The strange story of calcium-rich supernovae

When SN 2005E exploded, it made a big splash — and not just because of what it ejected.

One of the unique characteristics of astronomy is that most of the objects that we’re interested in can’t be studied in a laboratory. Essentially, the entire observable universe is our laboratory, and all we humans can do is look around. Take, for instance, supernovae. There’s no way to create one here on Earth, fortunately, so astronomers have to survey the sky, watching and waiting for something extraordinary to happen.

Once in a while, nature hands us something spectacular. In 2005, the light from an exploding star in the lonely galaxy NGC 1032 reached Earth, and astronomers at Lick Observatory catalogued the event as SN 2005E (Perets et al. 2010). After doing a number of follow-up observations, they realized what the truly extraordinary bit of the supernova was: Much of the matter expelled by the explosion was, inexplicably, calcium. Add to that the fact that the supernova occurred in the far outer reaches of NGC 1032, and SN 2005E suddenly became a target of interest.

NGC 1032, as seen through the Schulman Telescope at the Mount Lemmon Observatory in 2011. SN 2005E is not visible, as the image was taken six years later. Image credit: Adam Block, Mount Lemmon SkyCenter, University of Arizona, under the Creative Commons Attribution-Share Alike 3.0 United States license

The supernova eventually became the prototype of a class of objects known as calcium-rich supernovae. As their name suggests, about half of their ejecta consists of calcium, rather than hydrogen or helium. They occur either far from the center of a galaxy or in intergalactic space, somewhere in the middle of a galaxy cluster and far from any individual galaxy. 14 years later, our picture of these events is slowly starting to come into focus, but the precise mechanisms behind them remain unknown.

This week’s blog post talks about the strange case of SN 2005E, further calcium-rich supernova discoveries, and the unlikely sequence of events we think are needed to produce one. Rare, dim, and mysterious, they provide a new window into what goes into enriching interstellar gas and dust.

50% calcium, 100% unique

Fig. 1, Perets et al. (a) shows an SDSS image of NGC 1032 before the supernova; (b) shows a KAIT image of it afterwards, with SN 2005E clearly visible.

Many images of supernovae show face-on galaxies, with the explosion a bright point of light sprinkled in somewhere. We can’t get photographs of SN 2005E that look anything like this, however, because NGC 1032 appears edge-on to us, presenting only a side view. In the initial Lick Observatory images — taken using the Katzman Automatic Imaging Telescope (KAIT) — the supernova appears only as a dot relatively far from the host galaxy, 22.9 kpc radially from the center and a surprising 11.3 kpc above the disk. If we didn’t have earlier images of NGC 1032 from the Sloan Digital Sky Survey (SDSS), showing only a blank space where the supernova appeared, it might be mistaken for a field star, or a separate background galaxy.

Nonetheless, follow-up observations and spectroscopy confirmed that SN 2005E was a supernova. The spectra showed no hydrogen lines, ruling out a Type II supernova, but also lacked the silicon features of a Type Ia supernova, initially leading the team to classify it as a Type Ib supernova, with a massive star that had been stripped of its hydrogen envelope as a progenitor. However, the spectra also indicated that only about 0.3 solar masses had been ejected, which seemed to imply that a massive star couldn’t have been responsible. Additionally, there was no star formation anywhere nearby.

Fig. 2, Perets et al. The strong calcium lines are clear in these spectra of SN 2005E, indicating that about 0.135 solar masses of calcium was ejected.

This presented a puzzle. The mass of the ejecta was also too low for a normal Type Ia supernova. Further study of it showed remarkably strong calcium lines, meaning that 40–50% of the ejecta, by mass, consisted of calcium. This had never been observed before, although some models of white dwarf-white dwarf systems predicted that it could happen through a detonation of donated helium on the surface of one of the components, with double-white dwarf progenitors required to explain the extreme low luminosity of SN 2005E, dim even compared to Type Ia supernovae.

Fig. S3, Perets et al. SN 2005E also showed peculiar amounts of nitrogen in its ejecta — amounts similar to those produced by Type II core collapse supernovae, but fractions similar to Type Ib supernovae. Additionally, the total mass ejected was smaller than normal for any class — closer to that expected for Type Ia supernovae, but remarkably dim.

One question remained: How did SN 2005E stray so far from NGC 1032? The most likely explanation was that the progenitor had been a hypervelocity star, originally ejected from the inner region of the galaxy after an encounter with a supermassive black hole (SMBH) or SMBH binary. The speeds needed to propel a massive star this far from the center before it exploded ranged from 300 km/s to 1600 km/s — not unreasonable. However, the rate at which such encounters happen for high-mass stars make this event unlikely to be seen by the telescope, and coupled with the observations, it remained unlikely that the progenitor was a single massive star, but instead a white dwarf binary system, similarly ejected from the galaxy’s center.

Are calcium-rich supernovae truly alone?

SN 2005E did not remain the only known calcium-rich supernova. The same team that discovered it also found a set of earlier supernovae in archival data that had many of the same properties, and other groups were able to detect new ones, as well as study prior supernovae in more detail. At this point, it became clear that all of the objects had a few properties in common: they had low-mass ejecta, a large fraction of which was calcium, and they lay far from the centers of galaxies, extending to distances of many tens of kiloparsecs.

Fig. 3, Lyman et al. SN 2007ke also occurred far from any nearby galaxy. The second and third frames are the inset seen by Hubble through different filters, showing other targets of interest.

The fact that the supernovae appeared to be taking place at large separations motivated searches for faint host galaxies. For instance, was it possible that SN 2003dr, offset from the disk of NGC 5714, was in fact located in a faint globular cluster, or even in a low surface brightness dwarf galaxy? No, said Lyman et al. 2016, as far as Hubble could see. The team found similar results for a number of calcium-rich supernovae, confirming the idea that these supernovae really do happen far from the core of any galaxy.

It was also striking that some of these supernovae were clearly occurring in the space between galaxies — not merely in their outer reaches. In fact, when studying some calcium-rich supernovae like PTF12bho and PTF11kmb, it became a challenge to determine just where they originated from (see Lunnan et al. 2017). For instance, while the latter is a staggering 150 kpc from NGC 7265, that galaxy remains the most likely to be the progenitor’s source, as other candidates have redshifts that place them farther from the galaxy cluster.

Fig. 10, Lunnan et al. PFT12bho in two different SDSS fields of view of its home, the Coma Cluster.

With a larger sample size of calcium-rich supernovae, astronomers have determined that they all have a number of factors in common, including intermediate luminosities and fast-changing light curves when compared to normal supernovae. This has made it easier to identify new candidates among previously-known supernovae, but we still don’t know of many. Hopefully, with a little bit of luck, that number will grow.