Composition of Transiting and Transiting-only super-Earths

The relatively recent detections of the first three transiting super-Earths mark the beginning of a subfield within exoplanets that is both fruitful and challenging. The first step into characterizing these objects is to infer their composition given the degenerate character of the problem. The calculations show that Kepler-10b has a composition between an Earth-like and a Mercury-like (enriched in iron) composition. In contrast, GJ 1214b is too large to be solid, and has to have a volatile envelope. Lastly, while three of the four reported mass estimates of CoRoT-7b allow for a rocky composition, one forbids it and can only be reconciled with significant amounts of water vapor. In addition to these three transiting low-mass planets, there are now more than one thousand Kepler planets with only measured radius. Even without a mass measurement (“transiting-only”) it is still possible to place constraints on the amount of volatile content of the highly-irradiated planets, as their envelopes, if present, are flared. Using Kepler-9d as an example, we estimate its water vapor, or hydrogen and helium content to be less than 50% or 0.1% by mass respectively.

💡 Research Summary

The paper presents a comprehensive study of the internal composition of the first transiting super‑Earths—Kepler‑10b, GJ 1214b, and CoRoT‑7b—and extends the analysis to a “transiting‑only” case, Kepler‑9d, for which only a radius is known. The authors combine a solid‑interior model (Valencia et al.) that reproduces the core‑mantle structure of terrestrial and icy bodies with the CEPAM gaseous envelope model (Guillot & Morel) that solves the temperature‑pressure profile of H₂O‑ and H‑He‑rich atmospheres. At the solid‑gas interface, continuity of pressure and mass is enforced, allowing a self‑consistent mass‑radius‑composition relationship for any assumed bulk mass and radius.

Because all four planets orbit extremely close to their host stars (periods < 2 days for the three transiting super‑Earths and 1.6 days for Kepler‑9d), they receive intense stellar irradiation. The authors therefore incorporate a simple hydrodynamic escape model to estimate the lifetime of volatile envelopes. Water vapor is found to escape on ~1 Gyr timescales, whereas hydrogen–helium envelopes are lost within a few million years under the same irradiation conditions. This stark difference provides a powerful constraint: any H‑He present today must be a minute fraction of the planet’s mass, and even water vapor can survive only if the planet is relatively young or the envelope mass is modest.

Kepler‑10b – With a well‑determined mass of 4.56 ± 1.17 M⊕ and radius 1.416 ± 0.035 R⊕, its bulk density (~8 g cm⁻³) places it between Earth‑like (33 % iron core) and Mercury‑like (≈ 63 % iron) compositions. The escape calculations rule out any H‑He and limit water vapor to < 10 % by mass, confirming that Kepler‑10b is essentially a bare rocky planet.

CoRoT‑7b – The mass is highly uncertain. High‑mass estimates (≈ 7–8 M⊕) are compatible with a purely rocky body ranging from Earth‑like to Mercury‑like. The low‑mass estimate (1–4 M⊕) would require a substantial volatile layer; the authors show that up to ~40 % water vapor could be retained, while H‑He would be negligible. Because the estimated escape time for water (~1 Gyr) is comparable to the system’s age (~8 Gyr), a modest water envelope remains plausible for the low‑mass solution.

GJ 1214b – Its radius (2.68 R⊕) far exceeds that of any solid planet of comparable mass (6.55 M⊕), demanding a thick gaseous envelope. Using a 1000 K, 10‑bar boundary condition, the authors find that a pure water‑vapor envelope could explain the radius, but formation constraints (dust‑to‑gas ratios in protoplanetary disks) suggest a more realistic composition of an Earth‑like core capped by up to ~8 % H‑He by mass. The paper discusses conflicting transmission‑spectroscopy results (Croll 2010 vs. Bean et al. 2010) and proposes hazes as a possible reconciliation.

Kepler‑9d (transiting‑only) – Only the radius (1.64 R⊕) is measured; radial‑velocity non‑detections set an upper mass limit of ~15–20 M⊕. Because any H‑He envelope would be highly inflated and rapidly lost, the authors demonstrate that for masses below ~5 M⊕ the H‑He mass fraction must be < 0.01 %. Water vapor can also be ruled out at > 50 % for masses below ~2 M⊕. Consequently, a purely rocky Kepler‑9d would need a mass ≥ 3 M⊕, with higher iron fractions pushing the allowed mass up to ~30 M⊕.

The study emphasizes that, despite the intrinsic mass‑radius degeneracy, the combination of precise radius measurements, plausible interior models, and robust atmospheric escape estimates can place stringent limits on volatile content. This methodology is especially valuable for the growing population of short‑period super‑Earths discovered by Kepler, many of which will remain “transiting‑only” for years. The authors conclude that distinguishing truly rocky worlds from those harboring even modest envelopes is essential for interpreting the composition distribution of super‑Earths and for guiding future observational strategies aimed at probing their atmospheres and potential habitability.