophysical settings like supernovae and neutron star binary merger remnants, which are characterized by large neutrino densities. In these settings, simulations in the mean-field approximation show that neutrino-neutrino interactions can overtake vacuum oscillations and give rise to fast collective flavor evolution on timescales $t\ensuremath{\propto}{\ensuremath{\mu}}^{\ensuremath{-}1}$, with $\ensuremath{\mu}$ proportional to the local neutrino density. In this work, we study the full out-of-equilibrium flavor dynamics in simple multiangle geometries displaying fast oscillations in the mean field linear stability analysis. Focusing on simple initial conditions, we analyze the production of pair correlations and entanglement in the complete many-body-dynamics as a function of the number $N$ of neutrinos in the system, for up to thousands of neutrinos. Similarly to simpler geometries with only two neutrino beams, we identify three regimes: stable configurations with vanishing flavor oscillations, marginally unstable configurations with evolution occurring on long timescales $\ensuremath{\tau}\ensuremath{\approx}{\ensuremath{\mu}}^{\ensuremath{-}1}\sqrt{N}$, and unstable configurations showing flavor evolution on short timescales $\ensuremath{\tau}\ensuremath{\approx}{\ensuremath{\mu}}^{\ensuremath{-}1}\mathrm{log}(N)$. We present evidence that these fast collective modes are generated by the same dynamical phase transition which leads to the slow bipolar oscillations, establishing a connection between these two phenomena and explaining the difference in their timescales. We conclude by discussing a semiclassical approximation which reproduces the entanglement entropy at short to medium timescales and can be potentially useful in situations with more complicated geometries where classical simulation methods starts to become inefficient.