Context. The reionisation of the Universe is a process that is thought to have ended around z ∼ 6, as inferred from spectroscopy of distant bright background sources, such as quasars (QSO) and gamma-ray burst (GRB) afterglows. Furthermore, spectroscopy of a GRB afterglow provides insight in its host galaxy, which is often too dim and distant to study otherwise. Aims. For the Swift GRB 130606A at z = 5.913 we have obtained a high S/N spectrum covering the full optical and near-IR wavelength region at intermediate spectral resolution with VLT/X-Shooter. We aim to measure the degree of ionisation of the intergalactic medium (IGM) between z = 5.02−5.84 and to study the chemical abundance pattern and dust content of its host galaxy. Methods. We estimated the UV continuum of the GRB afterglow using a power-law extrapolation, then measured the flux decrement due to absorption at Lyα, β, and γ wavelength regions. Furthermore, we fitted the shape of the red damping wing of Lyα. The hydrogen and metal absorption lines formed in the host galaxy were fitted with Voigt profiles to obtain column densities. We investigated whether ionisation corrections needed to be applied. Results. Our measurements of the Lyα-forest optical depth are consistent with previous measurements of QSOs, but have a much smaller uncertainty. The analysis of the red damping wing yields a neutral fraction x H i < 0.05 (3σ). We obtain column density measurements of H, Al, Si, and Fe; for C, O, S and Ni we obtain limits. The ionisation due to the GRB is estimated to be negligible (corrections <0.03 dex), but larger corrections may apply due to the pre-existing radiation field (up to 0.4 dex based on sub-DLA studies). Assuming that [Si/Fe] = +0.79 ± 0.13 is due to dust depletion, the dust-to-metal ratio is similar to the Galactic value. Conclusions. Our measurements confirm that the Universe is already predominantly ionised over the redshift range probed in this work, but was slightly more neutral at z > 5.6. GRBs are useful probes of the ionisation state of the IGM in the early Universe, but because of internal scatter we need a larger statistical sample to draw robust conclusions. The high [Si/Fe] in the host can be due to dust depletion, α-element enhancement, or a combination of both. The very high value of [Al/Fe] = 2.40 ± 0.78 might be due to a proton capture process and is probably connected to the stellar population history. We estimate the host metallicity to be −1.7 < [M/H] < −0.9 (2%−13% of solar).

Long gamma-ray bursts (GRBs), among the most energetic events in the Universe, are explosions of massive and short-lived stars, so they pinpoint locations of recent star formation. However, several GRB host galaxies have recently been found to be deficient in molecular gas (H 2 ), believed to be the fuel of star formation. Moreover, optical spectroscopy of GRB afterglows implies that the molecular phase constitutes only a small fraction of the gas along the GRB line of sight. Here we report the first ever 21 cm line observations of GRB host galaxies, using the Australia Telescope Compact Array, implying high levels of atomic hydrogen (H i), which suggests that the connection between atomic gas and star formation is stronger than previously thought. In this case, it is possible that star formation is directly fuelled by atomic gas (or that the H i-to-H 2 conversion is very efficient, which rapidly exhaust molecular gas), as has been theoretically shown to be possible. This can happen in low-metallicity gas near the onset of star formation because cooling of gas (necessary for star formation) is faster than the H i-to-H 2 conversion. Indeed, large atomic gas reservoirs, together with low molecular gas masses, stellar, and dust masses are consistent with GRB hosts being preferentially galaxies which have very recently started a star formation episode after accreting metal-poor gas from the intergalactic medium. This provides a natural route for forming GRBs in low-metallicity environments. The gas inflow scenario is also consistent with the existence of the companion H i object with no optical counterpart ∼19 kpc from the GRB 060505 host, and with the fact that the H i centroids of the GRB 980425 and 060505 hosts do not coincide with optical centres of these galaxies, but are located close to the GRB positions.

This work studies the optical emission line properties of a sample of 155 low-redshift bright X-ray selected ROSAT Seyfert 1 type AGN for which adequate signal-to-noise ratio spectroscopic observations are available. We measured emission line properties by performing multi-component fits to the emission line profiles, covering the effect of blended iron emission. We also obtained continuum parameters, including 250eV X-ray luminosities derived from the ROSAT database. In addition, the measured properties are gathered for a correlation analysis, which confirms the well-known relations between the strengths of Fe II, [O III] emission and the X-ray slope. We also detect striking correlations between Hβ redshift (or blueshift), flux ratios of Fe II to Hβ broad component and [O III] to Hβ narrow component. These trends are most likely driven by the Eddington ratio. 4 The selection criteria of the sample are: an alternative high X-ray-to-optical flux ratio criterion, i.e., log CR ≥ −0.4 R + 4.9, where CR and R represent X-ray count rate and R magnitude respectively; declination δ ≥ 3 • ; galactic latitude |b| ≥ 20 • ; optical counterparts within a circle with radius r = r 1 + 5 ′′ , where r 1 is the RASS position error given by Voges et al.(1996); optical counterparts with R magnitudes between 13.5 and 16.5.