Formation of the Neptune system

Neptune is the outermost of the four giant planets in our solar system. The region in which Neptune orbits has longer dynamical time scales and probably had a lower surface mass density of solids than zones closer to the Sun. Both of these properties suggest that Neptune took significantly longer to accrete than did the other giant planets. "Runaway accretion" may permit growth of isolated ~ 10 M_⊕ cores in the outer solar system on time scales of 10^7 to 10^8 yr. It remains unclear, however, whether Neptune's core could have grown massive enough to trap > 1 M_⊕ of H_2 and He before the protoplanetary disk was dispersed. If core formation did occur on that time scale, models for the accretion of gas onto a growing condensible element core provide insights into nebular conditions required for the growth of Neptune-like planets. An alternative model is that ice-rich planetesimals of mass ≳ 0.1 M_⊕ trapped substantial quantities of solar nebula gas mixed with evaporated steam in high mean molecular weight atmospheres. The final stages of Neptune's growth could then have occurred after dispersal of the solar nebula, alleviating the possible time scale problem associated with the single core model. The observed similarity of the condensible element masses of four giant planets in our solar system is not an obvious consequence of current models. The decrease in H_2 and He masses from Jupiter to Saturn to Uranus and Neptune may be understood in terms of the dispersal of the solar nebula. However, several other dynamical problems remain unsolved. In particular, dynamical models of planetary growth which have been successfully applied to the terrestrial region produce far less satisfying results in the Uranus-Neptune region. The larger planetary masses and weaker influence of solar gravity cause mutual scattering to spread planetary embryos over a wide range of heliocentric distance. Several residual planets much more massive than Pluto become isolated in trans-Neptunian orbits. The fact that such planets have not been observed, even though many of them could have been detected by published surveys, implies that physical processes not yet quantitatively evaluated were quite important. The orbits of Neptune's inner satellites are nearly circular and have low inclinations with respect to the planet's equatorial plane. These properties properties imply satellite growth within a circum-Neptunian disk. We review five different models for the formation of such a disk: coaccretion, disruptive capture, giant impact, direct hydrodynamic accretion and spin-out. The high inclination of Triton's orbit suggests a capture origin.