Stars that exploded just a little bit

Title: A class of partly burnt runaway stellar remnants from peculiar thermonuclear supernovae

Authors: R. Raddi, M. A. Hollands, D. Koester, J. J. Hermes, B. T. Gaensicke, U. Heber, K. J. Shen, D. M. Townsley, A. F. Pala, J. S. Reding, O. F. Toloza, I. Pelisoli, S. Geier, N. P. Gentile Fusillo, U. Munari, J. Strader

First Author’s Institution: Friedrich-Alexander-Universität Erlangen-Nürnberg, Germany

Status: Submitted to MNRAS [open access on arXiv]

In today’s bite, we are going to talk about some pretty weird stars.

A couple of years ago, a white dwarf named LP40-365 was identified as a high-velocity star with a very unusual chemical composition. It was proposed that the star is a partially burnt white dwarf that had survived a (subluminous) type Ia supernova. Basically, the white dwarf went supernova but was not completely destroyed in the explosion, resulting in a fainter supernova that left a remnant behind.

This, put mildly, is super exciting stuff! We still don’t know whether type Ia supernovae result from the thermonuclear explosion of a white dwarf accreting material from a main-sequence companion, or from the merger of two white dwarfs. Observing a surviving remnant can reveal a lot about how thermonuclear supernovae actually work.

Today’s paper reports the discovery of three more stars that are similar to LP40-365, all supernova survivors. The authors selected high-velocity stars by mining Gaia DR2 and took their spectra. By analyzing stellar spectra, one can identify various elements present in the star’s atmosphere.

Very metal

The authors found three stars that, along with LP40-365, form a distinct class of chemically peculiar runaway stars. The spectra for the four stars look very similar to each other and quite different from other classes of stars. Their atmospheres are dominated by neon, followed by oxygen and then magnesium. There is no hydrogen or helium. They are also enriched with similar amounts of elements made in oxygen and silicon burning during thermonuclear explosions.

Fig 1 shows the atmospheric composition of three stars, as well as the bulk composition predicted by theoretical models of exploding carbon-oxygen and oxygen-neon white dwarfs. The theoretical predictions look very different from the observed composition but it is important to remember that we’re just looking at the atmospheres of these stars. The predicted bulk composition need not match the atmospheric composition since internal mixing can move elements around and change what we see.

Fig 1. The histogram bars represent the mass fraction of detected elements. The first three bars show the atmospheric composition of three partly burnt stars, and the rest show theoretical calculations. The Fink+14 models are pure deflagrations of carbon-oxygen white dwarfs. The Jones+18 model is an oxygen-neon white dwarf. Figure 13 in the paper.

In Fig 2, we see the elemental abundance patterns for the same three stars. The values are given relative to the ratios in the sun. The main thing to note here is how similar the three patterns look! By looking at such patterns, it is sometimes possible to make a guess about the conditions under which the various elements were made. Here, for example, we can see the Cr-, Co- and Ni-to-Fe ratios are higher than solar. The high central densities conducive to producing these high ratios are found in near-Chandrasekhar mass white dwarf explosions. This in turn tells us that the explosion was probably the result of the white dwarf accreting mass from a companion, giving us another clue about thermonuclear supernovae.

Fig 2. Black circles, blue squares, and red diamonds represent LP40−365, J1603−6613, and J1825−3757, respectively. The x-axis gives the name of the element and the y-axis gives its ratio to iron, relative to the corresponding solar ratio. Fig 11 in the paper. 

Light, large and moving fast

Thanks to Gaia parallaxes, the masses and radii of these stars can be estimated pretty well. It turns out that their masses are much lower than that of a typical white dwarf. The stars also seem to be inflated, i.e. have large radii, probably as a consequence of the supernova explosion! Their high velocities also make sense if they were flung out of a binary system that was disrupted in the supernova.

Fig 3 shows the location of the stars on a theoretical Hertzsprung-Russell (HR) diagram. Other classes of stars are shown for comparison. The portion of the HR diagram that they occupy is not occupied by any long-lived phase of stellar evolution.

Fig 3. Luminosity versus effective temperature for LP 40−365 (black circle), J1603−6613 (light-blue square), and J1825−3757 (red diamond). The white circles are white dwarfs, orange ones are main-sequence stars, and the blue circles are hot subdwarfs. Fig 14 in the paper.


Partly burned runaway stellar remnants form a distinct class of stars. At present, the class contains four members with the following characteristic properties:
1. Ne-dominated atmospheres, followed by O and Mg.
2. Broadly homogeneous composition of trace elements (C, Na, Al, Si, S, Ca, iron group, Sr).
3. Low mass, below the canonical white dwarf mass.
4. High ejection velocity in the range of 550-600 km/s.

Let’s wrap up with some really good news: the authors estimate that around 20 such stars should be detectable within 2 kpc from the Sun at the end of the Gaia mission. This story is just beginning.

About Sanjana Curtis

I'm a grad student at North Carolina State University. I'm interested in extreme astrophysical events like core-collapse supernovae and compact object mergers.

Leave a Reply