Down but not out: the white dwarf survivors of low-luminosity thermonuclear supernovae


We now know that there are a large variety of thermonuclear supernovae (SNe) with white-dwarf (WD) progenitors, of which Type Ia supernovae (SNe Ia) are the most common class. Roughly 10–30% of WD transients are relatively fast, low-luminosity, and low-energy events that likely have different progenitors from SNe Ia. While SNe Ia should fully disrupt their progenitor WD, lower-energy explosions may not unbind the WD, leaving behind a battered and bruised star that should have distinct observational properties. Given the rate of peculiar transients, there should be \(\sim\)10\({}^{6}\) such stars in the Milky Way. Such an odd WD was recently discovered, further indicating that the Milky Way holds unique opportunities to understand peculiar transients. There should be one of these stars at the center of peculiar thermonuclear transient remnants. We propose to observe the central stars of potential WD SN remnants to search for these stars.



While Type Ia supernovae (SNe) are the most common kind of thermonuclear white dwarf (WD) transient, we have discovered over the last decade that a significant fraction of WD transients (Perets et al., 2011; Foley et al., 2013) are relatively fast-evolving, low-luminosity, low-energy explosions. These peculiar thermonuclear transients (PTTs) likely have different progenitor systems and/or explosion mechanisms from SNe Ia. Understanding more about the outliers should also inform our understanding of the more normal objects. The discovery rate of PTTs is still relatively low, and we should utilize all possible information to further constrain their nature.

For at least two classes of PTTs, SNe Iax (Foley et al., 2013) and Ca-rich SNe (Perets 2010), we do not think that the explosion unbinds the WD. For both classes, we measure ejecta masses of \(\sim\)0.5 M\({}_{\odot}\), from which we imply a bound WD with \(M\approx 0.5\)–1 M\({}_{\odot}\). Immediately after the explosion, the outer layers of the progenitor “WD” are expected to be non-degenerate with a significant amount of injected energy from a combination of fallback and radioactive decay of bound, burned material. In fact for SNe Iax, we have significant evidence of a bound WD driving a strong, optically thick wind (Foley et al., 2014; Foley et al., 2016; Fox et al., 2015). This star will contract on a Kelvin Helmholtz timescale (\(\sim\) thousands of years or more) back to a typical WD radius, but should have very odd abundances. These are strong predictions for this model.

Earlier this year, a particular odd WD was discovered (Kepler 2016). This star has an oxygen atmosphere (there are very strong limits on the amount of H and He), but has a relatively low mass (Figure \ref{fig:kepler}). It is unclear how to form such a star without it having significant explosive burning. This star is a possible “postgenitor” star for a PTT and its discovery illuminates a new path to understanding PTTs: searching for more of these stars in the Milky Way.

We propose to firmly establish the relationship between PTTs and WD progenitor systems by directly detecting and characterizing bound WD postgenitors at the centers of SN remnants. Dozens of SN remnants (SNRs) in the Milky Way have been inferred to arise from type Ia supernovae, but if \(\sim\)15% have arisen from PTTs, we expect to find several postgenitors near their centers. We will obtain spectra of promising candidates near the centers of Ia-like remnants to identify WD remnants. This will allow us to map out the enigmatic connections between various subclasses of PTTs and their progenitor systems.