Rogue planets can be designated as rogue planets due to one of two possible scenarios: 1) the "rogue planet" formed as a sub-brown dwarf, or 2) the planet was ejected from its home star system. I want to make it clear that here, I'm interested in objects of the second category - objects that are similar to the planets we find in our solar system.

Could a rogue planet, having been ejected from its home star system, be gravitationally re-captured by another star system? If so, would it be likely to disrupt other bodies in the stellar system (I know this depends on size; let's say that the rogue planet is a "Hot Jupiter")?

I realize that we don't know a whole lot about these objects (let's face it, we only know of a handful), so I'm not looking for examples, but any logical answers would be pretty helpful.

  • 1
    $\begingroup$ voting without commenting is fine and should be the norm. As to your question, I think your first paragraph contains an error: you are interested in objects of the second category (not first); please correct. $\endgroup$
    – Walter
    Sep 5, 2014 at 9:39
  • $\begingroup$ @Walter Thanks for the correction; I should have seen that before. $\endgroup$
    – HDE 226868
    Sep 6, 2014 at 20:05
  • $\begingroup$ "but why are you interested?" I'm hoping to get my current writing project completed before I can no longer work on it. I'm writing a science fiction/fantasy novel about a planet that went rogue long ago, when its star system was grazed by a very large event (something like a rapidly approaching neutron star(s)?) The story will end by being captured by a star the planet has been approaching. While the story takes place in a universe not ours, and not necessarily bound by all the laws we know of in our universe, I do want to maintain as high a level of believability as possible. $\endgroup$
    – user7358
    May 31, 2015 at 0:20

2 Answers 2


The processes by which planets can be removed from their parent stars are discussed in some detail by Davies et al. (2014). These include direct ejection through interactions with other stars in dense birth environments; the ejection of planets due to planet-planet interactions, again usually taking place fairly early in a star's life; the later ejection of planets after an instability due to a passing star. There is also a possibility of exchange interactions, i.e. a planet will swap stars during an interaction. Again, this is most likely to happen in a dense birth environment rather than later in a star's life.

If the ejection of the planet happens while they are still in the dense environs of their birth, then recapture by another star is possible. A very recent paper by Wang et al. (2015) explores this very possibility using N-body simulations, but there are other papers to look at by Parker & Quanz (2012), or Perets & Kouwenhoven (2012). These papers seem to agree that the process of recapture is not common (but not rare either) and will happen sooner rather than later, before the cluster disperses. The latter of these papers has some particularly eye-catching results: they find that if the number of free floating planets is comparable to the number of stars then about 5% of stars recapture a planet. The capture tends to put the planet in a wide, eccentric orbit (hundreds of au or more). Capture is more likely for massive stars, but they claim that even brown dwarf-planet or planet-planet pairs could be formed in small numbers.

Once the planet has left the cluster environment then, as is the case of star-star interactions in the Galactic field, the probability of encountering another star and being captured is very remote.

  • $\begingroup$ Nice to see you quote Melvyn (et al). One thing all these studies (as well as my answer above) ignore, I think, it the possibility of tidal capture (or the aid of tidal dissipation in dynamical capture) and the role of tides in the orbital evolution of captured planets. $\endgroup$
    – Walter
    May 31, 2015 at 17:12

For a start, the hot jupiters (HJ) are not similar to the planets of the Solar system (they are much hotter and very close to their host star). It's essentially impossible for HJs to be ejected away from their star's sphere of influence. The reason is that they are gravitationally rather tightly bound to their star, so that any gravitational interaction with a third body that would gain them enough energy for ejection needs to be closer than is possible without it leading to a collision with that third body.

So rogue planets that didn't form independent of a star must have been much less bound to their parent star, similar to the outer planets of the Solar system. Of course, once ejected (free-floating), such an object can in principle be re-captured into another planetary system or a binary system. However, this is quite unlikely, since the planet must move with a velocity very close to that of its capturer. Also, it must involve a third body (either another planet or star bound to its capturer) to absorb (into its orbit) the orbital energy (and angular momentum) released by the capture.

The only possibility is that the planet was ejected very early on, when the stellar association / group / open cluster within which it was born had not yet been dissolved. Then it may find another home with one of the other members of that association.

I don't think that anybody has estimated the actual probabilities for this to happen, but why are you interested?

As a side remark, it seems clear that orbiting planets form from / in gas discs surrounding protostars, but it's not clear whether brown dwarves cannot also form this way. Whether or not an object formed this way becomes a gas planet, a brown dwarf, or even a star depends entirely on its mass and not the formation process.

  • $\begingroup$ "However, this is quite unlikely". Can you back that up since it is actually the crux of the question? Any ejection/recapture events are going to take place early on when the system is in a dense environment. $\endgroup$
    – ProfRob
    May 31, 2015 at 10:56
  • $\begingroup$ Actually, I take some of that back. Direct ejection events take place in dense environments. Instabilities leading to later ejection could also happen on longer timescales. $\endgroup$
    – ProfRob
    May 31, 2015 at 11:22
  • $\begingroup$ Sorry for waiting for almost 8 months to get back to you on this, but I was interested for the purpose of mentioning it in a story - an idea which I discarded quite some time ago. $\endgroup$
    – HDE 226868
    May 31, 2015 at 18:24

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .