Title: Detection of Thermal Emission from a Super-Earth
Authors: Brice-Olivier Demory, Michael Gillon, Sara Seager, Bjoern Benneke, Drake Deming and Brian Jackson
First Author’s Institution: Massachusetts Institute of Technology, Cambridge, MA
Last year, I reported on how the mystery of 55 Cancri e was resolved. This exoplanet was originally identified as a Neptune-sized planet on a short, 2.8 day orbit, but subsequent analysis by Dawson & Fabrycky indicated that it might actually be a smaller planet on an even shorter orbit, completing one orbit in just 18 hours. Independent observations by Winn et al. and Demory et al. confirmed 55 Cancri e as a super Earth planet with a radius twice Earth’s and a mass nearly 8 times Earth’s (see also Dan’s astrobite). This gives a mean density of 4.8 grams per cubic centimeter, not too much less than Earth’s own 5.5 g/cc!
In this Letter, Demory et al. observe the secondary eclipses of 55 Cnc e (when the planet passes behind the star) constituting a direct detection of this planet. One of the goals of such a measurement is to constrain the temperature of the planet and how much of the star’s light it reflects (the planet’s “albedo”). The depth of the secondary eclipse is given by the ratio of planetary to stellar light. Using our knowledge of the the star’s flux and the ratio between the stellar and planetary radii, we can extract from this the total amount of light coming from the planet. What’s the source of this planetary radiation? The planet receives a lot of energy from the star and, because it’s in a state of equilibrium, it emits exactly the same amount of energy; this radiation is detected during secondary eclipse. Knowledge of the amount of light emitted by the planet allows the authors to infer its temperature.
The authors find that 55 Cancri e is a scorching 2,360K, or 3,800 degrees Fahrenheit! That’s a thousand degrees higher than the melting point of iron, at 2,800 degrees F: this super Earth is not looking like a good vacation spot.
What does this mean for 55 Cancri e? If we were to assume that the planet absorbs all of the sunlight incident upon it and then re-radiates that energy uniformly across its surface, we would find that the planet’s equilibrium temperature should be 1,950K, less than the measured temperature by 400K. This means that some of the assumptions behind this calculation are wrong. What are the possibilities? The planet does likely reflect some of the light that hits it, but this would actually make the planet cooler, not hotter, so we have to look to other ideas. I’ll look at the other two possibilities presented in this Letter in a bit more detail: that the planet does not radiate uniformly and that these observations uncovered structure in the planet’s atmosphere.
Maybe, 55 Cnc e doesn’t radiate uniformly and we’re seeing the hot side. This is a plausible explanation: a planet orbiting so close to its host star will be tidally locked, that is, it’s always day time on one side and always night time on the other. Energy is only absorbed on the day side, so in order for such a planet to radiate uniformly, it would have to redistribute some of that energy to the night side very efficiently. Non-uniform radiation would make sense if 55 Cnc e had only a thin atmosphere (like the Moon or Mercury), which would not transport heat efficiently. However, if 55 Cnc e was rock with a minimal atmosphere, its density requires that it be made almost entirely of silicates, which doesn’t fit with our understanding of planet formation. An alternative composition is that of a small rocky core surrounded by water in an usual state (“supercritical” water), similar to the cores of the ice giants Uranus and Neptune.
Alternatively, the planet could have a substantial atmosphere and it happened that the observations, which were done using the Spitzer infrared space telescope, probe a layer that’s hotter than the equilibrium temperature. This phenomena has been observed in hot Jupiters as well as in our own Solar System, but the whopping 400K difference between the equilibrium and observed temperatures is not easy to explain in this scenario. (However, the authors point out that the uncertainty on their temperature measurement is 300K, so the situation might not be as bad as it seems. On the other hand, it could be worse!)
Where does this leave us? For one, with the first detection of light from a super Earth! It also demonstrates the importance of Spitzer in exoplanet follow-up observations and leaves us with intriguing possibilities for the nature of the still-mysterious 55 Cancri e.