On the Impact Origin of Phobos and Deimos. I. Thermodynamic and Physical Aspects

Abstract

Phobos and Deimos are the two small moons of Mars. Recent works have shown that they can accrete within an impact-generated disk. However, the detailed structure and initial thermodynamic properties of the disk are poorly understood. In this paper, we perform high-resolution SPH simulations of the Martian moon-forming giant impact that can also form the Borealis basin. This giant impact heats up the disk material (around ∼ 2000 K in temperature) with an entropy increase of ∼ 1500 J K −1 kg −1 . Thus, the disk material should be mostly molten, though a tiny fraction of disk material (< 5%) would even experience vaporization. Typically, a piece of molten disk material is estimated to be meter sized due to the fragmentation regulated by their shear velocity and surface tension during the impact process. The disk materials initially have highly eccentric orbits (e ∼ 0.6 − 0.9) and successive collisions between meter-sized fragments at high impact velocity (∼ 3 − 5 km s −1 ) can grind them down to ∼ 100µm-sized particles. On the other hand, a tiny amount of vaporized disk material condenses into ∼ 0.1µm-sized grains. Thus, the building blocks of the Martian moons are expected to be a mixture of these different sized particles from meter-sized down to ∼ 100µm-sized particles and ∼ 0.1µm-sized grains. Our simulations also suggest that the building blocks of Phobos and Deimos contain both impactor and Martian materials (at least 35%), most of which come from the Martian mantle (50-150 km in depth; at least 50%). Our results will give useful information for planning a future sample return mission to Martian moons, such as JAXA's MMX (Martian Moons eXploration) mission.

A dynamical context for the origin of Phobos and Deimos

“…Simulations of the formation of possible protolunar disk (e.g. Citron, Gendra & Ida 2015;Canup & Salmon 2016;Hyodo et al 2017) generically produce moons that are too large. This can potentially be mitigated depending on where the mass the deposited relative to the Roche limit (Rosenblatt & Charnoz 2012) or if there is substantial evolution in the Moon system that leaves only a fraction of the original mass visible today (e.g.…”

“…If we postulate that the last giant impact experienced by each TMA produces a disk of debris that assembles to form a Moon (Hartman & Davis 1975;Cameron & Ward 1976;Craddock 2011;Rosenblatt & Charnoz 2012;Citron, Gendra & Ida 2015;Canup & Salmon 2016;Rosenblatt et al 2016;Hyodo et al 2017), then we can examine the degree to which the continued perturbations resulting from the close passages of other bodies will disturb such a disk. This leads to the hypothesis that the much smaller mass of the Martian moons is due to the dynamical erosion of the original reservoir.…”

“…Nevertheless, modern simulations with the best available equations of state do produce moon-forming disks (e.g. Hyodo et al 2017), but the physics of such simulations are necessarily limited. Furthermore, even if disks do form, in many cases they can be completely removed by dynamical perturbations, as we have seen in the prior section.…”

“…Ahrens & O'Keefe 1972) is comparable to the escape velocity from the surface of Mars. The most detailed calculations (Hyodo et al 2017) invoke impactors a few percent the mass of Mars, producing a debris disk whose mass is a few percent that of the impactor. This is still orders of magnitude larger than the current satellite masses, but it is suggested that much of that material fell into Mars on shorter timescales.…”

“…Ronnet et al propose that direct gas-solid condensation in a more extended, purely gaseous, disk (e.g. Rosenblatt & Charnoz 2012) could yield small enough grains to match the observations, but such extended disks are not favoured by the simulations of impacts (Citron et al 2015, Hyodo et al 2017). This has led to revised scenarios in which the remaining satellites are formed from the outer edge of the original disk and were shepherded outwards by larger moons that subsequently migrated inwards due to tides and were disrupted (Rosenblatt et al 2016;Hesselbrock & Minton 2016).…”