Kronos: Devourer of Worlds Script (Patreon)

A team of scientists recently discovered a star that appears to have consumed its planets, like some sort of infanticidal titan from Greek mythology. It’s the story of HD 240430 and HD 240429, or as the authors of Oh et al 2017 have aptly named them, Kronos and Krios. It’s also what we’ll be discussing on today’s Space Time Journal Club.

Our story begins where every story begins. In the heart of a vast cloud of gas and dust drifting through interstellar space. It is here that stars are born. We’ve talked about this process before, so in short: essentially all stars form in these stellar nurseries, and they form in the many thousands before being ejected from their birth clusters to wander the galaxy. Our Sun wanders alone, and we have no idea where its sibling stars might be. 

However if the gravitational connection between a pair of stars is strong enough they might be ejected from the cluster as a binary pair. This is pretty common actually; around half of all stars are binaries. Binary stars typically have the same chemical composition as each other, having formed from the same cloud. They might be a close binary - whizzing around each other with periods as short as days or even hours, or they might be a wide binary - separated by up to a few light years, and just barely gravitationally bound.

These wide binaries are actually extremely useful tools for understanding star formation and the environment of the galaxy they live in. They’re so loosely bound that even a small gravitational nudge can break apart the pair, for example, by interacting with other massive objects.

So by counting the number of wide binaries of different separations, we can fine-tune our theoretical models of the galactic mass distribution, including the numbers of near-invisible stellar objects like black holes and neutron stars, as well the distribution as gas and dark matter. They also allow us to test to what degree stars that formed in the same cloud share a chemical signature. If we understand this chemical tagging, perhaps one day we’ll be able to locate the Sun’s lost siblings.

Wide binaries are useful, and should also be very abundant, but they’re hard to identify. They’re so, well, widely separated that it’s hard to tell if they’re actually gravitationally bound. Semyeong Oh of Princeton University and collaborators have taken up the hunt for these systems. Their primary tool is Gaia: an orbiting telescope launched by the European Space Agency in 2013, on a five year mission to measure the precise positions and distances of about a billion objects in the Milky Way. They identified plenty of stars that are close enough together that they could be gravitationally bound. They then made spectroscopic observations to measure Doppler shift velocities. That told them which pairs were actually traveling together, rather than passing each other by. That same spectroscopy also measured chemical composition 

Thats is how the team came upon HD 240430 and HD 240429 - Kronos and Krios - two stars nearly two light years apart from each about 326 light years from our solar system. They’re both G-type stars like our Sun, and both are around four billion years old. The stars velocities tell us that they're moving in lock-step together around the galaxy. That means they’re almost certainly a wide binary pair, with an orbital period of around ten thousand years. That means they should be made of the same stuff right? The researchers checked their spectrum and this is where things got weird.

Stellar spectra are thick with sharp emission and absorption features that result from electron transitions in atoms in the star’s atmosphere. The strength of these lines can tell us the relative abundance of non-hydrogen and helium elements within the star - in astrospeak, the star’s metallicity, although for an astronomer anything heavier than helium is called a metal. Yeah, sorry about that. So it turns out that Kronos has a higher metallicity than Krios. And elements with high condensation temperatures - like silicon and most actual metals - were in extreme overabundance. Low condensation temperature elements - so-called volatiles like carbon, nitrogen and oxygen  - were only slightly more abundant. We’ll get back to this, but it’s important. One volatile - Lithium - does show up in bizarrely high abundance in Kronos,  that’s also important, and we’ll also get back to it.

These stars definitely formed from the same molecular cloud. It’s incredibly unlikely that they would end up tracking each other’s motion so closely unless they were ejected from their birth cluster as a binary pair. But if they formed from the same cloud they should be made of the same stuff, right?  So why are their chemical differences much higher than ever observed for a wide binary pair?

Oh and collaborators have a story. Just like it’s mythological namesake, perhaps Kronos ate its children. Perhaps the extra elements are the vaporized remains of its own planetary system. This is a dramatic but actually plausible story. The extra elements found in Kronos are pretty much exactly what you’d expect from a star nomming on a bunch of terrestrial planets. Terrestrial or rocky planets are made of elements with high condensation temperatures. This is the first stuff to clump together in the early days of an inner solar system as things are cooling down. Volatiles are less abundant in those inner rocky planets because they condense at cooler temperatures. They remain vaporized in the inner solar system but clump together further out, giving us gas giants and comets and stuff. The researchers tested the hypothesis by throwing a bunch of Earth-like planets into a Sun-like star. Mathematically, I mean. The Princeton ethics policies are strict about planetary destruction. Oh et al calculated that 15 Earth masses of raw Earth material would produce the observed abundances very nicely. 

The other compelling bit of evidence is Kronos’ over-abundance of lithium. Lithium gets depleted in the early years of stars like the Sun - it gets dragged down to the core by convection where it’s destroyed in a fusion reaction. Kronos has an unreasonably high lithium abundance for a star of its age, while Krios has the expected amount. Even if Kronos and Krios didn’t form together, that lithium content suggests that  Kronos accreted a bunch of new material some time well after its formation - and after the formation of any planetary system. 15 Earth masses would about do the trick there as well. This is all good info to have - we now know that binary stars can have very different metallicities, and we’ll need to take that into account for future searches.

So what caused Kronos to devour it’s worlds? Well computer simulations of planet formation do show that planets can fall into their home stars. This can happen if a gas giant ends up in the inner part of the system, either by migration due to interactions between planets, or if the gas giant is perturbed by a passing star into an elliptical orbit. We talk a bit about the effects Jupiter may have had on our solar system in this episode. Future Gaia data will be used to search for rampaging outer planets around Kronos. It may also reveal other planet-eating stars, which will shed light on the whole planet formation process. We’ll be sure to keep you updated on future episodes of Space Time.