An outline of the paper: Formation of hot Neptunes by evaporation of hot Jupiters by G. Boue et al.
Close-in gas giant planets, “hot Jupiters”, are subjected to extreme irradiation sue their close proximity to their parent stars. It is theorised that such high irradiation levels can cause the atmospheres of these hot Jupiters to be evaporated into space, and thus for them to lose mass. This mass loss can either be via direct Jeans escape (where the high temperatures cause the kinetic energy of the gas to be high enough to break out of the planet’s gravitational pull) or by vertical winds on the planets dayside. Such mass loss processes could result in an anisotropic ejection of the atmosphere (see Figure 1. below).
Boue et al. model this atmospheric ejection as a driving force that, if anisotropic, could cause hot Jupiters to migrate to wider orbits. They model the known distribution of hot Jupiters (116 with masses greater than Saturn and periods less than 10 days, with an average of 3.5 days) as evaporating bodies that migrate outwards due to their mass ejection. Interestingly the model solutions create a hot Neptune population that matches the observed distribution well, with the average period 1.5 times that of hot Jupiters. The models also predict that for outflow apertures less than 2φ=60° (i.e. being sufficiently anisotropic) the outflow direction is offset by about 25° from the sub-stellar point. This agrees well not only with atmospheric models, but also with maps of the surface temperature of the hot Jupiter HD 189733 b by Knutson et al. 2007.
There is a clear caveat to this work. That is while the hot Jupiter distribution used is well populated, the hot Neptune distribution contains 35 planets thinly spread in period below 10 days. A much larger Neptune population would be needed to fully support these models.
On the other hand, our own team has also looked at the question of how hot Neptunes are produced. Pont et al. 2011 found that a single mechanism can explain the position of hot Neptunes: tidal interactions between the planet and star. In this case, a hot Neptune would not need to begin life as a hot Jupiter and then get stripped of its atmosphere. It would simply form in the usual way that hot Neptunes form, and would migrate inwards under tidal interaction until it stops on the mass-period relation of Mazeh et al. 2005.
A map of the day-night contrast of the extrasolar planet HD 189733b : Knutson et al. 2007
Feature image: Image courtesy of NASA/CXC/M.Weiss.