Simulating a white dwarf dominated Galactic halo

Observational evidence has suggested the possibility of a Galactic halo dominated by white dwarfs (WDs). While debate continues concerning the interpretation of this evidence, it is clear that an initial mass function (IMF) biased heavily toward WD precursors (1<~m/Msolar<~ 8), at least in the early Universe, would be necessary in generating such a halo. Within the framework of homogeneous, closed-box models of Galaxy formation, such biased IMFs lead to an unavoidable overproduction of carbon and nitrogen relative to oxygen (as measured against the abundance patterns in the oldest stars of the Milky Way). Using a three-dimensional Tree N-body smoothed particle hydrodynamics code, we study the dynamics and chemical evolution of a galaxy with different IMFs. Both invariant and metallicity-dependent IMFs are considered. Our variable IMF model invokes a WD-precursor-dominated IMF for metallicities less than 5 per cent solar (primarily the Galactic halo), and the canonical Salpeter IMF otherwise (primarily the disc). Halo WD density distributions and C,N/O abundance patterns are presented. While Galactic haloes composed of >~5 per cent (by mass) of WDs are not supported by our simulations, mass fractions of ~1-2 per cent cannot be ruled out. This conclusion is consistent with present-day observational constraint