An outline of the paper: The atmospheric chemistry of GJ 1214 b: photochemistry and clouds by Miller-Ricci-Kempton et al.
GJ 1214 b is at present the only intermediate-mass exoplanet (between Earth and Neptune) accessible to atmosphere observations (see post on the transmission spectrum and atmospheric circulation of this planet). Measurements with the HST and Spitzer space telescopes have shown that the transmission spectrum is broadly flat from the near- to mid-infrared, to a level that seems to exclude the molecular features expected for a cloud-free hydrogen-rich atmosphere. This is interpreted as indicating either a water-vapour atmosphere, or the presence of clouds or thick hazes in a hydrogen atmosphere. There is one discrepant data point though in the transmission spectrum data: Croll et al. (2011) find a larger effective radius in the K band (2.2 microns).
The paper by Miller-Ricci et al. study the possibility of clouds and hazes in the atmosphere of GJ 1214 b, assuming a hydrogen atmosphere. They consider two types of clouds/hazes:
(1) condensation clouds, i.e. clouds formed as the temperature in the atmosphere drops below the condensation points of one of its constituents,
(2) photochemical hazes, i.e. aerosols formed by the accumulation of products of non-equilibrium chemistry in the upper atmosphere, provoked by the incoming stellar radiation.
For the first type of clouds, the authors find only two candidate compounds, KCl and ZnS. However, they point out that these molecules should be too rare in the atmosphere of GJ 1214 b to provide enough material to have an effect on the transmission spectrum. For photochemical hazes, they find that methane photochemistry should produce compounds such as C2H2, C2H4, C2H6, that can in turn react to form longer hydrocarbons (not included in their calculation) and accumulate as aerosols. This is what forms the haze in the atmosphere of Titan. They find, however, that while such a photochemical haze can explain the flat spectrum in the near-infrared, the molecular bands of methane should still be observed in the mid-infrared, in contrast to the actual observations.
To resolve the quandary, Miller-Ricci-Kempton et al. propose an interesting solution: if the photochemical cross-section of methane is much higher than predicted by the models, then most methane could be photolysed in the upper atmosphere, explaining the lack of methane features in the transmission spectrum. This would also explain the anomalous K-band measurement, since there is strong absorption from water vapour at these wavelengths.
Thus we are left with two very different possibilities, each with a major problem:
(1) the planet has a water-vapour atmosphere. The transmission spectrum is featureless throughout, because of the small scale height. In that case the K-band measurement would be incorrect. The problem, though, is that without a hydrogen envelope the planet would have to have a very low density to account for the observed radius, which would imply an anomalous absence of rocks and iron in its interior.
(2) the planet has a hydrogen envelope, and the transmission spectrum is sculpted by extremely efficient methane photochemistry, producing a haze opaque in the visible and depleting methane above the infrared photosphere.
What kind of measurements could settle the issue? Obviously, a re-measurement of the K-band radius would be useful. Measuring the transmission spectrum in the visible could also lift the metaphorical haze, as a water atmosphere would show no feature, whereas a photochemical haze may show increased absorption in the blue.
Feature Image: Zinc Sulfide powder, found here.