Dust reddening and extinction curves toward gamma-ray bursts at z > 4 - Bolmer et al. (A&A, 2018)
Dust is known to be produced in the envelopes of asymptotic giant branch (AGB) stars and the expanded shells of supernovae (SN) remnants. During the first Gyr after the Big Bang, there has not been enough time to form AGB stars in high numbers, so the dust at this epoch is expected to be purely from SN. The time period corresponding to z ~ 5-6 is thus expected to display the transition from SN-only dust to a mixture of both formation channels as we know it today. Here we aim to use afterglow observations of Gamma-ray Bursts (GRBs) at redshifts larger than z > 4 in order to derive host galaxy dust column densities along their line-of-sight and to test if a SN-type dust extinction curve is required for some of the bursts. GRB afterglow observations were performed with the 7-channel „Gamma-Ray Optical and Near-infrared Detector“ (GROND) at the 2.2m MPI telescope in La Silla, Chile (ESO) and combined with quasi-simultaneous data gathered with the XRT telescope on board of the Swift-satellite. We increase the number of measured A_V values for GRBs at z > 4 by a factor of ~2-3 and find that, in contrast to samples at mostly lower redshift, all of the GRB afterglows have a visual extinction of A_V < 0.5 mag. Analysis of the GROND detection thresholds and results from a Monte-Carlo simulation show that, although we clearly suffer from an observational bias towards highly extinguished sight-lines, GRB host galaxies at 4 < z < 6 seem to contain on average less dust than at z ~ 2. Additionally, we find that all of the GRBs can be modeled with locally measured extinction curves and that the SN-like dust extinction curve provides a better fit for only two of the afterglow SEDs, indicating that there is no need to assume different dust properties at z > 4. For the first time we also report a photometric redshift of z ~ 7.88 for GRB 100905A, making it one of the most distant GRB known to date.
Fig. 2: Contour plot of the ∆χ2 values for each of fitted host-intrinsic visual extinction AV and redshift (z) parameters for the best-fit power-law index of β = 0.45. For three degrees of freedom, the significance levels of 1σ (68.27%), 2σ (95.45%), 3σ (99.73%) and 4σ (99.99%) correspond to ∆χ2 = 3.53, 8.02, 11.35 and 21.11, respectively. As shown in the left inset, the GROND SED is best-fit with the SMC extinction curve, a power-law slope of β = 0.45, no dust extinction (AV = 0.00 mag) and a photometric redshift of z ~ 7.88. In the inset on the right we also show the corresponding redshift probability density function. The gray shaded area indicates the 1σ confidence interval.
Fig.1: Spectral energy distribution for the 22 GRBs analyzed in my paper with increasing redshift from the bottom to the top (as labeled). The data for GRBs detected with GROND is plotted with circles, data from other instruments with diamonds. The X-ray spectrum, if not available simultaneously to the NIR/optical data, was flux normalized to the mid-time of the chosen GROND exposure. The dashed lines indicate the unabsorbed best-fit models. The solid lines indicate the best-fit model including absorption: in the X-rays due to galactic plus host intrinsic absorption by medium weight metals; in the NIR/optical range due to host intrinsic absorption by dust.
Evidence for diffuse molecular gas and dust in the hearts of gamma-ray burst host galaxies
The true nature of high-redshift GRB DLAs
Bolmer et al. (in prep)
Damped Lyman-α absorbing systems associated to gamma-ray burst afterglows offer a unique way to study the physical conditions in high-redshift star-forming galaxies. Here we build up a large sample of 22 GRBs at redshift z > 2 observed with VLT/X-shooter to determine the abundances of hydrogen, metals, dust, and molecules, and study the metallicity, dust depletion and nucleosynthesis effects in the high-redshift interstellar medium, to answer the question whether (and why) there might be a lack of H2 in GRB-DLAs. We have developed new, state-of-the-art methods, based on the Python Bayesian inference package PyMC, to fit absorption lines, in order to measure the column densities of 10 different metal species as well as neutral atomic and molecular hydrogen. The derived relative abundances are further used to fit dust depletion sequences and determine the dust-to-metals ratio and the host-galaxy intrinsic visual extinction. Additionally, we searched for absorption from carbon monoxide and vibrationally excited H2. There is no lack of H2-bearing GRB-DLAs. We detect absorption from molecular hydrogen in 6 out of 22 GRB spectra, with molecular fractions between f ≃ 5 · 10−5 and f ≃ 0.04, and claim tentative detections in three more cases. For the rest of the bursts we measure, depending on S/N, spectral coverage and instrumental resolution, more or less constraining upper limits. The GRB DLAs in our sample have on average low metallicities ([X/H] ≈ −1.2), comparable to extremely strong QSO DLAs. Furthermore, we find that all H2-bearing systems have significant depletion factors, [Y/Fe] > 0.3, and dust-to-metals ratios, DTM > 0.4. Also, all H2-bearing DLAs have neutral hydrogen column densities larger than log N(HI) > 21.7. Compared to QSO-DLA samples, the fraction of H2 detections is, with 27% (or 37% for z < 4), about 20% higher, which is likely a result of the fact that GRB-DLAs on average probe higher neutral hydrogen column densities, that indicate that the absorbing gas is associated with the inner regions of the host galaxy, where the gas pressure is higher and the conversion of HI to H2 takes place. This diffuse molecular gas is likely at distances of > 500 pc, unrelated to the star-forming region where the GRB progenitor was born.