The diverse shapes of binary asteroid satellites born from sub-escape-velocity moonlet mergers

TL;DR

The paper investigates how sub-escape-velocity moonlet mergers in circumasteroidal debris disks can produce atypically shaped binary-asteroid satellites like Dimorphos and Selam, followed by tidal evolution around a Didymos-like primary. Using the polyhedral DEM code GRAINS, the authors simulate numerous non-spherical moonlet mergers and track their post-merger deformation under tides, exploring how pre-merger geometry, debris-disk presence, and granular physics shape final morphologies. A key finding is that final shapes—classified as LAB, OAB, or SAB—depend sensitively on the initial orientation of the largest moonlet, with tidal forces causing mass loss or homogenisation in distance-dependent, discrete regimes, including OS-like Dimorphos-analogues and necked bilobates akin to Selam. The work further shows that higher-resolution SFDs and irregular grain packing alter deformation, yet bilobate outcomes persist, underscoring the necessity of non-spherical, polydisperse modeling to accurately capture post-merger evolution and its links to observed satellite shapes in the solar system.

Abstract

Recent direct observations of atypically shaped rubble-pile satellites of sub-km asteroids in form of the spherically oblate Dimorphos and bilobate Selam challenge classical binary asteroid formation theories, which only explain the predominantly elongated population. This study further explores a rubble-pile satellite formation scenario for binary asteroid systems involving debris disks by investigating how mergers between moonlets with impact velocities below the mutual escape speed (sub-escape-velocity mergers) and tidal disruptions can create atypically shaped moons. We simulated sub-escape-velocity mergers between moonlets and studied the resulting structural evolution of the formed moon in a tidal environment using the polyhedral discrete elements method N-body code GRAINS. Firstly, we find that the shapes of rubble-pile moons formed by mergers in this regime are highly dependent on the shape and initial orientation of the involved moonlets. This can be explained by the moonlets largely retaining their individual structures during the impact. Secondly, we observe that mass-loss via tidal disruption for a bilobate object occurs in discrete regimes of distance to the primary. Closer to the primary, the innermost lobe is completely stripped off, while only a small piece of it is lost further out. Due to moonlets largely retaining their shape after undergoing a sub-escape-velocity merger, it is necessary to account for their non-sphericity to accurately model satellite formation in circumasteroidal debris disks. Moreover, the reshaping of merged objects via tidal disruption and distortion can produce oblate spheroid moons such as Dimorphos and highly elongated bilobate satellites with distinct necks such as Selam.

Figures (20)

• Figure 1: Visualisation of how new initial conditions are generated by rotating the most massive moonlet counter-clockwise around its barycenter by an angle $\alpha$, here set to 45$^\circ$. On the left is the original configuration from the Wimarsson_et_al_2024 merger, with a vector $\boldsymbol{\hat{a}}_0$ parallel to the aggregate's semi-major axis. We then perform the rotation for each particle, ending up with a new aggregate with its semi-major axis parallel to the vector $\boldsymbol{\hat{a}}_1$.
• Figure 2: Evolution of semi-major axis (top), eccentricity (middle) and inclination (bottom) for the moons formed in the Wimarsson_et_al_2024 merger with (solid blue) and without a disk present (dashed orange).
• Figure 3: Shapes and orientations of each formed moon after 3 h of simulation time after a sub-escape-velocity merger between the two moonlets in Table \ref{['tab:moonlet_initial_data']}. The value $\alpha$, represents the counter-clockwise rotation relative to the initial position of moonlet A in Fig. \ref{['fig:moonlet_rotation_diagram']} around its barycenter. The particles originally belonging to moonlet A have been coloured in blue, while the corresponding particles from moonlet B are orange. The black arrow in each plot is pointing in the direction of the primary.
• Figure 4: The angle, $\gamma$, between the longest principal axis of the DEEVE of moonlet A (blue), $\boldsymbol{a}$, and the vector $\boldsymbol{r}_{AB}$, which is the positional vector between the barycenters of moonlet A and moonlet B (orange).
• Figure 5: Same as Fig. \ref{['fig:merged_moon_3h']} but after putting each body on a circular orbit with a semi-major axis of $1.1r_0$ and evolving the system for an additional 45 h (48 h in total), where $r_0$ is the distance from the primary barycenter at 3 h.