≡ Menu

White Dwarf Clues to Unusual Planetary Composition

The surge of interest in white dwarfs continues. We’ve known for some time that these remnants of stars like the Sun, having been through the red giant phase and finally collapsing into a core about the size of the Earth, can reveal a great deal about objects that have fallen into them. That would be rocky material from planetary objects that once orbited the star, just as the planets of our Solar System orbit the Sun in our halcyon, pre-red-giant era.

The study of atmospheric pollution in white dwarfs rests on the fact that white dwarfs that have cooled below 25,000 K have atmospheres of pure hydrogen or helium. Heavier elements sink rapidly to the stellar core at these temperatures, so the only source of elements higher than helium — metals in astronomy parlance — is through accretion of orbiting materials that cross the Roche limit and fall into the atmosphere.

These contaminants of stellar atmospheres are now the subject of a new investigation led by astronomer Siyi Xu (NSF NOIRLab), partnering with Keith Putirka (California State University, Fresno). Putirka is a geologist, and thus a good fit for this study. Working with Xu, an astronomer, he examined 23 white dwarfs whose atmospheres are found to be polluted by such materials. The duo took advantage of existing measurements of calcium, silicon, magnesium and iron from the Keck Observatory’s HIRES instrument (High-Resolution Echelle Spectrometer) in Hawai‘i, along with data from the Hubble Space Telescope, whose Cosmic Origins Spectrograph came into play.

Their focus is on the abundance of elements that make up the major part of rock on an Earth-like planet, especially silicon, which would imply the composition of rocks that would have existed on white dwarf planets before their disintegration and accretion. The variety of rock types that emerge is wider than found in the rocky planets of our inner Solar System. Some of them are unusual enough that the authors create new terms to describe them. Thus “quartz pyroxenites” and “periclase dunites.” None have analogs in our own system.

The finding has implications for planetary development, as Putirka explains:

“Some of the rock types that we see from the white dwarf data would dissolve more water than rocks on Earth and might impact how oceans are developed. Some rock types might melt at much lower temperatures and produce thicker crust than Earth rocks, and some rock types might be weaker, which might facilitate the development of plate tectonics.”

The paper goes into greater detail:

…while PWDs [polluted white dwarfs] might record single planets that have been destroyed and assimilated piecemeal, the pollution sources might also represent former asteroid belts, in which case the individual objects of these belts would necessarily be more mineralogically extreme. If current petrologic models may be extrapolated, though, PWDs with quartz-rich mantles…might create thicker crusts, while the periclase-saturated mantles could plausibly yield, on a wet planet like Earth, crusts made of serpentinite, which may greatly affect the kinds of life that might evolve on the resulting soils. These mineralogical contrasts should also control plate tectonics, although the requisite experiments on rock strength have yet to be carried out.

Image: Rocky debris, the pieces of a former rocky planet that has broken up, spiral inward toward a white dwarf in this illustration. Studying the atmospheres of white dwarfs that have been polluted by such debris, a NOIRLab astronomer and a geologist have identified exotic rock types that do not exist in our Solar System. The results suggest that nearby rocky exoplanets must be even stranger and more diverse than previously thought. Credit: NOIRLab/NSF/AURA/J. da Silva.

High levels of magnesium and low levels of silicon are found in the sample white dwarfs, suggesting to the authors that the source debris came from a planetary interior, the mantle rather than the crust. That contradicts some earlier papers reporting signs of crustal rocks as the original polluters, but Xu and Patirka believe that such rock occurs as no more than a small fraction of core and mantle components.

Adds Putirka:

“We believe that if crustal rock exists, we are unable to see it, probably because it occurs in too small a fraction compared to the mass of other planetary components, like the core and mantle, to be measured.”

The paper is Putirka & Xu, “Polluted white dwarfs reveal exotic mantle rock types on exoplanets in our solar neighborhood,” Nature Communications 12, 6168 (2 November 2021). Full text.


{ 7 comments… add one }
  • Bruce D. Mayfield November 3, 2021, 15:01

    Surely exoplanets do indeed contain far more combinations of mineral types than we have come across here on and in our own planet and it’s solar system.

    “The universe is not only stranger than we imagine, it is stranger than we can imagine.”

  • Alex Tolley November 4, 2021, 11:23

    I would like to read some critiques (if any) on this study beyond the 3 peer reviews.

    The WD stars selected are all with high Calcium spectra. The argument for the rock types is based on the normalized abundances of Mg, Fe, Si, Ca in the star’s spectra that are assumed to be the due to the consumption of rocky planets during the red giant phase – hence “polluted”.

    Looking at the data, I note that our sun has the same normalized abundances as the various rocky planets and asteroids in our system, even though we know that these planets have not polluted our sun. )I can only assume that the total abundances in WD are much higher, but that data is not provided anywhere).

    If one looks at the details in Table 1 it is not clear to me how the main rock types were extracted from the data (The supplementary information has the matrices of abundances but it needs a domain expert to understand how these relate to different rocks.)
    For example, WD1929 + 011 is classed as polluted with olivine (Dunite). Whereas WD1536 + 520 is classed as Periclase Dunite because there is more Mg, or less Fe in the spectrum? None of the diagrams explain the specifics, although the supplementary data suggests this is the case.

    I have no doubt that this paper is more geology centric and to astronomers with the relevant expertise, but I find the paper as presented more glitzy and less informative than I would prefer as a non-expert.

    • Alex Tolley November 4, 2021, 22:09

      Looking at the data, I note that our sun has the same normalized abundances as the various rocky planets and asteroids in our system, even though we know that these planets have not polluted our sun. )I can only assume that the total abundances in WD are much higher, but that data is not provided anywhere).

      I made an irrelevant statement here. I now understand that any metals would have sunk below the surface quite quickly, therefore ending their absorption lines in the star’s spectra. This should leave primarily H/He lines. Therefore any metal absorption lines must be from elsewhere, assumed to be infalling material from the nearer planets and debris.

      But another thought occurs. Now that we know that extrasolar systems do not necessarily have the same planetary arrangement as Sol’s, some of that debris may be from hot Jupiters and Neptunes in close orbits. For such worlds where the rain may be various metal oxides, how would that impact the spectra and the assumption of different mantle material of rocky worlds based on a mix of material sources?

  • Michael November 4, 2021, 17:21

    OT but it would not be beyond the realms of an advanced society building rings around these white dwarfs for energy which although small are very high in energy content. One of the smallest WD’s is around the size of the moon ! Encase it and the energy would be available for Trillions of years. Quite astonishing is the magnetic fields as well, one I think was clocked at 100 tesla. A globular cluster of these WD’s would be very valuable real estate indeed,.

  • Will Doyle November 5, 2021, 16:41

    I wonder what percentage of PWDs still have planets orbiting beyond their Roche limits. Does the pollution of the photosphere make it harder to detect orbiting planets using current methods?

    • Paul Gilster November 5, 2021, 19:54

      That question about polluted atmospheres is a good one (thanks, Will!), and one I hadn’t considered in relation to radial velocity detections. Does anyone have any leads on this in the literature?

      • Will Doyle November 5, 2021, 22:57

        After doing a bit of homework, it seems like a few people have looked at this. Undergrads at University of Washington searched for transiting debris (Wallach et al.), and a couple of UT Austin researchers predict that giant planets may be easier to detect around PWDs due to the presence of more spectral lines in the stars due to the introduced metals (Endl and Williams). Not at all what I had first thought.


Leave a Comment