The successful joint observation of the gravitational wave (GW) event GW170817 and its multiwavelength electromagnetic counterparts enabled us to witness a definite merger event of two neutron stars (NSs) for the first time. This historical event confirms the origin of short-duration gamma-ray bursts (GRBs) and, in particular, identifies the theoretically predicted kilonova phenomenon that is powered by radioactive decays of r-process heavy elements. However, whether or not a long-lived remnant NS could be formed during this merger event remains unknown; though, such a central engine has been suggested by afterglow observations of some short-duration GRBs. By invoking this long-lived remnant NS, we propose a model of hybrid energy sources for the kilonova AT 2017gfo associated with GW170817. While the early emission of AT 2017gfo is still powered radioactively, as is usually suggested, its late emission is primarily caused by delayed energy injection from the remnant NS. In our model, only one single opacity is required and an intermediate value of κ ≃ 0.97 cm² g⁻¹ is revealed, which could be naturally provided by lanthanide-rich ejecta that are deeply ionized by the emission from a wind of the NS. These self-consistent results indicate that a long-lived remnant NS, which must have a very stiff equation of state, was formed during the merger event of GW170817. This provides a very stringent constraint on the strong interaction in nuclear-quark matter. It is further implied that such GW events could provide a probe of the early spin and magnetic evolutions of NSs, e.g., the burying of surface magnetic fields.

The groundbreaking discovery of the optical transient AT2017gfo associated with GW170817 opens a unique opportunity to study the physics of double neutron star (NS) mergers. We argue that the standard interpretation of AT2017gfo as being powered by radioactive decay of r-process elements faces the challenge of simultaneously accounting for the peak luminosity and peak time of the event, as it is not easy to achieve the required high mass, and especially the low opacity of the ejecta required to fit the data. A plausible solution would be to invoke an additional energy source, which is probably provided by the merger product. We consider energy injection from two types of the merger products: (1) a post-merger black hole powered by fallback accretion; and (2) a long-lived NS remnant. The former case can only account for the early emission of AT2017gfo, with the late emission still powered by radioactive decay. In the latter case, both early- and late-emission components can be well interpreted as due to energy injection from a spinning-down NS, with the required mass and opacity of the ejecta components well consistent with known numerical simulation results. We suggest that there is a strong indication that the merger product of GW170817 is a long-lived (supramassive or even permanently stable), low magnetic field NS. The result provides a stringent constraint on the equations of state of NSs.

By fitting the bolometric light curves of 31 superluminous supernovae (SLSNe) with the magnetar engine model, we derive the ejecta masses and magnetar parameters for these SLSNe. The lower boundary of magnetic field strengths of SLSN magnetars can be set just around the critical field strength of the electron Landau quantization. In more details, SLSN magnetars can further be divided into two subclasses of magnetic fields of and , respectively. It is revealed that these two subclasses of magnetars are just associated with the slow-evolving and fast-evolving bolometric light curves of SLSNe. In comparison, the magnetars harbored in gamma-ray bursts (GRBs) and associated hypernovae are usually inferred to have much higher magnetic fields with a lower boundary of about . This robustly suggests that it is the magnetic fields that play the crucial role in distinguishing SLSNe from GRBs/hypernovae. The rotational energy of SLSN magnetars is found to be correlated with the masses of supernova ejecta. This possible correlation provides a clue to explore the nature of their progenitors. Moreover, the distribution of ejecta masses of SLSNe is basically intermediate between those of normal core-collapse supernovae and hypernovae. This could indicate an intrinsic connection among these different stellar explosions.

Neutron star mergers are believed to occur in accretion disks around supermassive black holes. Here we show that a putative jet launched from the merger of a binary neutron star (BNS) or a neutron star–black hole (NSBH) merger occurring at the migration trap in an active galactic nucleus (AGN) disk would be choked. The jet energy is deposited within the disk materials to power a hot cocoon. The cocoon is energetic enough to break out from the AGN disk and produce a bright X-ray shock breakout transient peaking at ∼0.15 days after the merger. The peak luminosity is estimated as , which can be discovered by the Einstein Probe from . Later on, the nonrelativistic ejecta launched from the merger would break out the disk, powering an X-ray/UV flare peaking at ∼0.5 days after the merger. This second shock breakout signal may be detected by UV transient searches. The cocoon cooling emission and kilonova emission are outshone by the disk emission and are difficult to detect. Future joint observations of gravitational waves from BNS/NSBH mergers and associated two shock breakout signatures can provide strong support for the compact binary coalescence formation channel in AGN disks.