Starless. Astronomers have the best evidence yet for free-floating planets.

NASA/JPL-Caltech/R. Hurt

'Homeless' Planets May Be Common in Our Galaxy

Our galaxy could be teeming with "homeless" planets, wandering the cosmos far from the solar systems of their birth, astronomers have found. The study could help clear up a long-running debate of whether free-floating planets really exist, and how common they are.

"The results are convincing enough that I suspect this paper will be cited for years to come as the best evidence of free-floating planets," says Dimitri Veras, an astrophysicist at the University of Cambridge in the United Kingdom, who was not involved with the study.

Over the past 2 decades, astronomers have identified more than 500 planetlike objects outside of our solar system. Most of these "exoplanets" orbit stars. The few that don't could be either free-floating planets or stars themselves; astronomers aren't sure, because their mass is so uncertain. Anything less massive than about 13 Jupiters is generally considered a planet, while anything between about 13 and 80 Jupiters is a small star known as a brown dwarf.

Astrophysicist Takahiro Sumi of Osaka University in Japan and colleagues—who form the Microlensing Observations in Astrophysics (MOA) and the Optical Gravitational Lensing Experiment (OGLE) collaborations—now appear to have figured out what is what. In a paper published online today in Nature, the researchers list 10 objects in our galaxy that are very likely to be free-floating planets. What's more, they claim that in our galaxy, free-floaters are probably so populous that they outnumber stars.

Most exoplanets have been discovered using one of two techniques: either seeing a star blink as a planet passes in front of it, or seeing a star wobble because of the shifting pull of a planet's gravity. Searching for planets that have no stars, however, requires a different approach: microlensing. In this technique, astronomers use a planet's gravity to bend light like a magnifying lens. As the planet passes in front of a distant star—one it isn't orbiting—it gives itself away by magnifying the star's light. Generally speaking, the shorter the magnifying time as the planet crosses the star, the smaller the planet.

Over 2 years, Sumi and others in the MOA collaboration monitored 50 million stars in our galaxy using the 1.8 meter MOA-II telescope at New Zealand's Mount John Observatory and the 1.3 meter Warsaw University Telescope at Chile's Las Campanas Observatory. They found 474 incidents of microlensing, just 10 of which were brief enough to be planets of around Jupiter's size. For each of those 10 planets, the researchers couldn't find any trace of a parent star within 1.5 billion kilometers—about the distance Saturn orbits around our sun. Although planets do orbit at greater distances, it is exceedingly rare for Jupiter-sized ones, and so the 10 planets were very likely to be free-floaters. The data were backed up by the OGLE collaboration.

The other 464 microlensing events were due to bigger objects--live stars, dead stars, and brown dwarfs. But because longer microlensing events like these are easier to spot, they skew the statistics. Taking that bias into account, the researchers estimated that there are nearly two free-floaters for every star in our galaxy.

"It is not surprising that there are such free-floating planets, as people have been expecting [they] exist," says Sumi. "But it is surprising that they are so common."

Perhaps the biggest question arising from the discovery is where free-floaters come from. One option proposed by astronomers was that they formed in the same way as stars—by using gravity to suck up nearby material. Yet that process is unlikely to have formed so many small objects, says Sumi. Instead, he and his colleagues think it's more likely that the free-floaters started out in planetary systems but were slingshotted away during a chaotic orbit. The question of formation is an important one, not least because—according to some—life on Earth could have originated from a free-floater that crashed into our solar system billions of years ago.

"These results suggest that violent, dynamical events are quite common in the history of planetary systems," says Sascha Quanz, an astronomer at the Swiss Federal Institute of Technology in Zurich, who was not involved with the research. "So, forming planets is one thing, but keeping them is another."