High-resolution simulations of the head-on collision of white dwarfs

Abstract

The direct impact of white dwarfs has been suggested as a plausible channel for type Ia supernovae. In spite of their (a priori) rareness, in highly populated globular clusters and in galactic centers, where the amount of white dwarfs is considerable, the rate of violent collisions between two of them might be non-negligible. Even more, there are indications that binary white dwarf systems orbited by a third stellar-mass body have an important chance to induce a clean head-on collision. Therefore, this scenario represents a source of contamination for the supernova light-curves sample that it is used as standard candles in cosmology, and it deserves further investigation. Some groups have conducted numerical simulations of this scenario, but their results show several differences. In this paper we address some of the possible sources of these differences, presenting the results of high resolution hydrodynamical simulations jointly with a detailed nuclear post-processing of the nuclear abundances, to check the viability of white dwarf collisions to produce significant amounts of 56 Ni. To that purpose, we use a 2D-axial symmetric smoothed particle hydrodynamic code to obtain a resolution considerably higher than in previous studies. In this work, we also study how the initial mass and nuclear composition affect the results. The gravitational wave emission is also calculated, as this is a unique signature of this kind of events. All calculated models produce a significant amount of 56 Ni, ranging from 0.1 M ⊙ to 1.1 M ⊙ , compatible not only with normal-Branch type Ia supernova but also with the subluminous and super-Chandrasekhar subset. Nevertheless, the distribution mass-function of white dwarfs favors collisions among 0.6 − 0.7 M ⊙ objects, leading to subluminous events.

Supernovae From Direct Collisions of White Dwarfs and the Role of Helium Shell Ignition

“…Similar differences due to the use of larger nucleosynthetic network have been seen in other works (García-Senz et al 2013), which arise due to the more efficient nuclear burning, producing larger abundances of higher elements (e.g., 56 Ni) and lower amounts of intermediate elements (e.g., 44 Ti).…”

“…Previous collision simulations include several works by different groups (Rosswog et al 2009;Raskin et al 2009Raskin et al , 2010Lorén-Aguilar et al 2010;Hawley et al 2012;Kushnir et al 2013;García-Senz et al 2013). In these simulations only pure CO-WD collisions were simulated (or CO-WD -He-WD collisions Rosswog et al 2009).…”

Observational Clues to the Progenitors of Type Ia Supernovae

“…Katz & Dong (2012) argued that, if a large fraction of all intermediate-mass stars are in the appropriate triple systems, most or all SNe Ia could come from such collisions. Kushnir et al (2013) and García-Senz et al (2013) have performed the latest hydrodynamic thermonuclear simulations of such collisions, and obtained promising agreement with some of the phenomenology of observed SN Ia explosions. Kushnir et al (2013), using high spatial resolution (∼ 1 km) in their simulations, and avoiding some numerical artifacts that affected previous work, find that a collision between, e.g., two 0.7 M WDs produces an explosion with a 56 Ni yield of 0.56 M , similar to a typical SN Ia.…”

Thermonuclear detonations ensuing white dwarf mergers

“…Such initial large hotspots are unrealistic and are not seen in the SPH simulations of WD mergers and they are more characteristic to WD collisions (Rosswog et al 2009a;Raskin et al 2009;AznarSiguán et al 2013;Kushnir et al 2013;García-Senz et al 2013;Papish & Perets 2015). The SPH interaction radius of the particles inside the hot envelope is below ∼ 1000 km, decreasing with increasing density, towards the core where the hotspots are located.…”