≡ Menu

Distant Companions: The Case of HD 106906 b

When the pace of discovery is as fast as it has been in the realm of exoplanet research, we can expect to have our ideas challenged frequently. The latest instance comes in the form of a gas giant known as HD 106906 b, about eleven times as large as Jupiter in a young system whose central star is only about 13 million years old. It’s a world still glowing brightly in the infrared, enough so to be spotted through direct imaging, about which more in a moment. For the real news about HD 106906 b is that it’s in a place our planet formation models can’t easily explain.

discovery image_2

Image: This is a discovery image of planet HD 106906 b in thermal infrared light from MagAO/Clio2, processed to remove the bright light from its host star, HD 106906 A. The planet is more than 20 times farther away from its star than Neptune is from our Sun. AU stands for Astronomical Unit, the average distance of the Earth and the Sun. (Image: Vanessa Bailey).

Start with the core accretion model and you immediately run into trouble. Core accretion assumes that the core of a planet like this forms from the accretion of small bodies of ice and rock called planetesimals. The collisions of these objects bulk up the core, which then attracts an outer layer of gas. The problem with applying this model to HD 106906 b is that it’s about 650 AU from its star, far enough out that the core doesn’t have enough time to form before radiation from the hot young star dissipates the already thin gas of the outer protoplanetary disk. Get beyond 30 AU or so and it gets harder and harder to explain how a gas giant forms here.

Gravitational instability is the key to the other major theory of planet formation, but it’s challenged by HD 106906 b as well. Here the instabilities in a dense debris disk cause it to collapse into knots of matter, a process that can theoretically form planets in mere thousands rather than the millions of years demanded by core accretion. But at distances like 650 AU, there shouldn’t be enough material in the outer regions of the protoplanetary disk to allow a gas giant to form.

Grad student Vanessa Bailey (University of Arizona), lead author on the paper presenting this work, describes one of two alternative hypotheses that could explain the planet’s placement:

“A binary star system can be formed when two adjacent clumps of gas collapse more or less independently to form stars, and these stars are close enough to each other to exert a mutual gravitational attraction and bind them together in an orbit. It is possible that in the case of the HD 106906 system the star and planet collapsed independently from clumps of gas, but for some reason the planet’s progenitor clump was starved for material and never grew large enough to ignite and become a star.”

Making this explanation problematic is the fact that the mass ratio of the two stars in a binary system is usually more like 10 to 1 rather than this system’s 100 to 1, but the fact that the remnants of HD 106906’s debris disk are still observable may prove useful in untangling the mystery. A second formation mechanism suggested for planets at this distance from their star is that the planet may have formed elsewhere in the disk and been forced into its current position by gravitational interactions. The preprint notes the problem with this scenario:

Scattering from a formation location within the current disk is unlikely to have occurred without disrupting the disk in the process… We also note that the perturber must be > 11 MJup; we do not detect any such object beyond 35 AU…, disfavoring formation just outside the disk’s current outer edge. While it is possible that the companion is in the process of being ejected on an inclined trajectory from a tight initial orbit, this would require us to observe the system at a special time, which is unlikely. Thus we believe the companion is more likely to have formed in situ in a binary-star-like manner, possibly on an eccentric orbit.

HD 106906 b was found through deliberate targeting of stars with unusual debris disks in the hope of learning more about planet/disk interactions. The team used the Magellan Adaptive Optics system and Clio2 thermal infrared camera mounted on the Magellan telescope in Chile. The Folded-Port InfraRed Echellette (FIRE) spectrograph at Magellan was then used to study the planetary companion. The Magellan data were compared to Hubble Space Telescope data taken eight years earlier to confirm that the planet is indeed moving with the host star.

This unusual gas giant thus joins the small but growing group of widely separated planetary-mass companions and brown dwarf companions that challenge our planet formation models. The massive, ring-like debris disk around HD 106906 helps us constrain the formation possibilities, but we’ll need more data from systems whose disk/planet interactions can be observed before we can speak with any confidence about the origin of these objects.

The paper is Bailey et al., “HD 106906 b: A planetary-mass companion outside a massive debris disk,” accepted at The Astrophysical Journal Letters and available as a preprint. This University of Arizona news release is also useful.


Comments on this entry are closed.

  • Demongo December 9, 2013, 12:10

    A “mini”-Dyson sphere…?

  • Klaus Lang December 9, 2013, 12:14

    Maybe HD 106906 b is a captured rogue planet?
    Klaus Lang

  • Dwight Williams December 9, 2013, 14:49

    Well, however this happened, it’s a record-breaker for the moment. It’ll be interesting to keep an eye on further developments.

  • Enzo December 9, 2013, 15:45

    Another 3 similar systems (large planet/brown dwarf far away from the main star), directly imaged, including around double red dwarfs :

  • David Cummings December 9, 2013, 19:03

    Two things we can be sure of in the future: computers are going to get ever faster and telescopes are going to see ever more detail. I wonder if there is something like a Moore’s Law for telescopic imaging.

  • Senua December 9, 2013, 20:06

    For so long we had only our own solar system to go on. Now we are realising that there appears to be other ways planets can form and that solar systems around other stars can be very different indeed to our own. Very interesting.

  • Anton Szautner December 10, 2013, 4:25

    One does not need to invoke capture or gravitational perturbation mechanisms for planetary bodies after they have become compact objects. In the diverse density conditions that obtain within star-forming environments it is entirely conceivable that local high-density condensations within dust or molecular clouds naturally form in close proximity to one another as a matter of course. Simulations have long since already and repeatedly hinted at this tendency. Its not JUST a matter of gravitational dynamics between discrete-object foci, nor is it all contingent on simplistic models of proplyd formation isolated from a real environment they are typically embedded in. There can also be significant fluid-(gas) dynamic interactions between dusty gas clouds that happen to cluster in close proximity and probably with a range of masses that follow some virial distribution. Given these considerations, one might expect such proximal condensations to undergo significant interactions (thermodynamic, ram-pressure, etc.) in order to end up producing a widely separated yet gravitationally bound system such as this. From this perspective, it should not be at all surprising that cloud clumps will occasionally end up to produce a wide range of configurations that our clean models do not accommodate, including seeing examples just like that described here. Nature remains, it must be allowed, far more rich and dirty than our simplistic models can thus far account for. We should quit the tedious characterization so beloved of the pop-press release brigade that spouts off on how ‘impossible’ a finding is, or describes something that “shouldn’t be there”. Whatever happened to the genuine acceptance of mystery that observation reveals WITHOUT the constant and instantaneous comparison to what we think we understand according to theory? From the PR standpoint – how the public gets the process – its looking very much as if we are becoming very much less concerned with paying attention to what we’re seeing and working to build upon existing theoretical explanations utilizing known factors to account for such observations than we are in a constant expression of hand-wringing over how much our existing models so horribly depart from their ability to explain what we discover. The shift in emphasis has been gradual and subtle, but troubling.

  • Kamal Ali December 10, 2013, 4:31

    “While it is possible that the companion is in the process of being ejected on an inclined trajectory from a tight initial orbit, this would require us to observe the system at a special time, which is unlikely.”

    My 2c:
    If we’ve observe thousands of such young stars, at least a few of them would
    be expected to be observed while being ejected ? I thought simulations
    had suggested ejections were quite common and so it’d be surprising if
    we never see companions with 10^3 or 10^4 AU separation.
    Orbital motion has been seen for HR8799 and perhaps will be seen for
    this system too (after a few years)
    telling us if the companion is traveling with a high velocity.

  • Michael December 11, 2013, 8:02

    If the Planet was ejected has the other been forced into a tight orbit or been swallowed up by the Star? On another note it would be interesting if this Planet has any moons which would be nice and warm and if we are lucky may be spotted by transit.

  • Michael December 12, 2013, 4:01

    On thinking about the possiblity of moons around the Planet would the absence of them point to the fact that it was eject from near the Star as these gravitational interactions are quite disruptional. Moons around this planet could point to the fact that it was created effectivily on its own.