Long duration gamma-ray bursts (GRBs), the brightest events since the Big Bang itself, are believed to originate in an ultra-relativistic jet breaking out from a massive stellar envelope. Despite decades of study, there is still no consensus on their emission mechanism. One unresolved question is the origin of the tight correlation between the spectral peak energy and peak luminosity discovered in observations. This Yonetoku relation is the tightest correlation found in the properties of the prompt phase of GRB emission, providing the best diagnostic for the radiation mechanism. Here we present three-dimensional hydrodynamical simulations, and post-process radiation transfer calculations, of photospheric emission from a relativistic jet. Our simulations reproduce the Yonetoku relation as a natural consequence of viewing angle. Although jet dynamics depend sensitively on luminosity, the correlation holds regardless. This result strongly suggests that photospheric emission is the dominant component in the prompt phase of GRBs.

We explore the physics of relativistic radiation mediated shocks (RRMSs) in the regime where photon advection dominates over photon generation. For this purpose, a novel iterative method for deriving a self-consistent steady-state structure of RRMS is developed, based on a Monte-Carlo code that solves the transfer of photons subject to Compton scattering and pair production/annihilation. Systematic study is performed by imposing various upstream conditions which are characterized by the following three parameters: the photon-to-baryon inertia ratio ξ_u*, the photon-to-baryon number ratio $$\tilde{n}$$, and the shock Lorentz factor γ_u. We find that the properties of RRMSs vary considerably with these parameters. In particular, while a smooth decline in the velocity, accompanied by a gradual temperature increase is seen for ξ_u* ≫ 1, an efficient bulk Comptonization, that leads to a heating precursor, is found for ξ_u* ≲ 1. As a consequence, although particle acceleration is highly inefficient in these shocks, a broad non-thermal spectrum is produced in the latter case. The generation of high energy photons through bulk Comptonization leads, in certain cases, to a copious production of pairs that provide the dominant opacity for Compton scattering. We also find that for certain upstream conditions a weak subshock appears within the flow. For a choice of parameters suitable to gamma-ray bursts, the radiation spectrum within the shock is found to be compatible with that of the prompt emission, suggesting that subphotospheric shocks may give rise to the observed non-thermal features despite the absence of accelerated particles.

We present the first global model of prompt emission from a short gamma-ray burst (GRB) that consistently describes the evolution of the central black hole (BH) torus system, the propagation of the jet through multicomponent merger ejecta, the transition into free expansion, and the photospheric emission from the relativistic jet. To this end, we perform a special relativistic neutrino-hydrodynamics simulation of a viscous BH-torus system, which is formed about 500 ms after the merger and is surrounded by dynamical ejecta as well as neutron star winds, along with a jet that is injected in the vicinity of the central BH. In a postprocessing step, we compute the photospheric emission using a relativistic Monte Carlo radiative transfer code. It is found that the wind from the torus leaves a strong imprint on the jet as well as on the emission, causing narrow collimation and rapid time variability. The dependence of the emission on viewing angle gives rise to correlations among the spectral peak energy, E _p , isotropic energy, E _iso, and peak luminosity, L _p , which may provide natural explanations for the Amati and Yonetoku relations. We also find that the degree of polarization is small for emission from the jet core (≲2%), while it tends to increase with viewing angle outside the core and can become as high as ∼10%–40% for energies larger than the peak energy. Finally, the comparison of our model with GRB 170817A strongly disfavors the photospheric emission scenario and therefore supports alternative scenarios, such as cocoon shock breakout.