Dark Matter Halo Environment for Primordial Star Formation

TL;DR

This study uses cosmological simulations to analyze the shape, spin, and environment of high-redshift halos hosting the first stars, revealing correlations with halo mass and implications for the initial mass function of primordial stars.

Contribution

It provides new insights into the statistical properties of early halos and links these properties to the initial mass function of Population III stars through feedback models.

Findings

Halos are predominantly oblate with low sphericity (~0.3).

Halo spin distributions follow a lognormal distribution with mean ~0.0184 at z=16.

Halo properties influence the initial mass function of the first stars, with feedback models predicting different mass peaks.

Abstract

We study the statistical properties (such as shape and spin) of high-z halos likely hosting the first (PopIII) stars with cosmological simulations including detailed gas physics. In the redshift range considered ($11 < z < 16$) the average sphericity is $<s> = 0.3 \pm 0.1$, and for more than 90% of halos the triaxiality parameter is $T \lesssim 0.4$, showing a clear preference for oblateness over prolateness. Larger halos in the simulation tend to be both more spherical and prolate: we find $s \propto M_h^{\alpha_s}$ and $T \propto M_h^{\alpha_T}$, with $\alpha_s \approx 0.128$ and $\alpha_T= 0.276$ at z = 11. The spin distributions of dark matter and gas are considerably different at $z=16$, with the baryons rotating slower than the dark matter. At lower redshift, instead, the spin distributions of dark matter and gas track each other almost perfectly, as a consequence of a longer time interval available for momentum redistribution between the two components. The spin of both the gas and dark matter follows a lognormal distribution, with a mean value at z=16 of $<\lambda> =0.0184$, virtually independent of halo mass. This is in good agreement with previous studies. Using the results of two feedback models (MT1 and MT2) by McKee & Tan (2008) and mapping our halo spin distribution into a PopIII IMF, we find that at high-$z$ the IMF closely tracks the spin lognormal distribution. Depending on the feedback model, though, the distribution can be centered at $\approx 65 M_\odot$ (MT1) or $\approx 140 M_\odot$ (MT2). At later times, model MT1 evolves into a bimodal distribution with a second prominent peak located at $35-40 M_\odot$ as a result of the non-linear relation between rotation and halo mass. We conclude that the dark matter halo properties might be a key factor shaping the IMF of the first stars.