Astronomical Mysteries No.4: What in the heavens is KOI-74b?
Update: There’s actually two of these objects. See Rowe et al. (2010) for KOI-81b, which if anything is even more extreme.
It’s a good time for exoplanets. The Kepler mission has recently released a list of around 750 exoplanet candidates, many of them in what are believed to be multi-planet systems. (As a point of comparison, on Monday evening, the list of known and confirmed exoplanets was ~450 – if we assume that even as few as 2/3rds of Kepler’s ‘maybes’ checkout, then this week has seen the planet count more than doubled.) The Corot mission has also been returning many new objects. And so far, the results are interesting. It’s led to a revolution in the way people think about exoplanets. The historical model was a nice, tidy one with terrestrial planets close in around a single star and gas giants futher out, and everything on nice, tidy orbits. This model is now basically dead. Hot Jupiters are everywhere. Eccentric orbits are normal. The Universe is, in short, a glorious, untidy mess. What this means for the physics of planet formation remains very much an open question.
However, bizarre as these planets are, they are not the strangest things unearthed by the planet-hunter satellites. That crown has to go to two things: KOI-74b and KOI-81b. I call these ‘things’, you see, because I have simply no idea whether to refer to them as planets or stars, or even if they are in some freakish new third category. Their properties are wholly unprecedented.
KOI-74 is an A-type star, somewhat like Sirius. It’s about 26 times brighter than our sun and it’s about one and a half times hotter, at T=9400K. So far, so normal. The CoRoT satellite observed a transit at this star – something passed between us and the star, causing its light to dim slightly. That’s interesting, but not yet earth-shaking. Here comes the weird bit, though. CoRoT kept watching it, measuring repeat eclipse cycles (this confirms the presence of a companion). However, halfway through each cycle, the system dimmed even more. Let me rephrase that, to make the weirdness explicit: when the companion ‘planet’ passed behind the star, we receive even less light than when the planet is transiting. The system is at its brightest when we can see both ‘planet’ and the star.
Now, getting a radius for a planet in an eclipsing system is a well-established and reliable technique. And it turns out that KOI-74b is smaller than Jupiter. In fact it’s about as big as Uranus or Neptune in our solar system. However, the behaviour of the lightcurve means it has to be about 3% as bright as the Sun.
To put this in comparison, your typical red dwarf is about 250,000 km across, or about ~0.17 solar radii. And these objects struggle to reach even 2% of the Sun’s output. KOI-74b has ~0.04 soalr radii – it’s nearly 4.5 times smaller than a red dwarf, and yet it’s actually brighter.
What. The. F**k?
Now, luminous flux depends on surface area and temperature. So, we can do the blackbody maths and estimate a temperature for it, assuming that its behaviour is in any way blackbody (and given what a freak this object already is, who knows?). Anyway, if we do this, this luminosity and radius require a temperature of over 12,000 degrees Celsius. The supposed planet is a third again hotter than its sun. By any reasonable planetary standard, this should be impossible.
I’ve already said WTF once, haven’t I?
At this stage, one possibility is that it might be a white dwarf. White dwarfs are compact stellar embers. They are very small, packing about the mass of the Sun into a volume more like that of the Earth. They’re dense enough that their light undergoes a measurable gravitational redshift. White dwarfs are the ultimate destination of solar-mass stars, after they exhaust their core hydrogen and expand through the red giant phase. The white dwarf is the small, dead ember left behind when the red giant’s ferocious stellar winds dissipate its atmosphere.
So is KOI-74b a white dwarf? Short answer: no, for two reasons.
One: 74-b is too wide. White dwarfs are thousands of kilometres in radius, not tens of thousands. And there aren’t really any fudge factors here either, for structural reasons. (White dwarfs are supported by electron degeneracy pressure, something which is a side-effect of the bizarre collision of quantum mechanics and relativity. It’s to do with what happens when the ‘speed’ of electrons with quantized energies starts to approach the speed of light – the Uncertainty Principle kicks in, and things get very, very odd very quickly.)
Two: 74-b is too light. White dwarfs weigh in at maybe 0.5 to perhaps 1.2 solar masses. They can’t be heavier than 1.44 solar masses, because if they are, electron degeneracy can no longer support them against their own weight (and a supernova happens at that point). 74-b weighs in at most 0.11 solar masses, and possibly as little as 0.02. This is more like the mass range of brown dwarfs, not white dwarfs.
So, 74b is too bright to be a conventional planet. It’s much too small to be a conventional star. It’s vastly too hot to be a brown dwarf. It’s both too large and too light to be a white dwarf. This raises the question of what exactly is it? A planet-sized critical lump of uranium, glowing in the light of its own chain-reaction? A beacon left behind by aliens? A lump of anitmatter, glowing in the light of the annihilation of its sturface with the sun’s stellar wind? Have both undergone some vast planetary collision, which has heated them to thousands of degrees? An accretion disk around a mini black hole?
The honest truth is that, at the moment, we don’t know. Believe it or not, the current data probably doesn’t rule out any of the above mad science hypotheses. (Well, possibly the antimatter one, although as far as I know no-one’s pointed an X-ray observatory at this object. That could be an interesting observation, though … if it’s pumping out X-rays then that would put the cat among the pigeons!)