Don’t Panic, But A Lot of Stars Seem to Eat Their Own Planets

At least one in a dozen binary stars shows signs of eating their planets

Artist Impression of a star devouring one of its planets. A planet skimming along a star’s surface

An artist's impression of a planet on the cusp of being engulfed and destroyed by its star.

International Gemini Observatory/NOIRLab/NSF/AURA/M. Garlick/M. Zamani (CC BY 4.0)

Is our solar system quotidian or quirky? It’s one of the greatest questions in astronomy, and scientists are getting a little bit of a handle on it as they examine the more than 5,500 exoplanets (and counting) discovered around other stars. Reaching an answer is, however, confounded by a rather dramatic problem: certain stars eat their own planets, which makes learning what counts as “normal” for planetary systems a little troublesome.

Now, if a new study published today in Nature is to be believed, planetary engulfment may be less of a glitch and more of a given of the cosmos. An international team of scientists looked at dozens of pairs of closely spaced co-natal stars—those born at the same time from the same cocktail of dust and gas. Both should start with essentially identical compositions. But scientists found that this wasn’t always the case: 8 percent of those pairs of twins had one star that was conspicuously belching out plumes of elements that are usually found on planets.

“The signals we see are really strong,” says Yuan-Sen Ting, an astrophysicist at the Australian National University and one of the study’s authors. The team’s finding of 8 percent is probably the lower limit; weaker, harder-to-confirm signals of planetary engulfment may be present, too.


On supporting science journalism

If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.


Astronomers have previously studied co-natal stellar twins to spot voracious stars on planet-based diets—but never so robustly. The study “ties so many things together. It’s really cool,” says Stephen Kane, an astrophysicist at the University of California, Riverside, who was not part of the research.

The authors suspect that these signals of stellar indigestion represent planetary meals that were consumed during the stars’ adulthood, well past their chaotic youth, when unsettled orbital paths should routinely send worlds plummeting sunward to fiery oblivion. This suggests that gravitational perturbations—perhaps from rogue passing stars or by migrating gas giants—can routinely launch rocky worlds into the maws of their star, even in mature planetary systems.

This insight also offers a way to make exoplanet hunting more efficient. “In the future, we could use the chemical abundance in stars to select the highest-priority targets for planet hunting,” says Lorenzo Spina, an astrophysicist at the Astronomical Observatory of Padua in Italy. “I would expect that stars that aren’t showing this signature of planetary engulfment, maybe they have a higher probability of hosting a planetary system that’s similar to our solar system.”

In the overarching quest to understand the life and death of planets, finding such smoking-gun (or burping-star) evidence for their late-stage stellar annihilation is, if not quite a coda, an elaboration on a long-running theme. For years, astronomers have puzzled over potential telescopic “tells”—stars showing strange flare-ups, odd rotations or peculiar chemistries—that hint at recent planetary snacking. Stars oddly enriched in refractory elements—iron and even harder-to-vaporize metals—are thought of as particularly suspicious because such durable matter often makes its way into planets.

But viewed in isolation, whether a star has gorged itself on a planetary buffet or if it just formed from an unusual interstellar cloud has proved difficult to tell. Twins of stars born at the same time from the same flotsam are ideal for studying this. “Just like in many sociology studies, twins are amazing because you take away all the other confounding factors,” Ting says. If one of those twins bears an uncanny elemental resemblance to its protostellar cloud but its sibling has oodles of silicon and iron, chances are the latter has nommed on a planet.

To get as large a sample size as possible, the team used the European Space Agency’s Gaia space telescope, which is especially good at tracking the voyages of billions of stars swirling through the Milky Way. Gaia is so good, in fact, that its data can be used to retrace a star’s celestial peregrinations and reveal close associations with other stars that may have since drifted light-years apart. From this, the researchers identified 91 pairs of co-natal stars, many of which were not too dissimilar from our sun.

The team then documented these stars’ composition using spectra from multiple large ground-based telescopes and identified 21 elements generally associated with planets, including carbon, oxygen, silicon, iron, nickel and zinc. “We can tell all of the subtle differences in these different elements,” says study co-author Fan Liu, an astronomer at Monash University in Australia. And in 8 percent of those pairs, the researchers spotted the grisly refractory fingerprints of star-swallowed worlds.

Numerous unanswered questions remain, such as what sorts of planets tend to be consumed and how to know with certainty whether any given star has wholly abstained from devouring members of its brood. Even so, “this work is super compelling,” says Melinda Soares-Furtado, an astrophysicist at the University of Wisconsin–Madison. “I’m excited about what we’re starting to find.”

But are 8 percent (or more!) of all stars likely to be planet-devourers? The circumstantial evidence tentatively suggests as much, says Ricardo Yarza, an astrophysicist at the University of California, Santa Cruz. But “estimating this rate is quite challenging.” Even a sample of 91 pairs of stellar twins is vanishingly tiny compared with the Milky Way’s hundreds of billions of stars—not to mention the innumerable suns of other galaxies.

Zooming back in for a moment to our own modest corner of the cosmos, what might this research reveal about our middle-aged sun? Our star has no known co-natal twin, but multiple studies of comparable stars elsewhere in the galaxy suggest the sun is not especially enriched in refractory elements. If anything, it seems to be depleted of them, making it anomalous in an alternative way. “The more you look at our solar system, the weirder it gets,” Kane says.

“We have found there is a huge diversity in planetary systems out there,” Spina says. In the context of this exotic menagerie, our own “seems a bit boring. It’s not chaotic. All our planets are well ordered.” But in light of this new study, perhaps this mundanity is worth celebrating. Earth may only exist because, unlike many other similar stars, our sun didn’t develop a penchant for planet-munching.