Planets form within circumstellar disks originating from the gravitational collapse of molecular clouds. There exist two primary pathways for planet formation: one involves the fragmentation of young, gravitationally unstable disks, while the other centers on the gradual accumulation of dust particles, resulting in the formation of increasingly massive objects (planetesimals, planetary embryos, terrestrial planets, and giant planets). I studied the critical physical processes in young discs that led to the early formation of Neptunes and Jupiter and set the stage for later-generation planet formation. I also investigated the formation of terrestrial planets in the solar system, highlighting a layered Earth at the end of its accretion or the beginning of its 4.5 billion years of geophysical evolution.
See my Google Scholar page for more details.
Top-down planet formation
The disk instability model has been proposed to form giant planets for a long time. Still, fundamental open questions remain, e.g., the disk cooling efficiency and the effects of magnetic fields. We employed a novel meshless finite mass method (Deng et al. 2019) to conduct the first global simulations that resolve the fine MHD turbulence interacting with spiral density waves in early massive disks (Deng, Mayer & Latter 2020, see animation). The spirals become flocculent, and the strong fields powered by spirals can keep fragments as light as Neptune from growing overmassive or disrupted (Deng, Mayer & Helled 2021). We thus proposed a top-down formation model for the most abundant intermediate-mass planets. We are working towards radiative magnetohydrodynamic simulations of young disks.
Level the stage
A circumstellar disk may accrete misaligned infall and thus become warped. However, we found that both spiral density waves in young disks and hydrodynamic instability in evolved disks are efficient in damping moderate warps. Strongly warped disks may break into connected or distinct parts depending on the disk thermodynamics. The evolution of broken disks and the interaction of disks and misaligned companions are poorly understood.
Terrestrial planets and giant impacts
The formation of terrestrial planets concludes with giant impacts. The Moon likely formed from the debris disk of the last giant impact on proto-Earth. To characterize the post-impact Earth, we employed a novel Lagrangian hydrodynamic method (Deng et al. 2019a), better at capturing turbulence and mixing than the traditional smoothed particle hydrodynamics method. We identified the global layering of post-impact Earth in both composition and thermal state due to the Moon-forming impact. It sets a heterogeneous initial condition for Earth's geophysical evolution (see the 4K animation). The lower mantle may remain unmixed and manifest as reflectors observed in the present-day middle mantle (Deng et al. 2019b). Some impactor materials, luckily sunk to the low mantle, may reorganize themselves as the continent-size anomaly in the deep mantle, i.e., LLSVP (Yuan et al., 2023). The broad implications of a layered early Earth is yet to be studied.
Based on N-body simulations of terrestrial planet formation (Fang & Deng 2020), we propose tidal mantle striping for proto-Merucy to explain its present-day iron excess. This alleviates the tension between high-temperature giant impacts and volatiles on Mercury's surface (Deng 2020).
Modern pebble accretion models with coupled dust evolution and planetesimal accretion are under construction for terrestrial planet formation.