In this Letter, we propose that a fast radio burst (FRB) could originate from the magnetic interaction between double neutron stars (NSs) during their final inspiral within the framework of a unipolar inductor model. In this model, an electromotive force is induced on one NS to accelerate electrons to an ultra-relativistic speed instantaneously. We show that coherent curvature radiation from these electrons moving along magnetic field lines in the magnetosphere of the other NS is responsible for the observed FRB signal, that is, the characteristic emission frequency, luminosity, duration, and event rate of FRBs can be well understood. In addition, we discuss several implications of this model, including double-peaked FRBs and possible associations of FRBs with short-duration gamma-ray bursts and gravitational-wave events.

Very recently Spitler et al. (2016) and Scholz et al. (2016) reported their detections of sixteen additional bright bursts from the direction of the fast radio burst (FRB) 121102. This repeating FRB is inconsistent with all the catastrophic event models put forward previously for hypothetically nonrepeating FRBs. Here we propose a different model, in which highly magnetized pulsars travel through asteroid belts of other stars. We show that a repeating FRB could originate from such a pulsar encountering lots of asteroids in the belt. During each pulsar-asteroid impact, an electric field induced outside the asteroid has such a large component parallel to the stellar magnetic field that electrons are torn off the asteroidal surface and accelerated to ultra-relativistic energies instantaneously. Subsequent movement of these electrons along magnetic field lines will cause coherent curvature radiation, which can account for all the properties of an FRB. In addition, this model can self-consistently explain the typical duration, luminosity, and repetitive rate of the seventeen bursts of FRB 121102. The predicted occurrence rate of repeating FRB sources may imply that our model would be testable in the next few years.

We use two model-independent methods to standardize long gamma-ray bursts (GRBs) using the E iso − E p correlation (log E iso = a + b log E p ), where E iso is the isotropic-equivalent gamma-ray energy and E p is the spectral peak energy. We update 42 long GRBs and attempt to constrain the cosmological parameters. The full sample contains 151 long GRBs with redshifts from 0.0331 to 8.2. The first method is the simultaneous fitting method. We take the extrinsic scatter σ ext into account and assign it to the parameter E iso . The best-fitting values are a = 49.15 ± 0.26, b = 1.42 ± 0.11, σ ext = 0.34 ± 0.03 and Ω m = 0.79 in the flat ΛCDM model. The constraint on Ω m is 0.55 < Ω m < 1 at the 1σ confidence level. If reduced χ 2 method is used, the best-fit results are a = 48.96 ± 0.18, b = 1.52 ± 0.08, and Ω m = 0.50 ± 0.12. The second method uses type Ia supernovae (SNe Ia) to calibrate the E iso − E p correlation. We calibrate 90 high-redshift GRBs in the redshift range from 1.44 to 8.1. The cosmological constraints from these 90 GRBs are Ω m = 0.23 +0.06 −0.04 for flat ΛCDM and Ω m = 0.18 ± 0.11 and Ω Λ = 0.46 ± 0.51 for non-flat ΛCDM. For the combination of GRB and SNe Ia sample, we obtain Ω m = 0.271 ± 0.019 and h = 0.701 ± 0.002 for the flat ΛCDM and the non-flat ΛCDM, and the results are Ω m = 0.225 ± 0.044, Ω Λ = 0.640 ± 0.082, and h = 0.698 ± 0.004. These results from calibrated GRBs are consistent with that of SNe Ia. Meanwhile, the combined data can improve cosmological constraints significantly, compared to SNe Ia alone. Our results show that the E iso − E p correlation is promising to probe the high-redshift Universe.