≡ Menu

White Dwarfs: Planetary System Rebirth?

Let’s catch up with white dwarfs, a kind of star that may spawn planetary systems of its own. For I’ve just found another case of archival data being put to good use in the form of a study of a white dwarf system called G238-44. Here, the data come from the Hubble instrument (specifically, its Cosmic Origins Spectrograph and Space Telescope Imaging Spectrograph), the Far Ultraviolet Spectroscopic Explorer (FUSE), and the Keck Observatory’s High Resolution Echelle Spectrometer (HIRES) in Hawaii.

What astronomers presented at a recent AAS conference is a picture of a system severely disrupted by its star’s transition to white dwarf status. Moreover, this is a star in the process of accretion with a distinct twist from earlier such discoveries. For the white dwarf – the remnant left behind after the system’s star went through its red giant phase – is actively drawing rocky and metallic material as well as ices from the debris of the disrupted system. These are the stuff of planet formation. We’re learning how extreme are conditions in what astronomers call an ‘evolved’ planetary system as it undergoes destruction and what may be a kind of rebirth.

Image: This illustrated diagram of the planetary system G238-44 traces its destruction. The tiny white dwarf star is at the center of the action. A very faint accretion disk is made up of the pieces of shattered bodies falling onto the white dwarf. The remaining asteroids and planetary bodies make up a reservoir of material surrounding the star. Larger gas giant planets may still exist in the system. Much farther out is a belt of icy bodies such as comets, which also ultimately feed the dead star. Credit: NASA, ESA, Joseph Olmsted (STScI).

In relation to main sequence stars, white dwarfs give us new assumptions about planet formation. Such a star contains half the mass of the Sun, for example, but while it’s only a bit bigger than the Earth, it sports a density of 1 x 109 kg/m3. The average white dwarf is 200,000 times as dense as the Sun, a remnant stellar core with a temperature in the range of 100,000 Kelvin. And the system it finds itself in, surviving the red giant phase of the star, is hardly a static place.

To analyze what is happening at G238-44, we have to take into account that the original red giant, perhaps much like the Sun in its earlier days, would have cast off its outer layers as nuclear burning ceased. This shedding of mass can cause asteroids and small moons to be scattered gravitationally by remaining large planets, their own orbits disrupted. Materials like these experience tidal forces that can tear them apart as they move inward toward the star. The result: A disk of gas and dust that, over time, settles onto the surface of the white dwarf and throws a distinct observational signal.

At G238-44, the white dwarf left behind is seen in the process of accreting two such objects, a process observed before in a number of white dwarf systems but never with both icy and rocky-metallic components in the mix. Now we have a case of a white dwarf evidently drawing on a planetary system that was once abundant in ices. As UCLA’s Benjamin Zuckerman, a co-author of the paper on this work, notes:

“Life as we know it requires a rocky planet covered with a variety of elements like carbon, nitrogen, and oxygen. The abundances of the elements we see on this white dwarf appear to require both a rocky and a volatile-rich parent body – the first example we’ve found among studies of hundreds of white dwarfs.”

Within about 100 million years of the white dwarf’s formation, the star will be capturing material from regions analogous to our asteroid and Kuiper Belt. The total mass involved in this study is relatively small, about that of a large asteroid. Nitrogen, oxygen, magnesium, silicon and iron have been measured in the debris disk here, and in interesting proportions. Lead researcher Ted Johnson, a colleague of Zuckerman’s at UCLA, sees a 2-1 mix of Mercury-like material – high in iron and suggestive of a metallic planetary core – mixing with the comet-like debris.

Image: This illustration shows a white dwarf star siphoning off debris from shattered objects in a planetary system. The Hubble Space Telescope detects the spectral signature of the vaporized debris that reveals a combination of rocky-metallic and icy material, the ingredients of planets. The findings help describe the violent nature of evolved planetary systems and the composition of its disintegrating bodies. Credit: NASA, ESA, Joseph Olmsted (STScI).

Terrestrial Planet in the Habitable Zone?

With this destruction derby in mind, let’s catch up with the white dwarf WD1054–226, found not so long ago to have objects – apparently of small moon or asteroid size – orbiting close to the star. Their presence is an indication, according to astronomers at University College London, that there may be a nearby planet in the star’s small habitable zone. This finding is based on data from the ESO’s 3.5m New Technology Telescope (NTT) at the La Silla Observatory in Chile. Fully 65 dips in the star’s light show the extent of the orbital material, whipping around the star in clouds every 25 hours.

A planet farther out seems the best explanation for how this arrangement stays in place, and if it is there, it would be in an orbit about 1.7 percent of the distance between the Earth and the Sun (roughly 2.5 million kilometers). That’s in the liquid water habitable zone, and the planet would be about the size of the Earth, based on these calculations.

What interesting scenarios stars like these represent. 95 percent of the stars in the galaxy will eventually become white dwarfs, with our Sun joining their ranks in four or five billion years. At WD1054-226, we’re hypothesizing the existence of a kind of planet that has yet to be confirmed around such a star. UCL’s Jay Farihi is lead author of the paper on this work:

“This is the first time astronomers have detected any kind of planetary body in the habitable zone of a white dwarf. The moon-sized structures we have observed are irregular and dusty (e.g. comet-like) rather than solid, spherical bodies. Their absolute regularity is a mystery we cannot currently explain. An exciting possibility is that these bodies are kept in such an evenly-spaced orbital pattern because of the gravitational influence of a nearby major planet. Without this influence, friction and collisions would cause the structures to disperse, losing the precise regularity that is observed. A precedent for this ‘shepherding’ is the way the gravitational pull of moons around Neptune and Saturn help to create stable ring structures orbiting these planets.”

Image: An artist’s impression of the white dwarf star WD1054–226 orbited by clouds of planetary debris and a major planet in the habitable zone. Credit Mark A. Garlick / markgarlick.com. License type Attribution (CC BY 4.0).

Tantalizing, but remember that the ‘planet’ here is unconfirmed. JWST data on the debris disk would be helpful as we learn more. White dwarf planets of terrestrial size should eventually turn up if they’re out there in any numbers. If we could find a transit, we’d be looking at a world as large as the star it orbits. The transit depth that would afford if we can find a system so aligned would make for an unforgettable light curve.

The paper is Farihi et al., “Relentless and Complex Transits from a Planetesimal Debris,” Monthly Notices of the Royal Astronomical Society Vol. 511, No. 2 (April 2022), 1647-1666 (full text).


Comments on this entry are closed.

  • Andrew Palfreyman June 21, 2022, 8:51

    A civilisation inhabiting a planet orbiting a white dwarf could in principle be billions of years older than ours, if I understand right. It also seems to be the case that a white dwarf, due to its exceptional density, would make a handier gravitational telescope for such a civilisation than would our sun. This pair of observations are probably more whimsical than useful!

    • Michael June 21, 2022, 16:59

      Perhaps they could collect the hydrogen blown off during the red giant phase to form red dwarfs in orbit. Maybe we could do this to our system later on by funneling the hydrogen onto jupiter.

  • Alex Tolley June 21, 2022, 9:33

    I think we are really pushing the limits of describing a habitable zone around a WD. Firstly, the energy emissions as the star cools from its “white” spectral class to that of a much cooler one means that this zone will collapse to an ever-smaller radius. Can a planet even stay in that HZ for long enough? Secondly, the star is going to be emitting peak energy in the UV, and also have strong X-ray emissions. This is hardly suitable for habitability – all organic compounds will be rapidly destroyed. Any planet that is in the right orbit to be in the HZ when the WD cools towards a more FKG spectrum will first have been fried. It would need infalling ices to restore the water for surface habitable conditions. It would certainly be tidally locked. How long would it remain in the HZ for life, even externally delivere by panspermia [natural or directed], to survive and evolve? What would living on such a world be like for [post] humans on a Starbase outpost be like? Could such a world be “terraformed” despite the conditions? Should it?

    • Adam Crowl June 21, 2022, 16:31

      Hi Alex
      Most of your questions have been considered. Eric Agol started the current search for habitable zone planets around white dwarfs with his paper:
      Eric Agol. Transit Surveys for Earths in the Habitable Zones of White Dwarfs. The Astrophysical Journal, 2011; 731 (2): L31 DOI: 10.1088/2041-8205/731/2/L31

      Of course there’s plenty of unknowns, but that’s why we keep looking.


      • Alex Tolley June 22, 2022, 11:34

        Adam, thanks for the Agol link. However, my questions are less answered than avoided. Agol specifically is looking at cool white dwarfs, and therefore avoiding the issue of their history after the red giant phase has ended. As for planets, he specifically notes that the issue of formation is outside the scope of the paper, which is focused on detection.

        As for characterization of such planets he says:

        Several topics require further study. The global climate models
        for determining theWDHZ should include fast synchronous
        rotation, magnetic fields, varied planet atmospheric composition,
        radiogenic heating, and white dwarf cooling. The WDHZ
        is a necessary but insufficient criteria for habitability. For example,
        planets that start hot may not retain their atmospheres,
        as has also been argued for planets orbiting M dwarfs (Lissauer
        2007); this may require volatile delivery from more distant bodies
        in the system (Jura & Xu 2010) or planetary outgassing.
        To retain an atmosphere might require a larger planet escape
        velocity, possibly favoring super-Earths for habitability.

        The Farihi et al paper, which looks at WD 1054–226 , a cool ~7900K WD, and a disk of dust and debris. The analysis suggests that the 25.5 hr transits are due to a dust cloud and not a planet.
        One suggestion is a possible eccentric orbit where the periastron is inside the Roche limit, breaking up any planet.

        Unlike stars in the main sequence where their luminosity starts high, then declines before a long period of slow inclease, a WD starys with an exteremely high flux and peak short wavelength, and thereafter cools. What this means that a planet in a fixed orbit around a main sequence star cannot enter the HZ from the outer edge as the star increases luminosity. The planet will always start from a cooler temperature (after the initial high luminosity phase) and become hotter as the inner edge of the HZ migrates towards teh planet.
        However, for a WD planet, this does not happen. Any formation from debris due to the red giant stage will have to start inside the inner edge of the HZ before cooling and a long stable period keep the planet in the HZ. Such a planet would be melted and crisped.

        [Chesley Bonestell did a painting of a planet exposed to a star going nova, and I suspect that a planet formed close to a WD duuring its early phase of exteremely high surface temperature and radiation flux might have a similar effect.]

        For any planet in the HZ of a cool WD, either it will have formed much later, or will have migrated inwards during to orbital perturbations caused by the start in the red giant phase. IDK the orbital implications, but I would guess that as teh WD mass is much less than of the star during its main sequence phase, it is more likely to lose its retinue of planets as their orbital velocity now exceeds that needed for the WD’s reduced gravitational attraction. This argument is perhaps countered by the observation that most WDs have polluted surfaces due to infalling debris.

        Is it possible that the close debris disk is formed by the breakup of a hot Jupiter or Neptune during the red giant phase, leaving just the less volatile rocky/metal core as a debris disk that reforms slowly as the WD cools. The planet may be in the WD HZ, but would not be actually a suitable environment for life until infalling cometary material replaced the needed volatiles during the WD’s cool phase. Any planetary analysis by surface probes would likely show the mantle and crust had once been subjected to extreme heating and transformed. [Would there be abundant crystals of carbon and other refractory compounds?] Would any carbonate rocks have likely formed from the infalling CO/CO2 from the comets, and needed to maintain the geological carbon cycle for stable temperatures as the star continues to cool? Paradoxically, unlike Earth, the atmospheric GHG pressures would have to increase with time to compensate for the star’s cooling. How would this occur given what we know about the Snowball Earth period and its end?

        In summary, while planets may be found in the WD HZ, I am skeptical that they could be life-bearing.

    • Martin June 21, 2022, 19:34

      Just eyeballing white dwarf cooling curves with crystallization it looks like there is a period of several billion years in which the luminosity is fairly stable. This occurs at around L~0.0001 suns so it’s output wouldn’t be particularly fierce at this point. Of course the planet would have already have been toasted, but hey, bring on the comets!

  • Martin June 21, 2022, 19:21

    A while back, just for fun, I calculated the gravitational lens focal length for a number of close white dwarfs. I used Wikipedia masses and diameters, so take these as approximations:
    Distances in Megamiles
    Procyon b 12.9
    Sirius b 3.7
    40 eridani c 19.9
    Van Maanen’s Star 9.1
    The sun (for comparison) 50874
    Those aliens have all the luck…

    • Robin Datta June 22, 2022, 1:07

      The aliens may indeed have ALL the luck: from crisps on cinder world to merfolk in ocean worlds pre- or post- cinder phase (or both).

      One critical transition for us was the incorporation of a foreign oxygen-using microbe, (which became a mitochondrion), into an anarobic (non-oxygen-using) cell. The amount of energy made available through the use of oxygen is much greater than that from methane or sulfur compounds, each of which is found in a few organisms.

      The other critical transition was hosts of features that first changed a shrew-like proto-primate into a primate, then into a hominin, and thence into a human.

      All of this quite iffy even with robust panspermia, both pre-white dwarf and post-white dwarf. Of course with sufficiently enormous numbers of instances of such habitable planets, there may be something similar to ourselves – by the luck of the draw, even without speculation such as Robert Forward’s Dragon’s Egg or Fred Hoyle’s The Black Cloud. And then again there is the gap between existence and discovery. But looking for prospective future planetary abodes would be prudent.

  • Henry Cordova June 21, 2022, 23:08

    It’s starting to look like all stars can have planets, regardless of their population, age, size, mass, evolution, or whether or not they are binaries, cluster members or collapsed stars. Perhaps there are no planetless stars, only planets we can’t detect, or that we haven’t detected yet.

    Maybe the question we should be asking is, “which stars DON”T have planets?” And why?

    • Michael Fidler June 22, 2022, 9:41

      Maybe, we should look for the planet less stars, that where the most advanced Zen civilizations live.

      • Henry Cordova June 23, 2022, 8:39

        You might just have a point. Advanced civilizations don’t need planets, they could survive by colonizing the small bodies orbiting a sun, using asteroid and cometary material and stellar energy to run their activities. Of course, this would imply they were exiles from a planet orbiting another star, a planet they left to form a colony elsewhere, or because their home world was polluted, invaded, overcrowded or otherwise made untenable. Perhaps they are just rebels and misfits or restless souls, Spirits Rebellious. “Zen civilizations”; what a concept.

        In any case, a planetless star would be less likely to be considered as worth visiting by strangers. It might be a good place to hide, or retire, or escape to. These guys might not want to be found.

        • Jeff Wright June 24, 2022, 3:45

          End of life worldships might finish up here.

  • Geoffrey Hillend June 22, 2022, 0:01

    Alex Tolley is correct. White dwarfs emit UV black body radiation which is strong enough to strip the electrons of rocks on planets. The surface temperature of a white dwarf is 35,000 Kelvin which is in the extreme ultra violet spectrum at 82 nanometers. According to Kirchhoff’s law of black body radiation wavelength tells the temperature.

    • Adam Crowl June 22, 2022, 5:54

      Hi Geoffrey
      Those really high temperatures only last a few million years. White dwarfs cool to under 10,000 K within ~1 Gyr typically. Then they’re more sedate with radiating away their remaining entropy.

      • Alex Tolley June 22, 2022, 16:59

        So jumping into a bath of boiling water is fine..because it will cool to room temperature in a few hours? Will your body heal with the cooling?

        A planet exposed to very high radiation fluxes will be subject to a number of irreversible processes. Will the effects be similar to the heating from the bombardment of Earth at formation, or will it be very different? If the “young” WD removes most of the volatiles in the debris disk, will there be sufficient volatiles to replace the losses sufficiently to allow seas and oceans to form on an Earth-sized world once the WD has cooled and the planet entered the HZ that is shrinking around the star?

        If we start from the assumption that we ignore the history prior to the cooler phase, does that make any sense? In my boiling water bath analogy, it is like assuming that a room-temperature bath is fine, and so there is no point even thinking about the blistered and peeling skin on the person because the boiling water event is ignored.

        If the planet forms during the cool phase, then I have no argument about the HZ being relevant to possible life on the planet but is that the case? Are there any remote observations that we can make to determine when the planet formed, and if during its hot phase, what would happen to the planet in an orbit that will eventually put it in the HZ when the WD has cooled?

        • Geoffrey Hillend June 23, 2022, 17:14

          Yes everything will be sterilized. Since a white dwarf is close to the mass of the Sun, and Earth sized planet in the goldilocks zone would have strong tidal forces and would be tidally locked. It might also have an atmosphere considering the tidal forces causing volcanism. 8,000 K to 10,000 K is still in the ultra violet spectrum. I can’t imagine any life without light in the visible spectrum and photosynthesis adapting to that kind of environment or after the searing heat of the red giant phase. Without oxygen, they can be no ozone or protection from UV.

        • Adam Crowl June 25, 2022, 17:27

          Hi Alex
          The infall of debris suggests there’s a flux of fresher material over time. Earth itself lost its primary atmosphere during the Sun’s own exuberant phase. What we have today is the long reprocessed remnants of the secondary atmosphere exuded from the mantle in the first billion years. Our cosmogonies are still descriptive attempts at imagining how a planetary system has formed after the fact, rather than predictive about the ultimate fate of potentially habitable planets.

  • Geoffrey Hillend June 22, 2022, 0:04

    White dwarfs emit extremely ionizing radiation.

  • Alastair Williams June 22, 2022, 3:59

    I suppose there are two possibilities for life here:
    1) A civilization that somehow survives the death of its star and is clinging on around a white dwarf (that’s a rather depressing thought).
    or 2) life or civilization that evolves in the billions of years a planet lingers in the habitable zone of the white dwarf.

    As another thought, has any consideration been made as to the fate of moons around gas giants? Say Europa or Titan. Such places might be habitable now, would they remain so around a white dwarf?

  • ljk June 22, 2022, 10:34

    There is also the possibility that any residents in a white dwarf star system are not original inhabitants, but visitors from other systems who came to mine, explore, etc.

    Such systems could be especially attractive to ETI since it is assumed any original inhabitants were either wiped out when the star changed or left for safer pastures before their sun turned into a red giant.

  • wdk June 23, 2022, 21:21

    Here’s an MNRAAS paper from about a year ago.

    PDF available.
    It looks at Hyades cluster WDs, age about 625 million years, assuming initial main sequence mass of about 3 Solar. No jovian mass stars detected as self luninous or otherwise. There are differences and similarities with what we are examining here. The main advantage of a new star cluster would be in the intrinsic brightness of newly formed
    jovian planets.