Searching for Type Ia Supernovae Progenitors through Circumstellar Material

Fig. 1: Artist impression of the two widely accepted type Ia progenitors.
(Top) A white dwarf accreting material from a non-degenerate companion. Copyright: David A. Hardy/AstroArt.org.
(Bottom) Two white dwarf stars spiraling in towards a merger due to gravitational wave emission. Copyright: GSFC/D. Berry.

Fig. 2: Multi-epoch spectra of SN 2006X showing the time-variable behavior interpreted as a signature of circumstellar material close to the exploding white dwarf. Image taken from Patat et al. 2007 (doi: 10.1126/science.1143005). 

Fig. 3: Multi-epoch spectra of SN 2008dt. The slight variability that is observed is within the noise level. These observations are consistent with a non-detection of time-variable features. Hence, there no sign of CSM along the line-of-sight to this event. Adapted from Sternberg et al. 2013. 

Signatures of the circumstellar material in the spectra of type Ia supernova might shed light on the progenitors of these cosmic explosions. Scientists at MPA have now analysed multi-epoch spectra of about a dozen supernova explosion of Type Ia and come to the preliminary conclusion that only a minority show clear signatures of such material that is consistent with a single-degenerate progenitor system.  

Type Ia supernovae (SNe Ia) are very luminous explosions that are used as standerdizable candles to measure distances on a cosmic scale. These measurements can then be used to reconstruct the expansion history of the Universe. Knowing the full nature of the progenitors of these events might help to standardize them more accurately, allowing for a more accurate reconstruction result. Moreover, SNe Ia are thought to be the end product of binary systems. Understanding their progenitors will help us to better understand the evolution and end product of certain binary stars. Hence, the nature of SN Ia progenitors is an important open question in need of an answer. 

It is widely accepted that a SN Ia event is the explosion of a carbon-oxygen white dwarf star. For a white dwarf to explode it needs to accrete material that will trigger carbon-burning. Due to the degenerate state of the white dwarf material this burning is a runaway process that produces enough energy to unbind and totally disrupt the white dwarf. The two leading models for the progenitors of these events are the single-degenerate model (Fig. 1, top) – in which material from a non-degenerate companion is transferred onto the white dwarf – and the double-degenerate model (Fig. 1, bottom) – in which a degenerate companion, another white dwarf, merges with the primary. 

One of the major discriminants between the different progenitor scenarios is the environment in which the white dwarf explodes. In the single-degenerate scenario the white dwarf if engulfed by circumstellar material (CSM) that was expelled from the system due to mass loss processes. This material should have relatively low outflowing velocities. In the double-degenerate scenario the white dwarf explodes in a cleaner environment. Even though some recent work suggests CSM might be present also in a certain double-degenerate progenitor, this would have different properties, namely higher velocities. 

Therefore, the detection of CSM in type Ia spectra can help disentangle the different progenitor scenarios and allow us to determine the binary pathway, or pathways, that lead to them. The question then arises what the manifestation of this CSM should be. Material that is close to the exploding white dwarf should be ionized by the ultra-violet radiation emitted during the explosion. As time progresses this material should recombine and return to its previous neutral/ionization-level. Thus, in early-time spectra, not long after the explosion, we expect to see little or no features of the neutral/low-ionization-level. In later-time spectra we expect to see these features emerge and/or intensify. Material that is further away at the time of explosion will not be ionized, but given its relatively low outflowing velocity it should manifest itself as a blue-shifted absorption feature. An ideal element to use in this search is neutral sodium, as it is a strong line, even when only small amounts of sodium are present. 

The first widely accepted detection of CSM in a type Ia was reported for SN 2006X by a group led by Ferdinando Patat from ESO (Fig 2). Following this detection two more events were reported to show signs of CSM – SN 2007le and PTF11kx – and three events for which such material was not detected – SN 2000cx, SN 2007af, and SN 20011fe. These mixed result might be due to viewing-angle effects that will cause the CSM to be visible only in part of the SNe Ia, *if there is* (?) a mix of at least two classes of progenitors - one with CSM and one without xxx– or more likely a combination of both xxx. (Both? Viewing angle and mixing? Delete or explain) The small size of this sample does not allow any robust conclusions to be made. A larger sample is needed to robustly conclude what the prevalence of cases like SN 2006X, SN 2007le, and PTF11kx is. Moreover, a larger sample including more cases with CSM detection will allow the study of the CSM properties. Non-detection of CSM can be used to estimate upper limits to the CSM mass. With these we can exclude implausible models and set constraints to the plausible ones. 

A group led by Assaf Sternberg showed that SNe Ia exhibit an overabundance of features indicative of outflowing material. This overabundance was shown to be consistent with circumstellar material. Nevertheless, as this analysis was based on single-epoch data, it can not be used to probe the properties of the CSM, as it is not possible to tell which individual features are circumstellar and which are interstellar. 

In collaboration with scientists world-wide we are leading a renewed effort to obtain a large multi-epoch spectroscopic sample of SNe Ia in hope to shed light on the elusive progenitors of SNe Ia. So far, we have already obtained multi-epoch spectra of 13 SNe Ia (Fig. 3), more then tripling the currently published sample. This enlarged sample suggests that only ~17% of SNe Ia exhibit time-variable absorption features that are associated with CSM. Though this result is in agreement with other previously published work, due to the size of our sample this result may still change. Moreover, in future analysis we will estimate upper limits for the CSM mass and will try to determine which binary pathways may be excluded as progenitors for the events in our sample. This is still work in progress. We hope to reach a sample size that is comparable with the Sternberg et al, single-epoch sample within the next couple of years, and that its analysis will help answer the long-standing type Ia progenitor question.

Assaf Sternberg and Wolfgang Hillebrandt

References

Patat, F., Chandra, P., Chevalier, R., et al. 2007 Science, 317, 924

Simon, J. D., Gal-Yam, A., Gnat, O., et al. 2009, ApJ, 702, 1157

Dilday, B., Howell, D. A., Cenko, S. B., et al. 2012, Science, 337, 942

Sternberg, A., Gal-Yam, A., Simon, J. D., et al. 2011, Science, 333, 856

Sternberg, A., Patat, F., Hillebrandt, W., et al, 2013, in preperation