Different Methods to Constrain Dark Matter

Abstract

The currently favored cosmological model suggests that over 85% of the matter in our universe is dark, yet the existence of dark matter is still to be confirmed by detecting it through interactions with normal matter. Direct detection experiments hope to observe signals from the scattering of dark matter particles off of cryogenic target nuclei. A null result from direct detection leads to an exclusion curve in the cross section-dark matter particle mass parameter space. Theoretical predictions for exclusion curves involve standard halo model, in which three astrophysical parameters are assumed to control the distribution of dark matter in the Milky Way. This thesis first discusses the uncertainties in these three parameters on the exclusion curve from the XENON1T experiment. Our estimate done with Monte Carlo simulations shows that at a low WIMP mass, the uncertainty in cross section can span six orders of magnitude. Dark matter self-annihilation might power the first-generation stars and form Dark Stars. The possibility of Dark Stars was originally proposed in the context of Weakly Interacting Massive Particle (WIMP) model. Although the WIMP model is successful in explaining large structures in the universe, it faces difficulties when applied to structures as small as dwarf galaxies. To overcome the small-structure problems, self-interactions between dark matter particles are introduced and the Self-Interacting Dark Matter (SIDM) model was proposed. In the second part of this thesis, we evaluate the probability that Dark Stars can be powered by SIDM. We first propose a simple particle physics model of SIDM that satisfies all the current constraints, and work out the phase space region in which Dark Stars can form. Then we investigate the gravothermal evolution of SIDM minihalos in the presence of a gas potential, and investigate whether it can lead to a sufficiently high dark matter density for Dark Stars to form. Finally, we present the first study of the properties of Dark Stars assuming they can reach hydrostatic equilibrium. Dark matter is a major player in the formation of Milky Way-like galaxies. Different dark matter models lead to a different accretion history of Milky Way-like galaxies. This thesis finally studies the recent accretion history of Milky Way-like galaxies using statistical cluster analysis. Stars from the same accreted satellite galaxy are clustered in action space. Since actions are conserved in slow enough gravitational evolution, the accreted satellites should remain clustered until today. We apply the cluster analysis algorithm Enlink to accreted star particles in action space from the halos of three simulated Milky Way-like galaxies in the FIRE-2 simulations. We compare the groups found by our cluster analysis with the actual accreted satellites from these galaxies, and find the well-recovered satellites. The results show that the member stars of satellites which fell into the galaxy less than 7.1 Gyr ago and were more massive than 4.0 times 10^8 solar mass can be well recovered by cluster analysis.