3.1.3. Magnetized (or Poynting-Flux Dominated) Outflows

The Poynting-flux dominated model speculates that the gravitational energy produces very strong magnetic fields, which may be crucial in the jet formation of GRBs, similarly to the Active Galactic Nuclei (AGN), where magnetic energy is converted into particle acceleration via Blandford–Znajek [83] or Blandford–Payne [84] mechanisms. The idea behind this model is that the collapse of a white dwarf (WD) induced by accretion from a massive star, the core collapse of a massive star, or NS merger does not immediately form a BH, but rather a rapidly-spinning (with a period of ∼ 1 ms) and highly-magnetized NS (with a magnetic field *B* & 10<sup>15</sup> G) NS, known as *magnetar* [70]. The maximum amount of magnetic energy that can be stored is <sup>∼</sup> <sup>2</sup> <sup>×</sup> <sup>10</sup><sup>52</sup> erg, and it can be extracted in a short timescale of ∼ 10 s and drives a jet along the polar axis of the NS powering the prompt emission [71]. The decay of rotational or magnetic energy may continue to power late time flaring or afterglow emission. The dipole radiation naturally produces a plateau phase up to the dipole spin-down time scale [15].

In this model, the magnetic field is essentially toroidal (i.e., <sup>~</sup>*<sup>B</sup>* <sup>⊥</sup> <sup>~</sup>*β*) and its polarity in the flow changes on small scale defined by the light cylinder in the central engine. The total luminosity is given by *L* = *L<sup>k</sup>* + *LM*, where *L<sup>k</sup>* = Γ*Mc* ˙ <sup>2</sup> is the kinetic part and *L<sup>M</sup>* = 4*πr* 2 *c*[*B* <sup>2</sup>/(4*π*)] is the magnetic part [85,86]. The key parameter is the magnetization *σ* ≡ *LM*/*L<sup>k</sup>* = *B* <sup>2</sup>/(4*π*Γ <sup>2</sup>*nmpc* 2 ), which plays a similar role to the baryon load in the classical model and defines the maximum attainable Lorentz factor Γmax ≈ *σ* 3/2, whereas, during the acceleration phase, one gets Γ(*r*) ∝ *r* 1/3 [85,86].

In this model, the rapid variability observed in GRBs and the low efficiency in dissipating the kinetic energy via shock waves in highly magnetized plasmas are still open issues. Recent recipes suggest that central engine variability leads to the ejection of magnetized plasma shells which expand due to internal magnetic pressure gradient and collide at a distance *rcol*. The ordered magnetic field lines of the ejecta get distorted and fast reconnection occurs. The induced relativistic turbulence may be able to overcome the low efficiency difficulty of the classical internal shock scenario [87].
