The early evolution of the Solar system produced a vast variety of bodies of different size, composition, and structure. On one hand, small bodies (asteroids, comets, dwarf planets) are clustered in the asteroid belt and in the outer Solar system, while rocky planets populate its inner part and gas rich planets its outer part. On the other hand, for small bodies one can distinguish between the iron-silicate asteroids with ordinary chondritic composition (like Lutetia and Vesta) and, e.g., ice-silicate bodies whose composition is related to the carbonaceous chondritic meteorites and which could contain a large amount of water ice (like the dwarf planet Ceres and the icy satellites). Furthermore, large variation of the surface properties can be observed. Vesta, the second-most massive body in the asteroid belt, displays a dry, basaltic surface produced by igneous processes indicating a differentiated structure, while Lutetia, having a smaller radius of ≈ 60 km, retained its native surface material and may still be partially differentiated. Also, among the bodies related to carbonaceous chondrites, Ceres´ surface is characterized by OH-bearing minerals, while the primary component of the Palladian surface material is a silicate that is low in iron and water. Most of the objects we observe now in the Solar system can be integrated into an accretion sequence beginning with kilometre-sized planetesimals which formed by collision and adhesion of dust grains. Larger planetesimals like Lutetia whose formation was dominated by gravitational attraction of smaller bodies follow, then the dwarf planets like Vesta which probably accreted in a runaway process dominating a swarm of smaller planetesimals, and, finally, planetary embryos like Mars, from which eventually Earth formed by a number of collisions.
Figure 1: The temperature fields (left panels) and the velocity fields (right panels) for a purely internally heated planetesimal with a radius of 260 km (approximately Vesta-sized) with temperature dependent viscosity. The upper panels show a conducting case and the lower panels a case where the interior convects at the Rayleigh number Ra equal 5·106 (DLR) .
The evolution of an asteroid begins with its accretion from the protoplanetary dust as a porous aggregate. Heating by radioactive decay combined with the self-gravity leads to the substantial compaction at rather low temperatures. Provided strong radiogenic heating by conduction, the iron phase and subsequently the silicate phase starts to melt upon reaching the respective solidus temperatures. The radioactive isotope 26Al partitions among other incompatible elements into the silicate melt. If melting is wide spread enough, the asteroid can differentiate (at least partially) via porous flow into an iron core and a silicate mantle. Differentiation is initiated by the iron melt sinking downward and silicate melt percolating towards the surface. The mantle and core regions, in which the partial melt increases above ≈ 50 %, start convecting. The asteroid cools due to convection and after the melt fractions drop below ≈ 50 % by conduction and melt heat transport to the sub-solidus temperatures. After cooling below the solidus, no further structural changes of the interior can be expected except by external influence such as impacts.
It is not easily understood, what are the driving forces of the processes, which have led to development of this structure variety early after the formation of the Solar system. As it is widely accepted, the differentiation of the planetary objects requires melting. In contrast to the terrestrial planets, one must account for significantly smaller radii, masses and gravity and significantly smaller accretion energies for planetesimals. Because of the latter and since the smaller sizes imply larger surface to volume ratios it must be concluded that the differentiation of planetesimals requires early, intense heat sources. These heat sources must have produced much more power than the decay of U, Th and K, the major heat source for the planets in addition to accretion heating. A widely accepted candidate is the radiogenic decay of the short-lived nuclide 26Al. Its widespread presence in the early solar system has been verified from the study of calcium-aluminium-rich inclusions (CAIs) and chondrules. In contrast to the long-lived radioactive elements, 26Al features several orders of magnitude higher specific power but a half-life time of only 0.717 Ma. Most recently revised estimates of the (60Fe/56Fe) concentration ratio in the early Solar system raise the possibility that the short lived nuclide 60Fe could also contribute significantly to heating and melting even in the absence of other heat sources. In particular, a redistribution of Fe with core formation makes this heat source specifically interesting for the discussion of the thermal and magnetic evolution of planetesimals.