Explaining mercury via a single giant impact is highly unlikely

Abstract

The classical scenario of terrestrial planet formation is characterized by a phase of giant impacts among Moon-to-Mars mass planetary embryos. While the classic model and its adaptations have produced adequate analogues of the outer three terrestrial planets, Mercury's origin remains elusive. Mercury's high-core mass fraction compared to the Earth's is particularly outstanding. Among collisional hypotheses, this feature has been long interpreted as the outcome of an energetic giant impact among two massive protoplanets. Here, we revisit the classical scenario of terrestrial planet formation with focus on the outcome of giant impacts. We have performed a large number of N-body simulations considering different initial distributions of planetary embryos and planetesimals. Our simulations tested the effects of different giant planet configurations, from virtually circular to very eccentric configurations. We compare the giant impacts produced in our simulations with those that are more likely to account for the formation of Mercury and the Moon according to smoothed hydrodynamic simulations. Impact events that could lead to Moon's formation are observed in all our simulations with up to similar to 20 per cent of all giant impacts, consistent with the range of the expected Moon-forming event conditions. On the other hand, Mercury-forming events via a single giant impact are extremely rare, accounting for less than similar to 1 per cent of all giant impacts. Our results suggest that producing Mercury as a remnant of a single giant impact that strips out the mantle of a differentiated planetary object with Earth-like iron-silicate ratio is challenging and alternative scenarios may be required (e.g. multiple collisions).