Giant planet seeks nursery place

Article metrics

A further discovery of a planet in a binary star system — this time close in — could prove a problem for accepted theories of planetary formation. The implication is that there are more planets out there than we thought.

The first discovery ten years ago of ‘Pegasi’ planets1 — giant planets 100 times closer to their star than the Solar System's giant planets are to the Sun — shook long-held conceptions of planetary-system formation, which took the Solar System as a template. To square the new observations with the conventional model, some astronomers adopted theories of violent planetary migrations, in which the planets move around after they form. The discovery by Maciej Konacki (page 230)2 of a giant planet in a system where the gravitational pull of a second star would disturb the planet's putative nursery will now place severe constraints on such theories.

Astrophysicists largely agree that giant planets form from a protoplanetary disk that surrounds an infant star — exactly how is a subject of debate. Before the discovery of extrasolar planets (planets beyond our Solar System), Jupiter was assumed to have grown from a core of solid material that, once its gravitational pull was great enough, attracted gas (mostly hydrogen and helium) until it reached its current size of 318 Earth masses. Such ‘core accretion’ only occurs if there is enough solid material to start with — this would preferentially be in cool regions of the protoplanetary disk far from the star, where the heavy elements are largely in a condensed form. The so-called ice line, beyond which this occurs, is typically at least 3 AU from the star. (1 AU, or astronomical unit, is the Earth–Sun distance.)

Thus, the discovery of 51 Peg1, a giant gaseous planet that orbits its parent star at only 0.05 AU, was problematic: if this planet had to have formed at a distance of at least 3 AU, what was it doing so close to its host star? Orbital migration, introduced by theorists to explain this discrepancy, holds that, once the star had accreted a critical mass of around one Jupiter mass, gravitational forces between it and the protoplanetary disk caused the planet to lose angular momentum and thus spiral towards the star. (How Jupiter remained at its original distance is unclear in this theory.) Competing theories of in situ formation of 51 Peg have been proposed3,4, but the inner regions of the protoplanetary disk have only enough material to form planets with sizes totalling a few Earth masses — too little to attract the hydrogen and helium gas needed to form a giant planet. Thus, the model of core accretion and ensuing orbital migration became widely accepted as explaining the formation of giant planets.

This model is challenged by Konacki's discovery2, which was made with the Keck Telescope on Hawaii. Konacki used the ‘Doppler wobble’, a now-standard technique that measures the subtle acceleration of a host star that results from the presence of an unseen companion. What Konacki found is another giant planet with a short orbital period (more than 30 are known). The importance of the finding lies in where he found it — in orbit around a star that is part of a close binary system called HD 188753.

More than 60% of stars in our Galaxy are gravitationally bound to a companion, so the Sun is in a minority. Most searches for extrasolar planets have avoided such binary systems because light from two stars complicates analysis, and it is not clear how the presence of a second star would disturb the process of planet formation. Planets have been discovered in binary systems, but only where the second star is too far away from the first for it to have much influence — the closest binary system where a planet had been discovered was the brighter component to γ Cephei5, where the orbital period of the two stars round each other is some 56 years.

The binary orbital period of HD 188753 is just 25.7 years, and the orbital separation of the stars, both of Sun size, is a mere 12.3 AU — about the distance from the Sun to Saturn. Konacki's velocity measurements reveal that the primary star (the more massive star, denoted HD 188753A) has a planetary companion of a minimum of 1.14 Jupiter masses that orbits the star every 3.35 days at a distance of about 0.05 AU. Yet according to the orbital migration theory, this planet should not exist. The secondary star is so close that its gravitational pull would have stripped away the protoplanetary disk of the primary star — where, even if it later migrated, the planet must have formed — reducing the disk to a radius of just 1.3 AU. But within this radius, ices are unlikely to last and so cannot contribute to the formation of a massive core. The alternative explanation — that the planet formed where it is — would challenge the standard picture, but runs into the problem of where the necessary solid material came from.

Further evidence6 that the region in which a giant planet can form might be smaller than current models allow comes from the exoplanet host star Gl 86, whose companion star, thought to be of brown-dwarf mass, has now been shown to be a white dwarf — an exhausted giant star that has collapsed in on itself. Depending on how big this second star was, there may also have been no ice-containing outer region of the protoplanetary disk in this system.

Although the planet in the HD 188753 system2 presents a conundrum to theorists, there might be an easy way out: abandon the make-do-and-mend migration theory to Occam's razor, and accept that not all planet-forming nebulae are similar to the solar nebula. Large and small protoplanetary nebulae of the same mass might differ only in their total angular momentum, such that in smaller nebulae more mass is closer in — nursing young giants.

The giant planets that orbit other stars exist in a diversity of systems and most are unlike the system of planets found around our own Sun. In our view, the diversity of planetary systems probably reflects a diversity of protoplanetary nebulae, and wherever sufficient mass is available, planets, even giant ones, may form. The neglected majority of double stars could thus fill the Galaxy with planets.

Sometimes we stumble across a planet in a place we did not expect to find one. Such a discovery can provide important steps towards understanding just how our Solar System, and others, formed. The discovery of the HD 188753 planet may be one of those steps.


  1. 1

    Mayor, M. & Queloz, D. Nature 378, 355–359 (1995).

  2. 2

    Konacki, M. Nature 436, 230–233 (2005).

  3. 3

    Wuchterl, G. Bull. Am. Astron. Soc. 28, 1108 (1996).

  4. 4

    Wuchterl, G. in Science with the VLT Interferometer (ed. Paresce, F.) 64–71 (Springer, Berlin, 1997).

  5. 5

    Hatzes, A. P. et al. Astrophys. J. 599, 1383–1394 (2003).

  6. 6

    Mugrauer, M. & Neuhäuser, R. Mon. Not. R. Astron. Soc. (in the press); preprint at

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.