Introduction
The coincident detection of gravitational waves (GW) and electromagnetic (EM) signals from the first established neutron star binary (NSNS) merger GW170817 provides a unique opportunity to study systematically the properties of compact objects, nuclear physics, and electromagnetism in strong gravity. The simultaneous detection of GW and EM signals from NSNS mergers is the prime target of multimessenger astronomy (MA) and can provide us with important information about high-energy astrophysical phenomena in strong gravity, the synthesis of heavy nuclei, the properties of dense, nuclear matter, etc. In particular, the observation of GW170817 coincident with the sGRB GRB170817A demonstrated that NSNS remnants can power sGRBs. In addition, this GW observation and its association with kilonova AT 2017gfo/DLT17ck indicate that, in contrast to black hole (BH) binary mergers, compact binary mergers where at least one of the companions is a NS are likely to be followed by various processes involving EM and neutrino emission. To systematically interpret multimessenger signals and their relation to the properties of the binary system, it is necessary to perform full general relativistic, magnetohydrodynamic (GRMHD) simulations incorporating detailed microphysical processes.
It has been long established that magnetic fields play a critical role in the fate of NSNS merger remnants. We have shown that NSNS remnants consisting of a BH + disk can launch a collimated, mildly relativistic outflow - an incipient jet - with duration and luminosity consistent with typical sGRB central engine lifetimes and magnitudes, as well as with the Blandford-Znajek mechanism (BZ) for launching jets and their associated Poynting luminosities. We also reported that a jet is launched following the delayed collapse of a hypermassive neutron star (HMNS) remnant if the initial pulsar-like magnetic field in the binary companions has a sufficiently large-scale poloidal component aligned to the orbital angular momentum of the system. It is worth emphasizing that the emergence of a jet does not require the development of a large-scale poloidal magnetic field component in the HMNS, but only initially, as in typical pulsars.
Neutrino processes (e.g., cooling and emission) may also have significant effects on the fate of NSNS merger remnants. It has been suggested that neutrino pair annihilation could carry a large amount of energy from the inner region of the disk. The thinning of the disk may result in a favorable geometry for jet launching, though the duration and energy of neutrino emission in NSNS mergers are likely to be insufficient for the outflows to break out from the ejecta shell and form relativistic jets. It is also believed that neutrino-driven winds, in which neutrinos absorbed in the disk can lift matter out of its gravitational potential, is a crucial mechanism of mass loss for NSNS mergers. Neutrinos may also have important effects on the magnetorotational instability (MRI), which is one of the main mechanisms to amplify the magnetic field to the strength required for jet launching. As the joint contribution and interaction between neutrino transfer and magnetic fields may produce copious interesting effects in compact binary mergers, numerical simulations of mergers with full global and microphysical ingredients are crucial in determining the real physics of the jet-launching mechanism underlying recent observations, such as GRB 170817A and GRB 160625B.
To achieve neutrino transport in compact binary merger systems, simplifications of the exact transport equations have been applied. These include the simplest leakage schemes, which are based on the assumption that the neutrino diffusion timescale is much longer than the weak interaction timescale. More sophisticated are truncated moment formalisms, in particular, the two-moment ("M1") scheme with analytic closure, and a mixed leakage-one moment scheme for evolving binary merger systems.
In this work, we perform GRMHD simulations of NSNS mergers modeled using a piecewise polytropic representation of the nuclear SLy nuclear equation of state (EOS) and initially endowed with a poloidal magnetic field extending from the stellar interior to the exterior, as in pulsars. Shortly after merger, we insert neutrino transfer using an M1 closure scheme. We performed two versions: i) a simplified "warm-up" version involving only one neutrino species ($\bar{\nu}_e$) and considering only charged-current interactions (called "Neutrinos-Simplified"); and ii) a full, more realistic version that evolves three neutrino species and considers additional microphysical processes ("Neutrinos-Full"). We compare our results to simulations that include neither magnetic fields nor neutrino transport ("Unmag") as well as those that include only magnetic fields ("Mag") and those that include only neutrino transport ("Unmag + Neutrinos-Simplified").
We find that neutrino processes enhance the angular momentum transport accelerating the collapse of the HMNS. As a consequence, in neutrino transport cases, the GW waveform is shorter in duration, though its strain amplitude remains above the sensitivity curve of next generation GW observatories. We also find that neutrinos do not have significant impact on the growth of the magnetic field, but they have the effect of clearing out the polar region above the BH poles, inducing a lower baryon-load in surrounding debris. A magnetically-driven jet is launched after $\gtrsim$ 10 ms following the collapse of the HMNS. However, the delay time between the peak GW (i.e. the binary merger) and the emergence of the jet is significantly shorter in neutrino radiation cases. The outgoing EM Poynting luminosity [$L_{\text{EM}} \sim 10^{53}\text{erg s}^{-1}$ ] is roughly consistent with sGRB models and the luminosity associated with the BZ mechanism. While the M1 scheme we adopt is not the most advanced and the results we report here are not likely the final answers, they are sufficient to generate a preliminary sketch of the combined influence of magnetic fields and neutrino transport on compact binary mergers and other astrophysical scenarios in strong gravitational fields.
Simulations were performed on Blue Waters, NASA High-End Computing and XSEDE supercomputer facilities and on SuperMuc_NG through PRACE Grant No. 2018194669. The Illinois GRMHD code, which implements the BSSN formulation of GR with moving-box adaptive mesh refinement, was used for all simulations.
