Because massive, low-metallicity population III (PopIII) stars may produce very powerful long gamma-ray bursts (LGRBs), high-redshift GRB observations could probe the properties of the first stars. We analyze the correlation between early PopIII star
s and LGRBs by using cosmological N-body/hydrodynamical simulations, which include detailed chemical evolution, cooling, star formation, feedback effects and the transition between PopIII and more standard population I/II (PopII/I) stars. From the Swift observed rate of LGRBs, we estimate the fraction of black holes that will produce a GRB from PopII/I stars to be in the range 0.028<f_{GRB}<0.140, depending on the assumed upper metallicity of the progenitor. Assuming that as of today no GRB event has been associated to a PopIII star, we estimate the upper limit for the fraction of LGRBs produced by PopIII stars to be in the range 0.006<f_{GRB}<0.022. When we apply a detection threshold compatible with the BAT instrument, we find that the expected fraction of PopIII GRBs (GRB3) is ~10% of the full LGRB population at z>6, becoming as high has 40% at z>10. Finally, we study the properties of the galaxies hosting our sample of GRB3. We find that the average metallicity of the galaxies hosting a GRB3 is typically higher than the critical metallicity used to select the PopIII stars, due to the efficiency in polluting the gas above such low values. We also find that the highest probability of finding a GRB3 is within galaxies with a stellar mass <10^7 Msun, independently from the redshift.
We study the implication of the collapsar model for Long Gamma-Ray Bursts (LGRBs) on the metallicity properties of the host galaxies, by combining high-resolution N-body simulations with semi-analytic models of galaxy formation. The cosmological mode
l that we use reproduces the Fundamental Metallicity Relation recently discovered for the SDSS galaxies, whereby the metallicity decreases with increasing Star Formation Rate for galaxies of a given stellar mass. We select host galaxies housing pockets of gas-particles, young and with different thresholds in metallicities, that can be sites of LRGB events, according to the collapsar model. The simulated samples are compared with 18 observed LGRB hosts in the aim at discriminating whether the metallicity is a primary parameter. We find that a threshold in metallicity for the LGRB progenitors, within the model galaxies, is not necessary in order to reproduce the observed distribution of host metallicities. The low metallicities of observed LGRB hosts is a consequence of the high star formation environment. The star formation rate appears to be the primary parameter to generate a burst event. Finally, we show that only a few LGRBs are observed in massive, highly extincted galaxies, while these galaxies are expected to produce many such events. We identify these missing events with the fraction of dark LGRBs.
We study the luminosity function (LF), the comoving rate and the detection rate of Long Gamma-Ray Burst (LGRBs) to high redshift, using galaxy catalogues constructed by combining high-resolution N-body simulations with semi-analytic models of galaxy
formation. We assume the collapsar model and different metallicity thresholds, and conclude that LGRBs are not good tracers of the star formation history in the universe. Then using the log N-log P diagram for BATSE bursts, we determine the LF (with and without evolution with redshift) and the formation rate of LGRBs, obtaining constraints on the slope of the power-law. We check the resulting redshift distribution with SWIFT data updated to 2009 August, finding that models where LGRBs have as progenitors stars with Z<0.3Z_sun and without evolution of the LF are in agreement with the data. We also predict that there are about ~1% of GRBs at redshift z>6.
We use galaxy catalogues constructed by combining high-resolution N-body simulations with semi-analytic models of galaxy formation to study the properties of Long Gamma-Ray Burst (LGRB) host galaxies. We assume that LGRBs originate from the death of
massive young stars and analyse how results are affected by different metallicity constraints on the progenitor stars. As expected, the host sample with no metallicity restriction on the progenitor stars provides a perfect tracer of the cosmic star formation history. When LGRBs are required to be generated by low-metallicity stars, they trace a decreasing fraction of the cosmic star formation rate at lower redshift, as a consequence of the global increase in metallicity. We study the properties of host galaxies up to high redshift (~9), finding that they typically have low-metallicity (Z<0.5 Z_sun) and that they are small (M<10^9 M_sun), bluer and younger than the average galaxy population, in agreement with observational data. They are also less clustered than typical L_* galaxies in the Universe, and their descendents are massive, red and reside in groups of galaxies with halo mass between 10^{13} M_sun to 10^{14} M_sun.
