Jets from Neutron-Star Merger Remnants and Massive Blue Kilonovae

We perform three-dimensional general-relativistic magnetohydrodynamic simulations with weak interactions of binary neutron-star (BNS) mergers resulting in a long-lived remnant neutron star, with properties typical of galactic BNS and consistent with those inferred for the first observed BNS merger GW170817. We demonstrate self-consistently that within $\lesssim 30\mathrm{ms}$ postmerger magnetized ($\sigma \sim 5–10$) incipient jets emerge with asymptotic Lorentz factor $\mathrm{\Gamma }\sim 5–10$, which successfully break out from the merger debris within $\lesssim 20\mathrm{ms}$. A fast ($v\lesssim 0.6c$), magnetized ($\sigma \sim 0.1$) wind surrounds the jet core and generates a UV/blue kilonova precursor on timescales of hours, similar to the precursor signal due to free neutron decay in fast dynamical ejecta. Postmerger ejecta are quickly dominated by magnetohydrodynamically driven outflows from an accretion disk. We demonstrate that, within only 50 ms postmerger, $\gtrsim 2\times {10}^{-2}{M}_{\odot }$ of lanthanide-free, quasispherical ejecta with velocities $\sim 0.1–0.2c$ is launched, yielding a kilonova signal consistent with GW170817 on timescales of $\lesssim 5\mathrm{d}$.

Figure 1

Top: average toroidal field in the remnant NS as a function of time after merger (blue solid line). The dashed line represents the expected amplification by the Kelvin-Helmholtz instability, the dot-dashed line the turbulent amplification by large-scale vortices, and the dotted line the expected linear growth by magnetic winding. Bottom: 3D rendering of magnetic field lines at 15 (left) and 50 ms (right), showing the conversion of turbulent fields into large-scale toroidal structures that give rise to twin jets.

Figure 2

Meridional snapshot showing the magnetic-to-fluid pressure ratio (upper half-plane) and the net specific neutrino heating rate (lower half-plane) once a stationary jet structure has emerged.

Figure 3

Meridional snapshot of radial velocity showing the successful breakout of a high-velocity, high-entropy jet structure from the surrounding merger debris (dynamical and postmerger wind ejecta; see Supplemental Material [67]). The white contour contains bound material according to the Bernoulli criterion (the accretion disk). Black contours contain dynamically unbound ejecta (geodesic criterion). The rest of the outflow domain is unbound according to the Bernoulli criterion.

Figure 4

Total unbound mass flux ${\dot{M}}_{\mathrm{tot}}$ through a spherical shell with radius 300 km and associated density-averaged local four-velocity $u=v^r\mathrm{\Gamma }$, magnetization $\sigma$, and expected velocity at the magnetosonic surface, ${\sigma }^{1/3}$, in polar regions ($\theta \lesssim 30°$).

Figure 5

Top: unbound mass distribution in terms of electron fraction (left; at 5 GK) and asymptotic expansion velocity (right) as sampled by tracer particles, normalized by the total ejected mass, with separate histograms for the polar outflows. Bottom: total final nucleosynthetic abundances at ${10}^9\mathrm{s}$ from reaction network calculations for postmerger ejecta, compared to observed solar abundances [77] (arbitrary normalization) with and without including neutrino absorption during $r$-process nucleosynthesis.

Figure 6

Kilonova light curves of the $\approx 2\times {10}^{-2}{M}_{\odot }$ postmerger ejecta in various UV and blue bands, compared to observed data (dots) and upper limits (triangles) of the GW170817 kilonova [8] and computed for a distance of 40 Mpc with an observer angle of 35° with respect to the rotational axis, as inferred for GW170817 [3, 80].