Abstract
Context. The first generation of stars were born a few hundred million years after the big bang. These stars synthesise elements heavier than H and He, which are later expelled into the interstellar medium, initiating the rise of metals. Within this enriched medium, the first dust grains were formed. This event is cosmologically crucial for molecule formation, as dust plays a major role by cooling low-metallicity star-forming clouds, which can fragment to create lower mass stars. Collecting information on these first dust grains is difficult because of the negative alliance of large distances and low dust masses. Aims. We aim to combine the observational information from galaxies at redshifts 5 ≲ z ≲ 10 to constrain their dust emission and theoretically understand the first evolutionary phases of the dust cycle. Methods. Spectral energy distributions (SEDs) are fitted with CIGALE and the physical parameters and their evolution are modelled. From this SED fitting, we built a dust-emission template for this population of galaxies in the reionisation epoch. Results. Our new models explain why some early galaxies are observed and others are not. We follow in time the formation of the first grains by supernovae later destroyed by other supernova blasts and expelled in the circumgalactic and intergalactic media. Conclusions. We find evidence for the first dust grains formed in the universe. But above all, this work underlines the need to collect more data and to develop new facilities to further constrain the dust cycle in galaxies in the reionisation epoch.
Highlights
Understanding the characteristics of the dust cycle is still an issue. This statement is even truer for Lyman break galaxies at all redshifts (LBGs, e.g. Burgarella et al 2007) and at redshifts z > 5 (Bouwens et al 2011; Finkelstein et al 2010) at the epoch of reionization (EoR). This is due to the severe limits due to submillimetre and millimetre observations and because we have a poor idea on the chemical conditions, the dust grain characteristics and even the stellar populations
A necessary preliminary phase was to perform a lot of initial fits on individual Low-zZ and Hi-z LBGs to understand the limits of the physical parameters to be explored (star formation history (SFH), dust attenuation, AGN fraction, etc.) in the CIGALE fits
We see an apparent maximum observed in IRX and sMdust for the present Hi-z LBG sample at sSFR ∼10−8 yr−1 that needs to be confirmed by future observational data. Both IRX and sMdust decline, and age increases to a locus at sSFR∼10−10–10−9 yr−1, where we find the LBGs with upper limits like those discussed in Capak et al
Summary
Understanding the characteristics of the dust cycle is still an issue This statement is even truer for Lyman break galaxies at all redshifts (LBGs, e.g. Burgarella et al 2007) and at redshifts z > 5 (Bouwens et al 2011; Finkelstein et al 2010) at the epoch of reionization (EoR). This is due to the severe limits due to submillimetre (submm) and millimetre (mm) observations and because we have a poor idea on the chemical conditions, the dust grain characteristics and even the stellar populations. SNe blasts produce a shockwave able to destroy dust grains in the ISM (e.g. Slavin et al 2015; Matsuura et al 2019; Nozawa et al 2003; Dwek & Cherchneff 2011)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
