External Enrichment of Mini Halos by the First Supernovae

Abstract

Recent high-resolution simulations of early structure formation have shown that externally enriched halos may form some of the first metal-enriched stars. This study utilizes a 1 comoving Mpc3 high-resolution simulation to study the enrichment process of metal-enriched halos down to z = 9.3. Our simulation uniquely tracks the metals ejected from Population III stars, and we use this information to identify the origin of metals within metal-enriched halos. These halos show a wide range of metallicities, but we find that the source of metals for ≳50% of metal-enriched halos is supernova explosions of Population III stars occurring outside their virial radii. The results presented here indicate that external enrichment by metal-free stars dominates the enrichment process of halos with virial mass below 106 M ⊙ down to z = 9.3. Despite the prevalence of external enrichment in low-mass halos, Population II stars forming due to external enrichment are rare because of the small contribution of low-mass halos to the global star formation rate combined with low metallicities toward the center of these halos resulting from metal ejecta from external sources mixing from the outside in. The enriched stars that do form through this process have absolute metallicities below 10−3 Z ⊙. We also find that the fraction of externally enriched halos increases with time: ∼90% of halos that are externally enriched have M vir < 106 M ⊙, and that pair-instability supernovae contribute the most to the enrichment of the intergalactic medium as a whole and are thus are the predominant supernova type contributing to the external enrichment of halos.

Connecting Primordial Star-forming Regions and Second-generation Star Formation in the Phoenix Simulations

“…Our presentation of N III is largely consistent with their estimate and highlights the dependence on the IMF to generate reasonable estimates of the number of enriching events. Average starforming halos in Hicks et al (2021) contained 20 enriching events, with the most massive halo having >100. Table 2 indicates that the average halo contained a few discrete PIII associations that led to the final state of the halo, while the most massive may have had several.…”

“…-M e yr −1 Mpc −3 , shown on the figure. The z  18 Population III SFRD approaches a relatively constant 10 −4 M e yr −1 Mpc −3 , representing the contribution from both newly collapsing halos in pristine gas and pristine regions of partially enriched halos; since so little of the volume has been enriched, this will continue until more IGM gas has been enriched and the newly collapsed halos are enriched prior to their formation (Hicks et al 2021).…”

“…Feedback (SNe, stellar winds) from each type of star particle contributes to a unique metal density field so that the contributions from Population III and Population II stars can be tracked independently. The star formation and feedback algorithms used in the PHX suite are well documented 4 and described in several prior works (Wise & Cen 2009;Wise et al 2012aWise et al , 2012bXu et al 2013;Hicks et al 2021), so the following is restricted to a high-level overview that includes the specific parameters used in the PHX suite. At each time step, every grid cell is evaluated for Population III star formation according to the following criteria: (1) baryon number density n b > 100 cm −3 , (2) H 2 fraction n H2 /n H > 10 −4 , (3) metallicity 5 Z < Z c for Z c = −5.5, (4) the AMR grid level that is most refined for that point in space, (5) a cooling time that is less than the freefall time, and (6) converging gas flow (∇ • ν < 0).…”

Connecting Primordial Star Forming Regions and Second Generation Star Formation in the Phoenix Simulations

“…Our presentation of N III is largely consistent with their estimate, and highlights the dependence on the IMF to generate reasonable estimates of the number of enriching events. Average star forming halos in Hicks et al (2021) contained 20 enriching events, with the most massive halo having > 100. Table 1 would indicate that the average halo contained a few discrete PIII associations that led to the final state of the halo, while the most massive may have had several.…”

“…Each type of star particle contributes to a unique metal density field so that the contributions from Population III and Population II stars can be tracked independently. The star formation and feedback algorithms used in the Phoenix suite are well documented 2 and described in several prior works Hicks et al 2021), so the following is restricted to a highlevel overview that includes the specific parameters used in the PHX suite. At each time step, every grid cell is evaluated for star formation according to the following criteria: 1) baryon number density n b > 100, 2) for Population III stars, H 2 fraction n H2 /n H > 10 −4 , 3) for Population III stars, metallicity 3 Z < Z c for Z c = −5.5, 4) AMR Grid level is most refined for that location, 5) the freefall time should be less than the cooling time, and 6) converging gas flow (∇ • ν < 0).…”