Global magnetohydrodynamic simulations of the inner regions of protoplanetary discs. II. Vertical-net-flux regime

The inner regions of protoplanetary discs, which encompass the putative habitable zone, are dynamically complex, featuring a relatively well-ionised, turbulent active zone located interior to a poorly ionised ‘dead’ zone. In this second paper, we investigate a model of the magnetohydrodynamic processes around the interface between these two regions, using five three-dimensional global magnetohydrodynamic simulations of discs threaded by a large-scale poloidal-net-flux magnetic field. We employ physically motivated profiles for Ohmic resistivity and ambipolar diffusion, alongside a simplified thermodynamic model comprising a cool disc and hot corona. Our results show that, first, the interface acts as a one-way barrier to inward transport of large-scale magnetic flux from the dead zone. This leads to magnetic flux depletion throughout most of the active zone, whereby it either advects inwards to the inner numerical boundary or accumulates just inside the interface. Second, two sources of strong variability emerge from the interface due to the difficulty of maintaining a constant, vertically integrated electrical current across distinct and evolving magnetic-field states. Third, despite the weak magnetothermal wind in the dead zone, a pressure maximum forms at the interface, leading to Rossby-wave-induced vortices. Fourth, unlike the model of Iwasaki et. al (2024), there is no ’transition zone’ devoid of magnetic flux and magnetic winds. Instead, multiple outflow zones span all disc radii reflecting the radially varying launch conditions, with an inner turbulent wind impinging upon an outer, more laminar one. Fifth, a heated corona prevents the ‘puffing up’ of poloidal-net-flux, active disc regions.

💡 Research Summary

This paper presents a suite of three‑dimensional global non‑ideal magnetohydrodynamic (MHD) simulations that explore the dynamics of the inner (< 1 au) region of protoplanetary disks when a weak vertical net magnetic flux threads the gas. Building on the zero‑net‑flux models of Roberts et al. (2025), the authors introduce a vertical‑net‑flux (VNF) regime, incorporate physically motivated Ohmic resistivity and ambipolar diffusion profiles, and adopt a two‑temperature prescription that separates a cool, dense disk from a hot corona. Five simulations are performed with the GPU‑accelerated code idefix, the fiducial run (NF‑BAZ) covering radii 0.1–100 au, and several variants that alter the location of the dead‑active zone interface, the radial domain, and the inner boundary condition.

Key findings are as follows. First, the dead‑active zone interface acts as a one‑way barrier to the inward transport of large‑scale magnetic flux from the dead zone. Consequently, magnetic flux is either advected inward to the inner numerical boundary or piles up just inside the interface, leaving the bulk of the active zone magnetically depleted. Second, maintaining a constant vertically integrated current across the interface proves difficult; two distinct sources of strong variability emerge: rapid current‑sheet reconnection events and slower, quasi‑periodic adjustments of the current density. Both are tied to the sharp contrast in non‑ideal coefficients (Rm and ΛA) between the ideal, inward‑drifting active region and the non‑ideal, outward‑drifting dead region. Third, despite a weak magnetothermal wind in the dead zone, a pressure maximum forms at the interface. This maximum triggers a Rossby‑wave instability that generates long‑lived vortices, providing a natural dust‑trapping mechanism that could aid planetesimal formation. Fourth, contrary to the recent Iwasaki et al. (2024) study, no “transition zone” devoid of magnetic flux or winds is observed. Instead, multiple outflow zones span the entire radial extent, reflecting radially varying wind‑launch conditions: an inner turbulent wind driven by the MRI‑active region impinges on an outer, more laminar wind emerging from the dead zone. Fifth, the hot corona suppresses the “puffing up” of the poloidal flux, keeping the active layer relatively thin and preventing excessive vertical expansion of the disk.

The authors compare their results with Iwasaki et al. (2024), highlighting the differences in flux transport and wind morphology, and discuss the implications for accretion rates, disk structure, and early planet formation. They conclude that the inclusion of a weak vertical net flux fundamentally alters the magnetic coupling at the dead‑active interface, leading to flux blockage, current‑sheet variability, pressure‑maxima‑driven vortices, and a layered wind architecture. Future work is suggested to incorporate Hall physics, more detailed ionisation chemistry, and synthetic observables to bridge the gap between theory and ALMA or JWST measurements.