Millisecond Magnetars As The Central Engine Of Gamma-ray Bursts

Gamma-ray bursts (GRBs), serendipitously discovered in the late sixties by the Vela satel-lites, have puzzled researchers for nearly five decades. The durations of GRBs have a bimodal distribution with short-duration bursts (SGRBs) lasting for less than-2 s and long-duration bursts (LGRBs) greater than-2 s. A large amount of observational evidence indicates that LGRBs originate from the collapse of massive stars. SGRBs, on the other hand, are believed to be the result of binary compact object mergers. The prompt emission of GRBs is mostly likely to be the result of synchrotron radiation of internal shock generated by the persistent activities of central engine. The afterglow of GRBs is observationally indicative of the emission of external shock generated as the relativistic ejecta drive into the surrounding media. Research in the past decades has enabled us to acquire detailed understanding of the host galaxies of LGRBs and SGRBs. LGRBs usually reside in the star-forming areas of young late-type galaxies, i.e., spiral galaxies or irregular galaxies. No LGRB is found in old elliptical galaxies. SGRBs, however, can reside in both young galaxies and old galaxies.Theoretically almost all LGRBs should be associated with supernovae (SNe). Such as-sociation is observed constantly. Current theory predicts that there is one kind of nova that is associated to SGRBs, viz. the kilonova, which is powered by the radioactive decay of r-process material synthesized as the neutron-rich matter unbound from neutron stars is dynam-ically ejected. The evidence for such theory recently gained strong support when researchers discovered an r-process kilonova associated with the SGRB 130603B.Now the study of GRBs is progressed to the stage to identify the nature of central engines, i.e., black holes or millisec-ond magnetars. To help identify the central engine, it is necessary to compare the differences of electromagnetic emission, gravitational wave radiation, and neutrino radiation coming from black holes and magnetars. We elaborate the progress in Chapter 1.The research on black hole as the central engine of GRBs is plenty accumulated. The last decade witnesses the rapid progress in numerical relativistic simulations, which support the idea of black holes as the central engine of GRBs since the simulations find the formation of jets by black holes. Some observational features, however, cannot be easily integrated into the black hole model, for example, the X-ray plateau, the extended emission of SGRBs, X-ray flares that last for 100-104 s. The most concise interpretation for these features is that they are powered by rapidly rotating magnetars. In the past decade much research has been afforded on the characteristics of prompt emission and afterglow of GRBs powered by magnetars.This work is elaborated under this situation. Usually we will first detect the prompt emis-sion of GRBs, and then their afterglow. If the central engine is a magnetar, however, the outcome could be different because the magnetar will dissipate its rotational energy by injecting Poynt-ing flux to the ejecta. Such energy injection will enable an observer outside the jet angle of the GRB to detect the electromagnetic signals. Assuming that the Poynting flux from the magnetar will quickly becomes lepton-dominated in Chapter 2, forward shock will develop when the rel-ativistic ejecta accelerated by the stellar wind drive into the medium. Reverse shock will also develop at the same time. We find that the recently discovered optical transient PTFllagg can be neatly interpreted as synchrotron emission of reverse shock powered by a millisecond mag-netar. PTF11agg-like transients, without the detection of SGRBs, open a unique observational window.In Chapter 3, we consider the absorption of reverse shock emission by the ejecta which is ignored when we study PTF11agg. We find that the ejecta become transparent when the first optical datum of PTF11agg was taken, justifying the treatment of Chapter 2. In this chapter we adopt a more realistic dynamics of the blast wave than adopted in Chapter 2. We find that these two dynamics are quantitatively similar. The ejecta is believed to be pure r-process material which is difficult to study in laboratory. We therefore explore the feasibility to study it astrophysically by observing the X-ray, UV, optical emission obscured by the ejecta. This is possible because the emission at different wavelengths depends on different opacities. It is found that at early time of the reverse shock emission, the opacity at X-ray band is dominated by elastic scattering off free electrons, the opacity at optical band dominated by bound-bound transitions, the opacity at UV band dominated by found-free transitions. As a result, ionization breakout is expected at UV wavelength.In Chapter 4 we consider the effect of inverse Compton (IC) scattering on the electron cooling that was not taken into account in previous chapters. Because the electrons in reverse shock are ultrarelativistic, it is expected that IC emission prominent. The effect of IC on syn-chrotron emission is to reduce its cooling frequency. It is found that the adoption of the more elaborated dynamics discussed in previous chapter and the account of IC is to give the cooling frequency very close to the value determined in Chapter 2 where PTF11agg was studied. To utilize the high-energy telescope to probe the birth of millisecond magnetars, we calculate the IC flux at 1 GeV and 100 GeV.It is found that Fermi/LAT and CTA can detect the IC emission powered by a typical magnetar up to～1 Gpc. The duration of IC emission is of the order of the spin-down timescale of the magnetar, i.e.,1 days.In Chapter 5 we discuss some topics that are on hot debate and the perspective for upcom-ing years.