Posted by In 1977 audiences watched Luke Skywalker stare at a double sunset on Tatooine, and for the next thirty-odd years most astronomers would have told you that a real solar system with two suns was possible in principle but probably rare and hard to prove. Then, in September 2011, a team using NASA’s Kepler telescope announced Kepler-16b, a cold gas giant roughly the mass of Saturn circling a pair of small stars every 229 days. Press releases called it Tatooine within hours. The name stuck, even though the planet is nothing like a desert world.
Since then more than a dozen such planets have turned up. They are called circumbinary planets, because they orbit both stars of a binary rather than one of them, and they are a useful test of how robust planet formation really is.
Two stars are normal; one is the oddity
It is easy to forget how common binaries are. Roughly half of Sun-like stars have at least one stellar companion, and the fraction is higher for more massive stars. Our Sun, a loner, is not the standard model. That means any honest estimate of how many planets exist in the galaxy has to deal with what happens when two stars share the same disk of raw material.
There are two basic arrangements. In wide binaries, such as Alpha Centauri A and B, a planet can orbit just one star while the other circles far away; this is called an S-type orbit. In tight binaries, where the stars are separated by a fraction of the Earth–Sun distance, a planet can only survive by looping around both, a P-type or circumbinary orbit. Kepler-16b is the second type.
Staying in orbit without being thrown out
The gravitational field near two stars is messy. A planet that ventures too close gets tugged first one way and then another, and simulations show it is likely to be flung out of the system altogether. A widely used rule of thumb, from simulations published in 1999, says circumbinary orbits are only stable beyond about two to four times the distance between the stars, depending on their masses and how elongated their mutual orbit is.
Here is the curious part. Many of the circumbinary planets Kepler found sit only just outside that line. Kepler-34b, Kepler-35b and others hug the edge of the stable zone. Planets probably did not form there, because the turbulence close to a binary should make it hard for dust to clump. The leading explanation is that they formed further out and migrated inward through the disk until they stalled near the inner edge, where the binary had cleared a cavity. It is a neat idea. Whether it is the whole story is not yet clear.
Finding a planet when the stars keep moving
For a single star, a transiting planet produces dips that repeat like clockwork. In a binary, the stars themselves are swinging around each other, so the planet crosses a moving target. Transits arrive early or late, change in depth and duration, and sometimes a planet crosses one star but not the other on a given pass. That irregularity is actually what confirms the planet, because nothing else produces that exact pattern, but it also makes these systems hard to spot with simple automated searches.
One of the best-known detections came from a seventeen-year-old. In 2019 Wolf Cukier, a high school student interning at NASA’s Goddard Space Flight Center, was checking TESS data by eye and noticed a transit-like signal in an eclipsing binary that the software had flagged as a stellar eclipse. It turned out to be TOI-1338b, a planet a bit smaller than Saturn, announced in 2020. A second planet in the same system was later identified through radial velocity measurements.
The list has kept growing in odd directions. Kepler-47 has at least two planets around its pair of stars, one of them a Neptune-sized world near the system’s habitable zone. Kepler-1647b, announced in 2016, is a Jupiter-sized giant with a three-year orbit. In 2025 astronomers reported a candidate planet on a polar orbit around a pair of brown dwarfs, passing over the poles of the binary instead of following its plane. It still needs confirmation.
How much light, and when
A planet with two suns does not automatically get twice the warmth. Kepler-16’s stars are both smaller and dimmer than the Sun, and the planet receives less light than Earth does. Brightness depends on each star’s luminosity and distance, and because the stars orbit each other every few weeks, the total light reaching the planet rises and falls in a cycle that runs on top of the planet’s own year.
Astronomers have worked out habitable zones for these systems the same way they do for single stars, combining the light of both. The zone wobbles over time, and the closer the binary pair, the smoother the combined illumination tends to be. A thick atmosphere and deep oceans would buffer short-term swings, much as Earth’s oceans smooth out the difference between day and night. Whether a real circumbinary planet has those things is a separate question, and so far nobody has measured the atmosphere of one.
Could anything live there?
No circumbinary planet found so far is a convincing candidate for life. Almost all the known ones are giants, partly because big planets produce the biggest transit signals and are easiest to detect. Smaller, rocky circumbinary planets are almost certainly out there. They are just harder to catch.
On paper, nothing about two stars rules out biology. A rocky planet in a stable orbit inside the combined habitable zone could, in principle, hold liquid water for billions of years. Moons of circumbinary giants are another possibility worth taking seriously, since a giant like Kepler-47c could host large moons of its own, and moons have been the focus of habitability thinking in our own Solar System for decades.
There are complications. A pair of stars with different temperatures would give a mixed spectrum, so any photosynthesis-like process would face light that shifts in colour as the stars move. Seasons could be layered, combining axial tilt, the planet’s orbit and the binary’s own cycle into something far less regular than ours. Whether life could cope with that is pure speculation. Life on Earth has handled worse.
What the next decade could tell us
The most interesting test will come when astronomers find a small rocky planet around a binary and point a large telescope at it. ESA’s PLATO mission, designed to monitor bright stars for long periods, should be good at catching long-period transits, including around binaries. Precise radial velocity surveys are also starting to find circumbinary planets that do not transit at all.
For now, a solar system with two suns sits in the category of real places that we know only in outline: masses, sizes, orbital periods and a lot of careful modelling. That outline already tells us something important. Planets form in environments far messier than our own, which widens the range of places where a habitable world might be hiding.
There is more on binaries, strange orbits and the science of habitable zones on SETIworld. If double sunsets have always appealed to you, it is a good place to keep looking.
