On the formation of hot Neptunes and super-Earths

Abstract

The discovery of short-period Neptune-mass objects, now including the remarkable system HD69830 (Lovis et al. 2006) with three Neptune analogues, raises difficult questions about current formation models which may require a global treatment of the protoplanetary disc. Several formation scenarios have been proposed, where most combine the canonical oligarchic picture of core accretion with type I migration (e.g. Terquem & Papaloizou 2007) and planetary atmosphere physics (e.g. Alibert et al. 2006). To date, published studies have considered only a small number of progenitors at late times. This leaves unaddressed important questions about the global viability of the models. We seek to determine whether the most natural model -- namely, taking the canonical oligarchic picture of core accretion and introducing type I migration -- can succeed in forming objects of 10 Earth masses and more in the innermost parts of the disc. This problem is investigated using both traditional semianalytic methods for modelling oligarchic growth as well as a new parallel multi-zone N-body code designed specifically for treating planetary formation problems with large dynamic range (McNeil & Nelson 2009). We find that it is extremely difficult for oligarchic tidal migration models to reproduce the observed distribution. Even under many variations of the typical parameters, we form no objects of mass greater than 8 Earth masses. By comparison, it is relatively straightforward to form icy super-Earths. We conclude that either the initial conditions of the protoplanetary discs in short-period Neptune systems were substantially different from the standard disc models we used, or there is important physics yet to be understood.