*2.4. Progenitors and Open Questions*

Beside the above discussion, the working definition of LGRBs and SGRBs suggests the existence of two different progenitor channels. In summary,


The huge observed isotropic equivalent energy release of <sup>∼</sup>1049–10<sup>55</sup> erg implies that: for LGRBs, up to ∼10 *M* are converted into radiation during the prompt emission duration of ∼100 s, whereas SGRBs up to ∼1 *M* are converted into radiation within ∼1 s [25]. The energy reservoir and the efficiency of the involved physical processes in producing the emitted energy represent a stringent requirement, especially for LGRBs<sup>4</sup> .

The commonly called *jets* substantially alleviate this issue by reducing the GRB energy release by jet's correction factor *f* = 1 − cos *θ*. Jets can be thought of, in an oversimplified picture, as outflows of relativistic matter ejected into a double-cone structure of opening angle *θ*. In general, the jet correction is poorly constrained because it requires very challenging measurements of *θ* and the observer viewing angle relative to the jet axis. This makes it troublesome to distinguish between geometric and dynamical effects. Indeed, very soft GRBs could be bursts viewed off-axis, whereas low luminosity GRBs may be the result of large jet opening angles [15].

Measurements of *θ* can be obtained by the predicted signature of the achromatic *jet breaks*, observable in the afterglow light curve at all frequencies. This feature can be explained by the dynamics of the GRB ejecta as follows. At the beginning, at high values of the bulk Lorentz Γ factor<sup>5</sup> , the ejecta is narrowly beamed into the jets while its Lorentz factor is Γ <sup>−</sup><sup>1</sup> < *θ* and, regardless of the hydrodynamic evolution, a GRB is observed only from a small fraction of the ejecta [15]. As the ejecta decelerates, Γ decreases below *θ* −1 , the beaming angle becomes larger, and a larger portion of the ejecta becomes observable. Continuous deceleration leads to the point that the entire surface of the jet is observable and the jet begins to spread sideways, producing a break in the light curve across the entire afterglow spectrum [27,28]. The sharpness of this break and the change in the afterglow decay rate depend on how long the jet remains collimated and on the jet radial density profile and energy distribution [29,30]. The time of the jet break is related to the jet opening angle, the bulk Lorentz factor, and the density of the circumburst medium (CBM). The above description has two effects:


In the pre-*Swift* era, simultaneous breaks in the optical and near-infrared (NIR) afterglow light curves were frequently interpreted as jet breaks. Nevertheless, the improved temporal and spectral coverage of GRB afterglows, especially in X-rays by *Swift*, have revealed within the first few hours after the prompt emission a complex structure made of flares, plateaus, and chromatic breaks [31–34]. The detected achromatic breaks are observed in a few cases. The absence of jet-break signatures in most GRB afterglows has been interpreted as due to the over-simplified assumption homogenous jets with sharp edges, whereas more complex models now include structured jets that produce several chromatic jet-breaks, or much smoother breaks, or jets that can keep their structure for longer than previously thought making difficult to detect breaks without a wide temporal coverage [30].

Besides the jet modeling issue, any GRB model has to deal with features like very luminous X-ray flares occurring up to a few 10<sup>4</sup> s after the GRB trigger and with shape and spectra similar to those flares observed during prompt emission and extended plateau phases that last for a few hours during the early afterglow evolution [33,34]. Both features imply an extended central-engine activity with a continuous source of energy injection lasting the above 10<sup>4</sup> s. In the standard picture, such long-lived energy injection requires the accretion of a significant mass onto the central BH via very large (∼ 1 M) and lowviscosity (*α* < 10−<sup>2</sup> ) accretion disk formed at the core collapse time, or via fall-back material continuously replenishing the accretion disk [35].
