Before I get into some NASA-funded exoplanet work that grew out of a study of the binary nature of Pluto and Charon, I want to mention that NASA TV will air an event of exoplanet interest on Tuesday the 7th, from 1700 to 1800 UTC. A panel of experts will be discussing the search for water and habitable planets, presenting recent discoveries of water and organics in our own system and relating them to the search for Earth-like worlds around other stars. NASA streaming video, along with schedules and other information, can be found here.
As to that Pluto/Charon work, it actually took Ben Bromley (University of Utah) and Scott Kenyon (Smithsonian Astrophysical Observatory) much deeper into space when they began relating it to the formation of planets in circumbinary orbits around binary stars. These are planets that orbit both stars rather than one of the two. In other words, the work, funded by NASA’s Outer Planets Program, has examined the familiar ‘Tatooine’ scenario from Star Wars, where a sunset view may consist of two stars approaching the horizon simultaneously, a scenario that for a long time was deemed impossible.
Bromley and Kenyon’s mathematical models show that binary stars orbited by planetesimals can produce Earth-class planets close enough to the stars to be in the habitable zone as long as they are in concentric, oval-shaped rather than circular orbits. Such orbits change our perspective about what can happen around binary stars. Says Bromley
“…planets, when they are small, will naturally seek these oval orbits and never start off on circular ones. … If the planetesimals are in an oval-shaped orbit instead of a circle, their orbits can be nested and they won’t bash into each other. They can find orbits where planets can form.”
And he adds:
“We are saying you can set the stage to make these things. It is just as easy to make an Earthlike planet around a binary star as it is around a single star like our sun. So we think that Tatooines may be common in the universe.”
Image: In this acrylic painting, University of Utah astrophysicist Ben Bromley envisions the view of a double sunset from an uninhabited Earthlike planet orbiting a pair of binary stars. In a new study, Bromley and Scott Kenyon of the Smithsonian Astrophysical Observatory performed mathematical analysis and simulations showing that it is possible for a rocky planet to form around binary stars, like Luke Skywalker’s home planet Tatooine in the ‘Star Wars’ films. So far, NASA’s Kepler space telescope has found only gas-giant planets like Saturn or Neptune orbiting binary stars. Credit: Ben Bromley, University of Utah
The conclusion is hardly intuitive, for two stars hosting an infant planetary system in this configuration perturb the region around them. It has been assumed that their gravitational interactions will clear out orbits to distances between 2 and 5 times the binary separation, with random motions among the planetesimals becoming high and destructive collisions frequent. More gentle merges and nudges are thought to be necessary for the formation of planets.
We do have seven circumbinary planets in inner orbits that have been identified by the Kepler telescope, but all are gas giants of Neptune-size or larger. Prevailing theories have suggested that such planets — the first discovered was Kepler-16b, a Saturn-mass planet at an orbital distance of 0.7 AU orbiting a K-class star and an M-dwarf — formed much further out in their systems and migrated closer to the stars because of gravitational interactions with another planet or the disk of gas surrounding the binary pair.
That conclusion may still hold for the gas giants so far detected, but the paper looks at all the Kepler circumbinary planets in light of the authors’ modeling and argues that there are other solutions beyond migration, particularly for smaller worlds. From the paper:
Toward understanding how circumbinary planets form, we re-examine a fundamental issue, the nature of planetesimal orbits around binary stars… [W]e describe a family of nested, stable circumbinary orbits that have minimal radial excursions and never intersect. While they are not exactly circular, these orbits play the same role as circular paths in a Keplerian potential. Gas and particles can damp to these orbits as they dynamically cool, avoiding the destructive secular excitations reported in previous work. Thus planetesimals may grow in situ to full-fledged planets.
These stable, non-Keplerian orbits, which the authors call ‘most circular,’ do not cross. Objects in such orbits respond to the mass of the central binary but also ‘experience forced motion, driven by the binary’s time-varying potential.’ It is this perturbation that keeps the planet from maintaining a circular orbit. A ‘most circular’ orbit is defined here as ‘having the smallest radial excursion about some guiding center, orbiting at some constant radius Rg and angular speed ?g in the plane of the binary.’
The result is that orbits inside a critical distance, pegged at twice the separation of the binary stars, are unstable, but beyond this distance, planet formation appears possible. After presenting their mathematical analysis, the authors go on to look at the Kepler circumbinary planets, finding that while in situ formation can produce rocky planets at the observed distances, this is unlikely to have occurred for the gas giants found by Kepler.
Most of the Kepler circumbinary planets seem to implicate migration of mass into the region where these objects are observed. Without larger samples of planets, it is impossible to distinguish between models where migration precedes assembly from those where migration follows assembly. All of the planets are too massive to allow in situ formation with no migration. However, the high free eccentricity observed in Kepler-34b and Kepler-413b are consistent with scattering events. Improved constraints on the orbits and bulk properties (mass, composition, spin, etc.) might allow more rigorous conclusions on their origin.
The argument here is not that migration cannot be a viable way to move planets into inner circumbinary orbits, but that both migration and in situ formation are possible. Outside of the critical region near the binary, planet formation can happen the same way it does around a single star, with planetesimal orbits allowing mergers, fragmentation and stirring of the material that will grow into stable worlds over time. In our own Solar System, the small moons of the Pluto/Charon binary may have formed in the same way.
We may not be seeing small, rocky worlds in inner circumbinary orbits that formed in situ because they are so much smaller than the gas giants thus far detected, but if the oval ‘most circular’ orbits the authors describe are indeed possible, we should start finding them. Migrating gas giants explain our current detections, in other words, but habitable rocky worlds in such systems should be able to form and remain stable in their original orbits.
The paper is Bromley and Kenyon, “Planet formation around binary stars: Tatooine made easy,” submitted to The Astrophysical Journal (preprint).
I think that “Tatooine”planets will see a lot of investigation in the future. There has been a lot of work looking at both the orbital stability of circumbinary or so called “p” planets and also habitability around around binary stars. ( planets that orbit just one star of a binary are called “s” type. As a rough rule of thumb their orbits are stable to about one fifth of the closest approach of the constituent stars of the binary , so for alpha centauri system this would be one fifth of 11 AU or 2.2 AU. This is very close to the most recent calculated value of 2.4 AU) .
This work has been based on the smallish number of planetary candidates discovered in binaries by Kepler, somewhat unexpectedly it must be said. no terrestrial planets as yet , though the authors comment that exomoon habitability could be increased too by alllwing larger Hill radii and freeing a moon from the potential harmful effects ofbtidal heating and the magnetosphere of the parent planet. ” Planetary aggression” ( see below )Orbital stability is based on the usual features such as semi major axis and also eccentricity. The distance varies also depending on whether a “p”planet is in a prograde or retrograde orbit with the latter being closer. The closest stable orbit being called ‘ a’ crit and is roughly 5 times the semi major axis of the binary for a retrograde planet and 6 for a prograde ( centred on the centre of gravity of the binary system) .
This is NOT the inner habitable orbit of course which will be significantly larger.More recently, Mason et al published in the Journal of Astrobiology on what they describe as the “binary habitability mechanism”, a process they theorise provides habitable “niches” for certain combinations of binary stars at certain semi major axes and eccentricities. They have calculated various combinations for these niches. Unsurprisingly they are all close binaries with periods of 50 days or less , though not “contact” binaries with very small semi major axes. To date few if any planets have been discovered around these with the exchange of angular momentum between the binary and protoplanetary disk being postulated for premature dissipation of the latter thus preventing planetary formation. Despite this the Mason paper is very optimistic for certain combinations of stars that they believe even have p planets that fall into Heller and Armstong’s “super-habitability” classification.
They found that the main driver is that the components of a some binary systems actively act as a brake on each other’s rotation thereby reducing their rotation rates far quicker than in isolation with a resultant reduction in magnetic field strength. Effectively “aging” the stars. Stellar magnetospheres have been blamed for excessive chromospheric activity which leads to the release of dangerous levels of X Ray with High energy U.V ( XUV) and strong stellar winds ( SW) in young ( rapidly rotating) stars . They term this ominously as “stellar aggression” . We all know that M stars in particular illustrate this when young and also before they enter the main sequence. They look a variety of real examples of different combinations of stars of different sizes and describe a primary and secondary star . The stellar sizes vary from Sun analogues to late K ( 0.7 Msun) for the primary in combination with a similar sized secondary all the way down to an early M class star ( 0.55Msun). In all cases the binaries demonstrated marked reductions in XUV and SW , the only variations being the time taken to achieve this via “pseudosynchronisation” of rotaion according to initial period and eccentricity. Each binary demonstrated the usual range of habitable zones as with single stars though they all tended to be bigger than for their single equivalents . The biggest reductions in XUV and SW were seen with a late K and early M combination , with increasingly long lasting habitable zones as the difference in mass between primary and secondary increased , to the point that planetary habitability here would be dependent on the features of the planet itself rather than related to stellar properties. Critically they also looked at levels of photosynthetically usable energy and found it to be higher for all binary combinations versus single stars. Not all binary combinations produce these “niches” and the group have produced a tool for calculating the binary habitability mechanism, available via the link from the paper or August’s arXiv preprint. All in all very exciting , especially as both TESS and PLATO could pick up such systems for subsequent examination.
One caveat to what is a very exciting and heart warming paper is that the authors did not look at what happens in these systens before they entered the main sequence . This is highly relevent as it is well known that this time is particulatly important for M stars especially , given they have bern shown to have an extended and active ( UV specifically) pre main sequence life.
If relatively close binary systems form would co-rotating or contra-rotation be preferred as it would a direct impact on planet formation? I would think co-rotation would have a negative impact as the material would tend to collide between their discs and fall inwards where as in the contra-rotation case it would allow a more stable environment for planet formation.
The wonderful thing about Keplerian Orbits is not just their complete stability in all two body systems, but that they are still stable after any single perturbation. I suspect that if one of these ‘most circular’ orbits is perturbed the planet will spiral inwards or outwards. It could be my poor imagination, but I just have difficulty seeing how the gain and loss of angular momentum from the co-orbiting primaries will tend to cancel each other without special circumstances.
The article by Benitez-Llambay in Nature that Paul describes in the “Migratory Jupiter ” post above suggests that planet /dusk integration is more complex than first thought. If accretion is sufficiently fast it produces a “heat torque” that actually causes Otward migration , though inward migration is still maintained at lower accretion rates. There was an article on circumbinary accretion in arxiv yesterday that talked of turbulent zones around binaries extending to 2-4AU that could prevent build up of more than small planetesimals . That’s obviously important as any habitable zone is likely to lie within this area so for habitability we would require inward migration . Trouble with that is that accretion further out produces icy planets that on melting could become “Ocean worlds” and no one can imagine clearly they would work , particularly in terms of habitability. Claim and counterclaim in the published literature .Too much theory at present. We need to get observing. Role on TESS and PLATO.