Unveiling the Secrets of Metallicity and Massive Star Formation Using DLAs along Gamma-ray Bursts

TL;DR

This study analyzes a large sample of Damped Lyman-alpha systems along Gamma-ray Burst sightlines to understand their metallicity and star formation environments across redshifts, revealing higher metallicities than quasar DLAs and supporting multiple GRB progenitor channels.

Contribution

It provides the largest publicly available dataset of GRB-DLAs, compares their properties with quasar DLAs, and discusses implications for GRB progenitors and galaxy evolution.

Findings

GRB-DLAs are more metal enriched than QSO-DLAs at z>3.

Metallicity in GRB hosts declines more slowly with redshift than in QSO-DLAs.

Results support the existence of multiple GRB progenitor channels.

Abstract

We present the largest, publicly available, sample of Damped Lyman-$\alpha$ systems (DLAs) along Gamma-ray Bursts (GRB) line of sights in order to investigate the environmental properties of long GRBs in the $z=1.8-6$ redshift range. Compared with the most recent quasar DLAs sample (QSO-DLA), our analysis shows that GRB-DLAs probe a more metal enriched environment at $z\gtrsim3$, up to $[X/H]\sim-0.5$. In the $z=2-3$ redshift range, despite the large number of lower limits, there are hints that the two populations may be more similar (only at 90\% significance level). Also at \hiz, the GRB-DLA average metallicity seems to decline at a shallower rate than the QSO-DLAs: GRB-DLA hosts may be polluted with metals at least as far as $\sim 2$kpc from the GRB explosion site, probably due to previous star-formation episodes and/or supernovae explosions. This shallow metallicity trend, extended now up to $z\sim5$, confirms previous results that GRB hosts are star-forming and have, on average, higher metallicity than the general QSO-DLA population. Finally, our metallicity measurements are broadly consistent with the hypothesis of two channels of GRB progenitors, one of which is mildly affected by a metallicity bias. The metallicity evolution of modeled GRB hosts agrees reasonably well with our data up to intermediate redshift, while more data are needed to constrain the models at $z\gtrsim 4$.