The regular satellites found around Neptune ($approx 17~M_{Earth}$) and Uranus ($approx 14.5~M_{Earth}$) suggest that past gaseous circumplanetary disks may have co-existed with solids around sub-Neptune-mass protoplanets ($< 17~M_{Earth}$). These di
sks have been shown to be cool, optically thin, quiescent, with low surface density and low viscosity. Numerical studies of the formation are difficult and technically challenging. As an introductory attempt, three-dimensional global simulations are performed to explore the formation of circumplanetary disks around sub-Neptune-mass protoplanets embedded within an isothermal protoplanetary disk at the inviscid limit of the fluid in the absence of self-gravity. Under such conditions, a sub-Neptune-mass protoplanet can reasonably have a rotationally supported circumplanetary disk. The size of the circumplanetary disk is found to be roughly one-tenth of the corresponding Hill radius, which is consistent with the orbital radii of irregular satellites found for Uranus. The protoplanetary gas accretes onto the circumplanetary disk vertically from high altitude and returns to the protoplanetary disk again near the midplane. This implies an open system in which the circumplanetary disk constantly exchanges angular momentum and material with its surrounding prenatal protoplanetary gas.
Stars form in dense cores of molecular clouds that are observed to be significantly magnetized. In the simplest case of a laminar (non-turbulent) core with the magnetic field aligned with the rotation axis, both analytic considerations and numerical
simulations have shown that the formation of a large, $10^2au$-scale, rotationally supported protostellar disk is suppressed by magnetic braking in the ideal MHD limit for a realistic level of core magnetization. This theoretical difficulty in forming protostellar disks is termed magnetic braking catastrophe. A possible resolution to this problem, proposed by citeauthor{HennebelleCiardi2009} and citeauthor{Joos+2012}, is that misalignment between the magnetic field and rotation axis may weaken the magnetic braking enough to enable disk formation. We evaluate this possibility quantitatively through numerical simulations. We confirm the basic result of citeauthor{Joos+2012} that the misalignment is indeed conducive to disk formation. In relatively weakly magnetized cores with dimensionless mass-to-flux ratio $gtrsim 5$, it enabled the formation of rotationally supported disks that would otherwise be suppressed if the magnetic field and rotation axis are aligned. For more strongly magnetized cores, disk formation remains suppressed, however, even for the maximum tilt angle of $90degree$. If dense cores are as strongly magnetized as indicated by OH Zeeman observations (with a mean dimensionless mass-to-flux ratio $sim 2$), it would be difficult for the misalignment alone to enable disk formation in the majority of them. We conclude that, while beneficial to disk formation, especially for the relatively weak field case, the misalignment does not completely solve the problem of catastrophic magnetic braking in general.
