Ultra-stripped supernovae: progenitors and fate

The explosion of ultra-stripped stars in close binaries can lead to ejecta masses <0.1 M-circle dot and may explain some of the recent discoveries of weak and fast optical transients. In Tauris et al., it was demonstrated that helium star companions to neutron stars (NSs) may experience mass transfer and evolve into naked similar to 1.5 M-circle dot metal cores, barely above the Chandrasekhar mass limit. Here, we elaborate on this work and present a systematic investigation of the progenitor evolution leading to ultra-stripped supernovae (SNe). In particular, we examine the binary parameter space leading to electron-capture (EC SNe) and iron core-collapse SNe (Fe CCSNe), respectively, and determine the amount of helium ejected with applications to their observational classification as Type Ib or Type Ic. We mainly evolve systems where the SN progenitors are helium star donors of initial mass M-He = 2.5-3.5 M-circle dot in tight binaries with orbital periods of P-orb = 0.06-2.0 d, and hosting an accreting NS, but we also discuss the evolution of wider systems and of both more massive and lighter - as well as single - helium stars. In some cases, we are able to follow the evolution until the onset of silicon burning, just a few days prior to the SN explosion. We find that ultra-stripped SNe are possible for both EC SNe and Fe CCSNe. EC SNe only occur for M-He = 2.60-2.95 M-circle dot depending on Porb. The general outcome, however, is an Fe CCSN above this mass interval and an ONeMg or CO white dwarf for smaller masses. For the exploding stars, the amount of helium ejected is correlated with Porb - the tightest systems even having donors being stripped down to envelopes of less than 0.01 M-circle dot. We estimate the rise time of ultra-stripped SNe to be in the range 12 h-8 d, and light-curve decay times between 1 and 50 d. A number of fitting formulae for our models are provided with applications to population synthesis. Ultra-stripped SNe may produce NSs in the mass range 1.10-1.80 M-circle dot and are highly relevant for LIGO/VIRGO since most (possibly all) merging double NS systems have evolved through this phase. Finally, we discuss the low-velocity kicks which might be imparted on these resulting NSs at birth.