After the first three direct detections of gravitational waves (GWs)--signals produced by the inspiral and merger of binary black hole (BHBH) systems--it may be just a matter of time before GWs from merging binary neutron stars (NSNS) or black hole-neutron star (BHNS) systems are detected as well. Estimates from population synthesis and the current sensitive volume of the Advanced Laser Interferometer Gravitational Wave Observatory (aLIGO) predict detection rates of $\lesssim 20$ events per year for NSNS systems, and $\lesssim 4$ events per year for BHNS systems.

Merging NSNSs and BHNSs are not only important sources of gravitational radiation but also promising candidates for coincident detections of electromagnetic (EM) counterparts. These systems are thought to be the progenitors of short gamma-ray bursts (sGRBs). Coincident detection of GWs with EM bursts could give new insight about the sources.

Recently, self-consistent calculations in full general relativity (GR) have shown that the outcome of the merging magnetized BHNS and NSNS systems is a collimated, mildly relativistic outflow--an incipient jet. In the first scenario, the key ingredient for jet launching is the existence of a strong poloidal B-field component after disruption, which can be achieved if initially the NS is endowed with a dipole B-field that extends from the NS interior into the exterior. In the NSNS scenario, by contrast, jets arise whether or not the B-field is confined to the NS interior. It was shown that binary NSNSs that start from the late inspiral and undergo $\textit{delayed}$ collapse to a BH launch jets ~ 44 (MNS/1.8M) ms following the NSNS merger. The burst duration found was Δt ~ 97 (MNS/1.8M) ms, which is consistent with short sGRBs.

Here we present numerical results for $\textit{prompt}$ collapse scenarios. They show that the magnetic energy is amplified during the NSNS merger, but the absence of a transient hypermassive neutron star remnant, which forms in the delayed collapse case prior to collapse, prevents the magnetic energy from reaching saturation levels. In contrast to the delayed collapse case, the magnetic energy decreases following BH formation, and after ~ 20 (MNS/1.8M) ms, we do not observe any indication of jet outflow or B-field collimation. Only NSNS systems that undergo delayed collapse are therefore viable models of engines that power sGRBs.

Simulations were performed on the Blue Waters supercomputer at UIUC. The Illinois GRMHD code, which implements the BSSN formulation of GR with moving box adaptive mesh refinement, was used for all simulations.