Planet embryos

The low Fe abundance in the lunar mantle suggests the Moon-forming impact happened late in Earth s accretion (Canup Asphaug 2001). It may have been the last collision with another embryo. Simulations of terrestrial-planet formation find that low-velocity, oblique impacts are common (Agnor et al. 1999), so that planets like Earth and Venus are likely to experience at least one such impact during their formation. This suggests large satellites may be a common outcome of terrestrial-planet formation. 

Energy released during impacts causes large embryos and planets to melt and differentiate, forming iron-rich cores and silicate mantles. Earth acquired most of its water before its core finished forming, possibly from the Asteroid Belt. It gained the last 1 % of its mass in the form of non-fractionated material after core formation was complete. 

The highly noncircular orbits of embryos and the long accretion timescales allowed considerable radial mixing of material over distances of 0.5-1.0 AU (Wetherill, 1994). It is likely that each of the inner planets accreted material from throughout the inner solar system, although the degree of radial mixing depends sensitively on the mass distribution of the embryos at this time (Chambers, 2001). The relative contributions from each part of the disk would have been different for... 

Energy released from impacts, together with heat from the decay of radioactive isotopes, led to differentiation in planetary embryos, once these objects became partially molten (Tonks and Melosh, 1992). Iron and siderophile elements (e.g., platinum, palladium, and gold) preferentially sank to the center to form a core, while the lighter silicates and lithophile elements formed a mantle. Differentiation was probably a continuous process rather than a single event, so that large planets like Earth accreted from embryos that were already partially or wholly differentiated. 

Attempts to model the accretion of Uranus and Neptune from planetesimals orbiting 20-30 AU from the Sun (the current locations of these planets) have met with severe difficulties. Long orbital periods in the outer solar system mean that accretion occurs very slowly. In addition, solar gravity is sufficiently weak here that gravitational interactions between planetary embryos would have ejected a substantial amount of mass from this region of the disk (Levison and Stewart, 2001). Numerical simulations show that it is unlikely that bodies larger than Earth could have accreted in situ at the locations of Uranus and Neptune, even if the nebula was substantially more massive than the minimum-mass nebula (Thommes et al., 2003). 

Chambers J. E. and Wetherill G. W. (1998) Making the terrestrial planets iV-body integrations of planetary embryos in three dimensions. Icarus 136, 304-327. 

Additional growth to form Earth-sized planets is thought to require colhsions between these planetary embryos. This is a stochastic process such that one cannot predict in any exact way the detailed growth histories for the terrestrial planets. However, with Monte Carlo simulations and more powerful computational codes the models have become quite sophisticated and yield similar and apparently robust results in terms of the kinds of timescales that must be... 

Wiechert U., Halliday A. N., Palme H., and Rumble D. (2003) Oxygen isotopes in HED meteorites and evidence for rapid mixing in planetary embryos. Earth Planet. Sci. Lett, (in press). 

Accretion of planetary embryos into planets 1027-1028 107-108 Giant impacts... 

Models for the formation of the giant planets suggest that a rocky planetary embryo of about ten Earth masses can form rapidly, within 10s years. Once this embryo is established these massive planetary embryos accumulate two Earth masses of solar nebular gas over 107 yr (Kortenkamp et al., 2001). 

Two features in this table deserve comments 1. - There is no planet corresponding to n == 5, that is, to = 2.8 AU. However, there is a belt made up of several asteroid rings approximately at this distance from the Sun. This suggests that the asteroid belt might be the embryo of a planet in formation, or the remnants of a planet that was hit... 

---