The giant impact hypothesis remains the leading theory for lunar origin. However, current models struggle to explain the Moon's composition and isotopic similarity with Earth. Here we present a new lunar origin model. High‐energy, high‐angular‐momentum giant impacts can create a post‐impact structure that exceeds the corotation limit, which defines the hottest thermal state and angular momentum possible for a corotating body. In a typical super‐corotation‐limit body, traditional definitions of mantle, atmosphere, and disk are not appropriate, and the body forms a new type of planetary structure, named a synestia. Using simulations of cooling synestias combined with dynamic, thermodynamic, and geochemical calculations, we show that satellite formation from a synestia can produce the main features of our Moon. We find that cooling drives mixing of the structure, and condensation generates moonlets that orbit within the synestia, surrounded by tens of bars of bulk silicate Earth vapor. The moonlets and growing moon are heated by the vapor until the first major element (Si) begins to vaporize and buffer the temperature. Moonlets equilibrate with bulk silicate Earth vapor at the temperature of silicate vaporization and the pressure of the structure, establishing the lunar isotopic composition and pattern of moderately volatile elements. Eventually, the cooling synestia recedes within the lunar orbit, terminating the main stage of lunar accretion. Our model shifts the paradigm for lunar origin from specifying a certain impact scenario to achieving a Moon‐forming synestia. Giant impacts that produce potential Moon‐forming synestias were common at the end of terrestrial planet formation.

During accretion, terrestrial bodies attain a wide range of thermal and rotational states, which are accompanied by significant changes in physical structure (size, shape, pressure and temperature profile, etc.). However, variations in structure have been neglected in most studies of rocky planet formation and evolution. Here we present a new code, the Highly Eccentric Rotating Concentric U (potential) Layers Equilibrium Structure (HERCULES) code, that solves for the equilibrium structure of planets as a series of overlapping constant‐density spheroids. Using HERCULES and a smoothed particle hydrodynamics code, we show that Earth‐like bodies display a dramatic range of morphologies. For any rotating planetary body, there is a thermal limit beyond which the rotational velocity at the equator intersects the Keplerian orbital velocity. Beyond this corotation limit (CoRoL), a hot planetary body forms a structure, which we name a synestia, with a corotating inner region connected to a disk‐like outer region. By analyzing calculations of giant impacts and models of planet formation, we show that typical rocky planets are substantially vaporized multiple times during accretion. For the expected angular momentum of growing planets, a large fraction of post‐impact bodies will exceed the CoRoL and form synestias. The common occurrence of hot, rotating states during accretion has major implications for planet formation and the properties of the final planets. In particular, the structure of post‐impact bodies influences the physical processes that control accretion, core formation, and internal evolution. Synestias also lead to new mechanisms for satellite formation. Finally, the wide variety of possible structures for terrestrial bodies also expands the mass‐radius range for rocky exoplanets.

In the giant impact hypothesis for lunar origin, the Moon accreted from an equatorial circumterrestrial disk; however the current lunar orbital inclination of 5 • requires a subsequent dynamical process that is still debated [1][2][3] . In addition, the giant impact theory has been challenged by the Moon's unexpectedly Earth-like isotopic composition 4, 5 . Here, we show that tidal dissipation due to lunar obliquity was an important effect during the Moon's tidal evolution, and the past lunar inclination must have been very large, defying theoretical explanations. We present a new tidal evolution model starting with the Moon in an equatorial orbit around an initially fast-spinning, high-obliquity Earth, which is a probable outcome of giant impacts. Using numerical modeling, we show that the solar perturbations on the Moon's orbit naturally induce a large lunar inclination and remove angular momentum from the 1 arXiv:1802.03356v1 [astro-ph.EP] 9 Feb 2018Earth-Moon system. Our tidal evolution model supports recent high-angular momentum giant impact scenarios to explain the Moon's isotopic composition [6][7][8] and provides a new pathway to reach Earth's climatically favorable low obliquity.The leading theory for lunar origin is the giant impact 9, 10 , which explains the Moon's large relative size and small iron core. Here we refer to the giant impact theory in which the Earth-Moon post-impact angular momentum (AM) was the same as it is now (in agreement with classic lunar tidal evolution studies 11, 12 ) as "canonical". In the canonical giant impact model 13 , a Mars-mass body obliquely impacts the proto-Earth near the escape velocity to generate a circum-terrestrial debris disk. The angular momentum of the system is set by the impact, and the Moon accretes from the disk, which is predominantly (> 60 wt%) composed of impactor material. However, Earth and the Moon share nearly identical isotope ratios for a wide range of elements, and this isotopic signature is distinct from all other extraterrestrial materials 4, 5 . Because isotopic variations arise from multiple processes 4 , the Moon must have formed from, or equilibrated with, Earth's mantle 5,14 . Earth-Moon isotopic equilibration in the canonical model has been proposed by Pahlevan and Stevenson 15 , but has been questioned by other researchers 16 , who suggest that the large amount of mass exchange required to homogenize isotopes could lead to the collapse of the proto-lunar disk.Cuk and Stewart 6 proposed a new variant of the giant impact that is based on an initially high AM Earth-Moon system. In this model, a late erosive impact onto a fast-spinning proto-Earth produced a disk that was massive enough to form the Moon, and was composed primarily of material from Earth, potentially satisfying the isotopic observations. Canup 7 presented a variation of a high-2 AM origin in which a slow collision between two similar-mass bodies produces a fast-spinning Earth and disk with Earth-like composition. Subsequently, Lock et al. 8 have argued that a range of hig...