Introduction
The exciting prospect of simultaneous observations of both gravitational waves and electromagnetic signals originating from the coalescence and merger of binary neutron stars (NSNSs) makes these systems, along with black hole-neutron star binaries (BHNSs), prime targets for the LIGO/Virgo scientific collaboration in the era of multimessenger astronomy. These systems had long been hypothesized as progenitors of the same central engines that power short-hard gamma-ray bursts (sGRBs), which was supported by the first detection of a kilonova associated with the sGRB ``GRB130603B''.
The strongest theoretical support for this hypothesis came from self-consistent, fully general relativistic magnetohydrodynamic (GRMHD) simulations of BHNS and NSNS mergers that showed that an incipient jet may be launched if the NS is suitably magnetized. Nevertheless, the detection of GW170817 coincident with a sGRB (event GRB170817A), as well as its association with kilonova AT 2017gfo/DLT17ck, provides the best direct observational evidence so far that some sGRBs are indeed powered by NSNS mergers, or at least by the merger of a compact binary where at least one of the companions is a NS. Note that the progenitor of GW170817 has been identified as an NSNS based on the masses of the companions; depending on the spin priors of the binary companions, their inferred masses are in the broad range of $0.86-2.26M_\odot$, though the total mass of the system is constrained to be $2.73-3.29 M_\odot$ with $90\%$ credibility. These masses are consistent with astrophysical observations of NSs, but it cannot rule out the presence of a stellar-mass BH. Recently, X-ray observations have strongly suggested that the rapidly rotating, giant star 2MASS J05215658+4359220 is the binary companion of a noninteracting $\sim 3M_\odot$ BH. So, there may be a population of stellar-mass BHs missed by X-ray observations that eventually may form GW170817-like binary systems.
We perform full 3D GRMHD simulations of NSNS configurations in quasicircular orbits that merge and undergo delayed collapse to a BH. The binaries consist of two identical, uniformly rotating NSs modeled with a $\Gamma=2$ polytropic EOS with spin $\chi_{NS}\equiv J_{ql}/(M/2)^2=0.36$, where $J_{ql}$ is the quasilocal angular momentum of the NS, and $M$ is the Arnowitt-Deser-Misner (ADM) mass of the system. The spins are aligned with the orbital angular momentum of the system $L$. We choose highly spinning NSNS configurations to reduce computational costs, because, as we recently showed, the higher the initial spin of the binary companions the shorter the jet launching time. Each star is initially endowed with a dipolar magnetic field of the same magnitude extending from the stellar interior into its exterior and whose dipole moment is either aligned or perpendicular to $L$. We consider the following configurations:
Ali-Ali case: The magnetic dipole moment in both stars is aligned to $L$.
Ali-Per case: The magnetic dipole moment in one of the stars is aligned to $L$, while in the other is perpendicular to it.
Per-Per case: The magnetic dipole moment in both stars is perpendicular to $L$.
Note that these three cases can be used to infer the outcome of general cases in which the dipole moment of the seed magnetic field is misaligned by an angle $\theta\leq 90^\circ$ to the spin of the NS. For comparison purposes, we also consider a second Ali-Ali case in which symmetry across the orbital plane (equatorial symmetry) is imposed.
