The question How are rogue planets discovered? describes the difficulties in finding planets (or planet-sized objects) that are floating through space without being under the influence of any star or Galaxy system.

What would the possible surface conditions of these planets be? Could chemotrophic life arise on such planets?

  • $\begingroup$ I have removed all questions about finding such objects and included the link to the other question so that this one is only about their surface conditions. Better not ask two questions at once on SE sites. $\endgroup$
    – user1569
    Commented Nov 9, 2018 at 15:11

2 Answers 2


Rogue planets have two formation mechanisms: Independent formation and ejection.

An independently formed rogue planet would have condensed out of the nebular material by itself and not formed from a young star's protoplanetary disk. We understand how individual stellar-mass objects condense and which have been tested by observation, but I'm not aware of any tested studies which predict rate at which Earth-massed objects independently form. (We certainly have not observed a large population of them, and the gravitational micro-lensing studies which have been done would have detected them if they existed.) But since Brown Dwarfs are fairly common, it seems safe to assume that there is a population of super-Jupiters that formed by direct condensation.

The planets formed by ejection should range in size from meteors right up to super-Jupiters and should be broadly similar to the size distribution of bound planets. (If small planets tend to be closer to their star, there may be some bias against them being ejected.)

Planets are mostly ejected early in a system's life, but it can happen at any point in a star's lifetime -- planetary system dynamics never becomes perfectly stable. And if there's a near passage of another star or a large rogue planet, ejections can occur even after billions of years of stability. (See Fritz Leiber's "A Pail of Air"!) But the population of ejected planets is probably very heavily weighted towards planets ejected soon after formation.

The distinction between independently formed and ejected planets is important because when a planet forms it is very hot from the gravitational energy released in formation, and most likely initially inimical to life forming. But it cools (the surface much faster than the interior) and eventually life can form.

If the planet remains in orbit around a star, the surface temperature drops asymptotically towards an equilibrium temperature where the sum of the star's radiational input and the heat leakage from the still-hot interior balances the IR radiation leaked by the planet into space. Frequently, that will be in a temperature range where life can form on the surface.

Once a planet is ejected, the equilibrium temperature will be much lower. For example, for Earth at 4.5 GY old, the Sun's radiation dumps 3000 times as much energy onto the Earth's surface than does leakage from the interior, (see https://en.wikipedia.org/wiki/Earth%27s_internal_heat_budget for details.) The time it takes for an Earth-sized rogue planet to cool to Earth's current temperature is on the order of 10 MY and after that it would just keep on cooling down to a surface temperature on the order of 30K.

So the question of life forming on an Earth-like planet is one of whether or not there is time before the surface layers freeze solid, since that would appear to halt the formation of life.

All bets are off for larger planets which (a) cool more slowly and (b) probably have enough more water to have subsurface liquid oceans even after the surface freezes solid.

For planetary bodies where life does have time to form, the only source of energy would be the heat leaking from the interior either directly due to the heat gradient (a very diffuse source), or indirectly from the equivalent of Earth's deep sea vents. The latter seems more likely since the steeper heat and chemical potential gradients are a lot easier to exploit.

Hal Clement wrote two scientifically excellent stories set on such planets. One, Star Light, has high-gravity aliens working with humans to explore the surface of Dhrawn, a Brown Dwarf. The other was short fiction and had humans meeting intelligent life living on an closer to Earth-sized rogue planet. (I think it was "Sortie" and sequels, but I'm not sure.)

In any event, it seems likely that there would be a great variety of different kinds of rogue planet that might potentially support life.

  • $\begingroup$ "All bets are off for larger planets " They also have moons..., but would that be true for ejected planets? Perhaps not. There is also (significant) heat from radioactive decay for rocky bodies. $\endgroup$
    – ProfRob
    Commented Nov 9, 2018 at 14:42
  • $\begingroup$ "An independently formed rogue planet would have condensed out of the nebular material by itself and not formed from a young star's protoplanetary disk. " Some would argue that this isn't a planet, it is a low-mass brown dwarf. There really isn't any way that small rocky bodies can form like this. $\endgroup$
    – ProfRob
    Commented Nov 9, 2018 at 14:43
  • $\begingroup$ @Rob Jeffries "Some would argue that this isn't a planet" The dynamical definition of planets has the advantage of being several thousand years old, but no other. I think it makes a lot more sense to call something a planet based on its physical attributes. Under the dynamical definition, if Earth were to be ejected from the Solar System, it would cease to be a planet. Yet it would still be studied by planetologists and not by stellar astronomers. Regards of what they say, nearly all astronomers use the physical rather than the dynamical definition. $\endgroup$
    – Mark Olson
    Commented Nov 9, 2018 at 14:47
  • $\begingroup$ @Rob Jeffries: I vaguely recall reading a paper which discussed ejections and lunar stability -- maybe on ArXiv? But the stability of lunar orbits would depend on how tightly the moon was bound compared with the tidal forces tending to disrupt the system. So it's pretty clear that a planet ejected by a passing stellar-mass object would take all but its most weakly bound moons with it, but ejections due to lower-mass objects would tend to disrupt the satellite system also. $\endgroup$
    – Mark Olson
    Commented Nov 9, 2018 at 14:51
  • 1
    $\begingroup$ As regards the other point, I would not have imagined you form a rocky planet from the collapse of a gas cloud that is 98% H and He. There is no suggestion that brown dwarfs have anything but a "normal" abundance. A protoplanetary disk allows the process of chemical differentiation to occur and the formation of rocky/icy planets. $\endgroup$
    – ProfRob
    Commented Nov 9, 2018 at 16:54

David Stevenson has theorized that a rouge planet could be ejected with considerable hydrogen in the atmosphere that would lead to a high-pressure hydrogen atmosphere. This is highly opaque to infrared and could conceivably hold the little bit of internally-generated heat well enough that water could exist on the surface. Here's an abstract of that paper: Life-sustaining Planets in Interstellar Space.

During planet formation, rock and ice embryos of the order of Earth's mass may be formed, some of which may be ejected from the Solar System as they scatter gravitationally from proto-giant planets. These bodies can retain atmospheres rich in molecular hydrogen which, upon cooling, can have basal pressures of 102 to 104 bars. Pressure-induced far-infrared opacity of H2 may prevent these bodies from eliminating internal radioactive heat except by developing an extensive adiabatic (with no loss or gain of heat) convective atmosphere. This means that, although the effective temperature of the body is around 30 K, its surface temperature can exceed the melting point of water. Such bodies may therefore have water oceans whose surface pressure and temperature are like those found at the base of Earth's oceans. Such potential homes for life will be difficult to detect.

Another such paper is: Constraints on the free-floating planets supporting aqueous life, by Viorel Badescu.


You must log in to answer this question.

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