Neutrino Signatures and the Neutrino-Driven Wind in Binary Neutron Star Mergers
We present VULCAN/2D multigroup flux-limited-diffusion radiation-hydrodynamics simulations of binary neutron star mergers, using the Shen equation of state, covering ≳ 100 ms, and starting from azimuthal-averaged two-dimensional slices obtained from three-dimensional smooth-particle-hydrodynamics simulations of Rosswog & Price for 1.4M☉ (baryonic) neutron stars with no initial spins, co-rotating spins, or counter-rotating spins. Snapshots are post-processed at 10 ms intervals with a multiangle neutrino-transport solver. We find polar-enhanced neutrino luminosities, dominated by ¯νe and "νμ" neutrinos at the peak, although νe emission may be stronger at late times. We obtain typical peak neutrino energies for νe, ¯νe, and "νμ" of ∼12, ∼16, and ∼22 MeV, respectively. The supermassive neutron star (SMNS) formed from the merger has a cooling timescale of ≾ 1 s. Charge-current neutrino reactions lead to the formation of a thermally driven bipolar wind with (M·) ∼ 10^−3 M☉ s^−1 and baryon-loading in the polar regions, preventing any production of a γ-ray burst prior to black hole formation. The large budget of rotational free energy suggests that magneto-rotational effects could produce a much-greater polar mass loss. We estimate that ≾ 10^−4 M☉ of material with an electron fraction in the range 0.1–0.2 becomes unbound during this SMNS phase as a result of neutrino heating. We present a new formalism to compute the νi ¯νi annihilation rate based on moments of the neutrino-specific intensity computed with our multiangle solver. Cumulative annihilation rates, which decay as ∼t^−1.8, decrease over our 100 ms window from a few ×1050 to ∼ 1049 erg s−1, equivalent to a few ×10^54 to ∼10^53 e−e+ pairs per second.
Additional Information© 2009. The American Astronomical Society. Received 2008 June 25; accepted 2008 September 4; published 2008 December 30. Print publication: Issue 2 (2009 January 10). We acknowledge helpful discussions with and input from Ivan Hubeny, Casey Meakin, Jim Lattimer, Stan Woosley, H.-Thomas Janka, Bernhard Müller, and Martin Obergaulinger. This work was partially supported by the Scientific Discovery through Advanced Computing (SciDAC) program of the US Department of Energy under grant numbers DE-FC02-01ER41184 and DE-FC02-06ER41452. C.D.O. acknowledges support through a Joint Institute for Nuclear Astrophysics postdoctoral fellowship, sub-award 61-5292UA of NFS award 86-6004791. E.L. acknowledges support from ISF grant 805/04. The computations were performed at the local Arizona Beowulf cluster, on the Columbia SGI Altix machine at the Ames center of the NASA High-End Computing Program, at the National Center for Supercomputing Applications (NCSA) under Teragrid computer time grant TG-MCA02N014, at the Center for Computation and Technology at Louisiana State University, and at the National Energy Research Scientific Computing Center (NERSC), which is supported by the Office of Science of the US Department of Energy under contract DE-AC03-76SF00098. Online-only material: color figures
Published - DESapj09.pdf