In the Star Wars saga, the Skywalker clan has its roots on Tatooine - a desert-covered planet revolving around two suns. A theoretical investigation has explored the likelihood for worlds like this to exist.

And it looks like the nearest Tatooine may be closer than a galaxy far, far away.

That's because more than half of the stars in our galaxy have a stellar companion. And yet, of the 130 or so currently known exoplanets (none of which are Earth-like), only about 20 of them are around so-called binaries. The percentage may grow higher. The current ratio is affected by an observational bias: planet hunters tend to avoid binaries because the star-star interactions can hide the planet signatures.

Scientists discussed the issue earlier this month at a gathering of exoplanet hunters at the Space Telescope Science Institute in Baltimore.

Bad to good

"A few years ago, it was thought that [binaries] were a very bad site to search for planets," says Michel Mayor of the Observatoire de Geneve. "So we carefully eliminated all binary stars from our sample."

But planets may be just as likely around binaries as around single stars. Recent numerical simulations have shown that Earth-like planets, known as terrestrials, form readily in double star systems.

"The most significant thing we found is that terrestrial planets around certain close and wide binaries can look similar to planets around a single star," said Jack Lissauer of the NASA Ames Research Center.

Wide binaries are those in which the two stars are separated by several astronomical units (AU), which is the distance between the Sun and the Earth. Planets could orbit around one of the pair, or each separately. So far, all the stellar binaries with exoplanets are wide binaries.

But close binaries, where the stars are less than about an AU apart, can potentially have planets in orbit around both stars - presumably as is the case for Tatooine. These planets, however, will be much harder to detect.

Lissauer and his collaborators have explored what binary star systems are favorable for planet formation. These limits could be useful in future planet searches.


The researchers used computer models that start with 14 large planet "embryos" and 140 smaller planetesimals in orbit around one star or both stars of a binary. Evolution of this material is influenced by gravity and collisions. The models are followed for the equivalent of about one billion years.

"All of our simulations have been able to form terrestrial planets," said Ames researcher Elisa Quintana, who presented a poster on these results at the symposium.

But not all of the models produce planets around 1 AU, which is often thought to be the most likely habitable zone for life. Quintana varied how the two stars revolve around each other to see what configurations allowed for stable planet orbits inside 1 AU.

For wide binaries, Earth-like planets formed as long as the two stars came no closer than 7 AU. Quintana said that about 50 percent of known binaries meet this constraint.

The research group also ran simulations that mimicked Alpha Centauri - the nearest binary system to Earth, where the closest the two stars come is about 11 AU. The secondary star apparently acts like Jupiter does in our solar system - limiting how far out planets can form. The results showed several terrestrial planets were possible around either of the stars.

Planets have not yet been seen in the Alpha Centauri system, but small mass planets cannot yet be ruled out.

For close binaries, if the two stars are about 0.1 AU apart, the planets that form are indistinguishable from those seen in simulations with only one star. But as this separation increases, or the orbit becomes highly non-circular, it is harder for Earth-like planets to exist.

"Perturbations from the stellar motions can eject matter into space or into one of the stars," Quintana said.

The simulation results can inform observers which binaries might be better targets for their telescopes.

Observational hurdles

That said, it will not be easy to see a planet around a binary, especially those where the stars are close to each other. Most planets have been found by the radial velocity technique that searches for Doppler shifts in the light spectra of stars.

"Finding the wobble from a planet in a stellar spectrum is hard enough without having another star orbiting the one you are looking at," Quintana said.

An alternative way of detecting planets is to look for the eclipse, or transit, of a planet in front of a star. Lissauer said that transiting searches could potentially discover planets around close binaries, but "there are complications."

For one thing, two stars are putting out light, so the eclipse of one star is less noticeable. Also, the transit searches look for certain patterns of dimming and brightening of a star. If there are two stars in a tight orbit, this pattern will be different, so special algorithms will be needed.

But there are situations where a binary could provide an advantage for detecting planets. If the two stars eclipse each other, a planet could change when this eclipse happens.

"If the timing of the eclipses is not periodic, maybe a planet is to blame," Lissauer said.

Besides the possibility of transit timing, eclipsing binaries make good targets because planets - if they exist - will likely orbit in the same plane as the two stars - meaning they will also eclipse the stars at some point.

Which of these detection methods will be most likely to find the first Tatooine-like planet? Lissauer is unwilling to say.

"Predictions are tricky because they deal with the future," he joked.