Many of the planets discovered by the Kepler satellite are close orbiting Super-Earths or Mini-Neptunes. Such objects exhibit a wide spread of densities for similar masses. One possible explanation for this density spread is giant collisions stripping planets of their atmospheres. In this paper we present the results from a series of smoothed particle hydrodynamics (SPH) simulations of head-on collisions of planets with significant atmospheres and bare projectiles without atmospheres. Collisions between planets can have sufficient energy to remove substantial fractions of the mass from the target planet. We find the fraction of mass lost splits into two regimes -- at low impact energies only the outer layers are ejected corresponding to atmosphere dominated loss, at higher energies material deeper in the potential is excavated resulting in significant core and mantle loss. Mass removal is less efficient in the atmosphere loss dominated regime compared to the core and mantle loss regime, due to the higher compressibility of atmosphere relative to core and mantle. We find roughly twenty per cent atmosphere remains at the transition between the two regimes. We find that the specific energy of this transition scales linearly with the ratio of projectile to target mass for all projectile-target mass ratios measured. The fraction of atmosphere lost is well approximated by a quadratic in terms of the ratio of specific energy and transition energy. We provide algorithms for the incorporation of our scaling law into future numerical studies.

Measures of exoplanet bulk densities indicate that small exoplanets with radius less than 3 Earth radii ($R_\oplus$) range from low-density sub-Neptunes containing volatile elements to higher density rocky planets with Earth-like or iron-rich (Mercury-like) compositions. Such astonishing diversity in observed small exoplanet compositions may be the product of different initial conditions of the planet-formation process and/or different evolutionary paths that altered the planetary properties after formation. Planet evolution may be especially affected by either photoevaporative mass loss induced by high stellar X-ray and extreme ultraviolet (XUV) flux or giant impacts. Although there is some evidence for the former, there are no unambiguous findings so far about the occurrence of giant impacts in an exoplanet system. Here, we characterize the two innermost planets of the compact and near-resonant system Kepler-107. We show that they have nearly identical radii (about $1.5-1.6~R_\oplus$), but the outer planet Kepler-107c is more than twice as dense (about $12.6~\rm g\,cm^{-3}$) as the innermost Kepler-107b (about $5.3~\rm g\,cm^{-3}$). In consequence, Kepler-107c must have a larger iron core fraction than Kepler-107b. This imbalance cannot be explained by the stellar XUV irradiation, which would conversely make the more-irradiated and less-massive planet Kepler-107b denser than Kepler-107c. Instead, the dissimilar densities are consistent with a giant impact event on Kepler-107c that would have stripped off part of its silicate mantle. This hypothesis is supported by theoretical predictions from collisional mantle stripping, which match the mass and radius of Kepler-107c.