A widely accepted theory among the scientists says that a giant collision is responsible for the high density of Mercury. For that reason, we accomplished three-dimensional simulations of giant impacts on proto Mercury followed by an approximate treatment of hydrodynamics of the ejected matter, whose fate is of key importance in determining the final size and differentiation of the post impact planet. Depending on the initial conditions (e.g. impact geometry, impact velocity and size of the projectile), we found out that most of the ejecta are subject either to vapor to liquid or to liquid to vapor phase transition, which results in formation of centimeter sized condensates and melt droplets. These small bodies, being subject to Poynting-Robertson effect, disappear into the Sun on astronomically very short time scale and cannot be accumulated by the destroyed planet.

In general, it appears that not only one specific giant impact but rather various randomly selected giant impacts may deplete the proto Mercury silicate mantle and yield a planet with dimensions and structure approximately equal to present day Mercury. We conclude that the strange density of Mercury is, with the utmost probability, a product of a stochastic,


Spacecraft missions (e.g. the NEAR flyby at Mathilde 253) and ground based optical and radar observations have provided an increasing evidence that many or even most asteroids are porous. Also comets are thought to have porous structures.  Furthermore, porosity may play an important role in the formation of planetesimals because collisions between porous objects can be highly inelastic and hence, porosity could act as a sticking mechanism between the growing bodies. So we need to understand how porous structures respond to impact.


We have performed (using our "smooth particle hydrodynamics" code) three dimensional simulations of giant impacts onto young Jupiter with the aim of exploring both the effects on subsequent evolution of post-impact gas planets and the possibility of direct detection of such events. We have shown that off-axis collisions tend to leave brighter-planets than head-on collisions due to better distribution of impact energy in the atmosphere. The former are then also easier to detect since the collision induced electromagnetic signals last on a longer time scale.

Four snapshots of the head-on collision covering about two hours of the real time history (clockwise ordered). The slices are taken in the plane of collision. The projectile consists of an iron core (red) and a silicate mantle (orange), which is partially peeled off due to mechanical ablation (right column). The last snapshot shows (lower left) the propagation of the shock wave through the atmosphere of the target.