≡ Menu

A Planet Engulfed by a Red Giant?

Polish astronomer Aleksander Wolszczan (Penn State) is best known as the discoverer of the first confirmed planet outside our Solar System. That was back in the early 1990s, when Wolszczan was working with Dale Frail (NRAO), using observations from the Arecibo dish to demonstrate that the pulsar PSR B1257+12 was orbited by two planets. These are relatively small worlds (3.9 and 4.3 Earth masses respectively), and in an era where new planet candidates number in the thousands, it’s easy to forget how striking Wolszczan’s work appeared at the time, and how it gave impetus to the developing exoplanet hunt.

A pulsar planet looks to be an extremely inhospitable place, but learning how planets are distributed among the stars involves studying every conceivable kind of world. Wolszczan’s latest work targets an equally hostile environment, the former habitable zone of a star that has begun expanding into a red giant. The star, BD+48 740, has 11 times the Sun’s radius and is significantly older. Wolszczan and an international team of astronomers have been studying its chemical composition by way of figuring out events in its recent past. Using data gathered at the Hobby-Eberly Telescope (UT-Austin), they discovered that BD+48 740 contains an unusual amount of lithium, an element created primarily in the early days of the universe.

Finding lithium in such high amounts is a flag to the astronomer:

“Theorists have identified only a few, very specific circumstances, other than the Big Bang, under which lithium can be created in stars,” Wolszczan added. “In the case of BD+48 740, it is probable that the lithium production was triggered by a mass the size of a planet that spiraled into the star and heated it up while the star was digesting it.”

Image: The first evidence of a planet’s destruction by its aging star indicates that the missing planet was devoured as the star began expanding into a “red giant” — the stellar equivalent of advanced age. “A similar fate may await the inner planets in our solar system, when the Sun becomes a red giant and expands all the way out to Earth’s orbit some five-billion years from now,” said Alexander Wolszczan, Evan Pugh Professor of Astronomy and Astrophysics at Penn State and the discoverer of the first planet ever found outside our solar system. Credit: Marty Harris/McDonald Obs./UT-Austin.

The work on BD+48 740 is part of the Penn State-Toruń Planet Search, which specializes in detecting planetary systems around stars more evolved than the Sun. Some 50 red giants are currently known to host planets or brown dwarf companions, some of them (like HD 102272) in multi-planet systems. Although earlier phase sub-giants are found with ‘hot Jupiters’ orbiting them, the more evolved red giants show few planets in close orbits and few planets whose orbit is eccentric.

That makes the discovery of a long-period planet in a highly eccentric orbit around a red giant an interesting find. BD+48 740 is orbited by a planet of at least 1.6 Jupiter’s mass in such an orbit, one the researchers call the most elliptical planetary orbit yet detected around an evolved star. The culprit should be gravitational interactions between planets, and the suspicion is that the missing planet was the cause, its dive into the star giving the outer planet a gravitational boost.

The elliptical orbit of the newly discovered planet is not in itself enough to imply the engulfment of an inner planet. The paper on this work points out that we have studied planetary systems like HAT P-13 b, c and HD 217107 b, c, both systems around main sequence stars in which there is a close-in planet in a nearly circular orbit and a more distant companion in an eccentric orbit. But the lithium abundance in BD+48 740 suggests a missing planet and a recent planet-planet scattering event, one that would have occurred after the star had left the main sequence.

Says team member Eva Villaver (Universidad Autonoma de Madrid):

“Catching a planet in the act of being devoured by a star is an almost improbable feat to accomplish because of the comparative swiftness of the process, but the occurrence of such a collision can be deduced from the way it affects the stellar chemistry. The highly elongated orbit of the massive planet we discovered around this lithium-polluted red-giant star is exactly the kind of evidence that would point to the star’s recent destruction of its now-missing planet.”

The paper is M. Adamów at al., “BD+48 740 – Li overabundant giant star with a planet. A case of recent engulfment?” Accepted at Astrophysical Journal Letters (preprint). A Penn State news release on this work is also available.


Comments on this entry are closed.

  • A. A. Jackson August 23, 2012, 12:09

    I am a little surprised that this seems to be the first detection of a planet in the extended envelope of a main sequence star. I thought there were other instances.
    It’s one of those things, but you have to be careful where you publish. In 1981 Dr. Rosemary Killen (now at Goddard Spaceflight Center) and I published a paper about the consequences of a ‘Jupiter’ becoming embedded in a Red Giant atmosphere. (I got the idea by thinking about the fate of Jupiter when the Sun goes off the Main Sequence.) :
    “Accretion shocks as evidence of extrasolar planets”, Jackson, A. A.; Killen, R.,British Interplanetary Society, Journal (Interstellar Studies), vol. 34, Nov. 1981, p. 494-497, 11/1981

    Tried to get this paper published in Nature and then Icarus, where it was rejected. I got flustered and asked Tony Martin if it fit the JBIS, he said sure.

    In 1982 Mario Livio published “On the origin of low mass cataclysmic binaries” in Astronomy and Astrophysics, vol. 112, no. 2, Aug. 1982, p. 190-194. Which has a discussion planets embedded in stellar giants atmospheres.
    Now cataclysmic binaries had been known for some time , where one star in a binary gets swallowed by its partner.
    (There are earlier qualitative mentions of it.)
    I don’t know if Killen and I were the first to propose the planet concept with a quantitative treatment.

    Livio went to study in a number of papers both stellar and planet embeddings. Since he did not cite us I wrote him, he wrote me back a nice letter , he did not chide me but did say we had published in an obscure journal! To astrophysicists , I think that’s probably true. However, I notice in later publications (as far as I know) he never cited our paper.
    If you are interested in interstellar flight JBIS is your journal … otherwise , alas!
    There is quite a list of papers (1990-2000’s) about planets embedded in Red Giant envelopes , I not sure a one of them cites our paper.
    Griper mode now turns off.

  • Mike August 23, 2012, 15:36

    Burp! Stellar indigestion?

  • qraal August 23, 2012, 16:34

    Hi Paul
    When our own Sun balloons outwards during its Red Giant phase current modelling suggests Earth will escape direct engulfment – at the same time as the radius expands, the Sun is losing mass and thus the Earth’s orbit will expand. However that simplistic analysis is dramatically modified when tidal forces are factored in – Earth’s gravity will create a tide in the “hot vacuum” of the Sun’s envelope and undergo a death-spiral.
    Normally that would be the end, but opacity might produce a different end result. The Sun’s mass-loss rate is driven by opacity – the resistance of the Solar envelope to outwards heat-flow. If it can be engineered just right, the Sun will balloon, but lose too much mass to turn into a Helium Main Sequence star. Instead it will subside directly into being a helium white dwarf. The recent discussion here of the prospects for habitable zones around white dwarfs suggests that the future guardians of the Earth might cause it to spiral inwards just slowly enough to match the Sun’s contraction.
    Of course how they might do so and how they would protect Earth during the Red Giant phase are questions that’ll need answering over the next 7.5 aeons before the Sun reaches the extreme end of its evolution as a Red Giant.

  • coolstar August 23, 2012, 19:12

    @Dr. Al Jackson It’s really too bad that Livio didn’t reference your earlier work; I know how annoying that can be as it’s happened to me also (but with much less of an excuse than in your case).
    I don’t know how well it’s remembered, but Aleksander Wolszczan’s and Dale Frail’s announcement of the “first” pulsar planets made for quite an event at the 1992 AAS meeting where Wolszcan gave an invited talk on the subject. He was proceeded by Andrew Lyne who gave a talk RETRACTING the discovery of a planet(s) around PSR 1829-10, which would have been the first confirmed extra-solar planet. That detection had a period of very near 365 days, as I remember, and was an artifact of their data reduction process. The large audience was so impressed by Dr. Lyne’s integrity and courage that he received a standing ovation, the only one I’ve ever seen at an AAS meeting.

  • A. A. Jackson August 23, 2012, 19:51

    That is very good summary , there has been a long study , well since 1980, of the fate of the Earth as the Sun moves off the Main Sequence. It was now you see it , now you don’t for a long time.
    The recent paper I saw was :
    “Distant future of the Sun and Earth revisited”, Schröder, K.-P.; Connon Smith, Robert, Monthly Notices of the Royal Astronomical Society, Volume 386, Issue 1, pp. 155-163, 2008.
    I don’t know what’s more recent, except that I do know with the confirmation of lots of extra solar planets now this is also a topic for other planetary systems, like what should we look for around white dwarfs, that sort of thing.

    Here is an odd thing, if I remember we will collide with the Andromeda Galaxy in 4 billion years , while the sun wont be a Red Giant until about 8 billion years.
    That’s weird!
    The Earth might get stripped out solar orbit before it’s destroyed by the sun!
    So gonna need interstellar flight before 4 billion years, boy how things sped up!

  • Rob Henry August 23, 2012, 20:12

    This seems to imply to me that a system of planets that has been stable for billions of years, still has a high chance of suddenly going horribly wrong. I will never laugh again on seeing a “Worlds in Collision” rapid planetary pinball scheme.

  • Rob Henry August 23, 2012, 20:13

    A. A. Jackson, I find that any subtle mistakes made in the first publication on a topic are often highly instructive to my understanding. I think that omitting such referances is a greater mistake than just etiquette.

  • Christopher Phoenix August 23, 2012, 22:57

    @A.A. Jackson
    Four billion years is a long time to develop interstellar travel- I’m not even sure if humans will still be in their modern form by that time. We’ve only existed in our modern form for 100,000 years if I remember correctly, and our recorded history spans only 5000 years. That is not long compared to 4 billion years.

    However, we may not be so lucky. Earth probably will not remain habitable all the way up to the point when the sun turns into a red giant. Our sun’s luminosity is slowly but steadily increasing- it was once much dimmer than it is today- and Earth will be experiencing a greater and greater solar insolation. This could lead to a runaway greenhouse effect, where the greenhouse effect is not compensated for by other factors (like increased cloud cover), and Earth could become a sizzling hell like Venus.

    Long before the sun becomes a red giant, Earth may have become uninhabitable. Even short of a runaway greenhouse effect, high temperatures will increase the number of high-intensity storms and drastically change our ecosystems, perhaps making Earth quite unpleasant to live on. I don’t see this happening for a long time, but it is something to think about when we consider the ultimate fate of our cosmic home.

    Long before then, we could have founded a vast human civilization among the stars. To any future interstellar empires, the slow spin of our galaxy, the wanderings of the stars, and the inevitable collision with the Andromeda galaxy may seem as unimportant in daily life as the motion of continental plates is to us. Heck, if you have incredibly fast FTL, you might reach the Andromeda galaxy long before it hits us…

  • Adam Crowl August 24, 2012, 1:46

    Hi Al
    The “collision” with M31 isn’t likely to strip Earth from the Sun. Both Galaxies are much too dispersed for that to happen. Saying it’s “not likely” and knowing the actual odds are two different things however. I hope we’ll have a better idea in 3-4 aeons time.

  • Ronald August 24, 2012, 8:53

    With due respect for Adam’s immense knowledge of, and experience with the universe, with ref. to his comment (qraal, August 23, 2012 at 16:34), but the departure from the Main Sequence by our sun may well be the least, or rather last, of our problems:
    A main sequence star like our sun grows about 10% brighter per gigayear, and we are already quite close near the inner edge of our Habitable Zone. Whereas the outer edge of the HZ is rather fuzzy and highly dependent upon planetary conditions (mass, atmosphere), and at least at 1.2 AU but possibly as far out as1.5 AU, most authors agree that the inner edge is rather sharply delimited and somewhere near 0.95 AU.
    In about 0.5 gy the earth will probably get too hot for (most) higher life and in about 1 gy so hot that the oceans will likely evaporate.
    This is also what makes me a rather sceptical about the prospects for living terrestrial planets around some old(er) solar analogs: their greater age and resulting increased luminosity may have rendered any earthlike planets unlivable, in other words: moved the HZ so far outward that these planets have left it on the inside.
    This may also be the case for Alph Cen A: its greater mass (1.1*solar) and age (estimates vary from 4.5 – 6.5 gy, Porto de Mello et al. in The Alpha Centauri binary system, 2008, estimate it between 4.5 and 5.3 gy) mean that it is probably also further in its evolution, which may be partly responsible for its very high luminosity of about 1.52 * solar.
    In the nearby (up to about 50 ly) stellar neighborhood this is also the case with a significant number of solar type stars, including: Delta Pavonis, Mu Arae, 51 Pegasi, Upsilon Andromedae, Nu Phoenicis, 104 Tauri, Tau Bootis, Gamma Cephei, Beta Virginis.

  • Ronald August 24, 2012, 8:59

    I should have added: slightly smaller mass, later spectral class solar type stars, such as Alph Cen B, are a very different story with a much longer period stable Habitable Zone.

  • ljk August 24, 2012, 9:40

    A. A. Jackson said on August 23, 2012 at 12:09:

    “If you are interested in interstellar flight JBIS is your journal … otherwise , alas!”

    Centauri Dreams/Tau Zero will remedy that situation.

    By the way, is your paper available online anywhere?

    Then A. A. Jackson said on August 23, 2012 at 19:51:

    “Here is an odd thing, if I remember we will collide with the Andromeda Galaxy in 4 billion years , while the sun wont be a Red Giant until about 8 billion years.
    That’s weird!
    The Earth might get stripped out solar orbit before it’s destroyed by the sun!
    So gonna need interstellar flight before 4 billion years, boy how things sped up!”

    Unless new data has come in that I have not seen, the Sun will be a red giant in about 5 billion years time. Yellow dwarf stars have lifetimes of about 10 billion years, then they turn into white dwarfs and eventually black dwarfs (aka a dead hunk of charcoal). Sol may not engulf Earth as previously believed, but if our planet is still around, it will be turned into a ball of molten slag.

    I have also read that the Sun will begin expanding enough that in about one billion years or so, Earth will become too warm to be inhabitable long before our star becomes a red giant. Ironically that is also around the same time that the Moon will have moved far enough away from our planet that we will no longer see total solar eclipses, only annular and partial ones. Assuming anything intelligent is still on Earth, or that Earth even exists in its current form by then.

  • A. A. Jackson August 24, 2012, 11:02

    Yes it’s true the Sun will make the Earth (at least at the surface) uninhabitable long before the Helium flash. So that’s out there more or less about 1.5 billion years, if I remember right.
    (The time scale for a possible extinction impactor or super volcano is a lot shorter than that. )
    If we survive! We may have either left for the better climes in the solar system or skedaddled altogether , or!, remember the guy with the idea of gradually moving the Earth outward with a repeating massive asteroid, I think. (Frankly I wouldn’t take the chance.)
    O well it’s very likely the solar system will be un-disrupted by the time of Andromeda collision. And yes, Adam, probably that would not take the solar system apart. Makes me wonder what it does dynamically to the local Solar stellar neighborhood?
    I have seen the simulations but can’t believe we know enough about running a million body problem like that in order to figure what it does to planetary systems, it may not be possible for a long time , to do that.

  • Phil August 24, 2012, 12:09

    Out of curiosity if the Sun retains (or looses) mass as it ages and it also expands, what is it that keeps planets from stealing material from the parent star? My naive view tells me that if any material came near the Earth, the Earth’s gravity would be stronger than the Sun’s in its local neighbourhood (the moon is a perfect example of this) and the Sun’s material would be pulled to the Earth. My limited knowledge of physics also tells me that the sun’s gravitational effect would still be the same as a point source at it’s center of mass even though it’s ballooned up.

    Where is my logic going wrong? Does the planet just loose more material than it gains due to heating?

  • Rob Henry August 24, 2012, 18:30

    A. A. Jackson, whether the Earth is worth saving from our Sun’s red giant phase, and whether we survive are two different questions – thus there is no risk involved and much reward possible in moving our planet then.

    The key is that the sort of (post)human civilisation that is likely to do it is unlikely to dwell here (apart, perhaps from a few primitive tribesmen and occasional visiting anthropologists and biologists). The place would be a revered nature reserve, not the seat of our civilisation.

  • Eniac August 25, 2012, 0:12

    Yellow dwarf stars have lifetimes of about 10 billion years, then they turn into white dwarfs and eventually black dwarfs (aka a dead hunk of charcoal). Sol may not engulf Earth as previously believed, but if our planet is still around, it will be turned into a ball of molten slag.

    Can’t resist my nitpicking urge:

    Dwarfs, white or black, are made of degenerate matter, which has nothing in common with charcoal, except that it may (but not necessarily does) contain the same nuclei. And, for all intents and purposes, the Earth already is a ball of molten slag.

  • Adam Crowl August 26, 2012, 20:28

    Hi ljk & All
    Larry, the Main Sequence ends when the Sun’s core hydrogen burning ends – at an age of 10 Gyr, so 5.5 Gyr from now. However the Red Giant state takes about 2 Gyr to fully evolve. Initially the Sun is only a bit brighter and a bit cooler, thus a bit bigger, and it stays that way for over 1.5 Gyr. Then it starts ramping up. The very large Sun phase, when it bloats to +100 times larger, comes right at the very end of the 2 Gyr process, and lasts about ~1 Myr. So very tiny period compared to the ponderous prelude.

  • Brett Bellmore August 26, 2012, 20:37

    Shielding the Earth against increased solar luminosity ought to be a relatively trivial task for a future civilization. WE could probably manage it in a pinch, given a couple decades warning of the need.

  • Securis August 28, 2012, 13:02

    Recent paper about a planet getting disintegrated by it’s sun:


    From the Abstract:

    Our quantitative analysis supports the hypothesis that the transit signal of KIC 12557548 b is due to a variable cloud of
    dust, most likely originating from a disintegrating object.



    The first image of the new weather satellite MSG-3 is worth a look.