Magnetically Dominated, Thermally Stable Disk Models

• Thermally stable magnetically dominated disks are defined by magnetic pressure that rivals or exceeds gas and radiation pressures to balance heating and cooling.
• Researchers employ ideal MHD equations with radiative transfer and global simulations to capture Maxwell stresses and MRI-driven turbulence in these disks.
• Enhanced angular momentum transport and energy loss via winds or coronae result in stable, thin disks that match spectral and variability features in AGN and TDEs.

A thermally stable, magnetically dominated disk is an accretion flow in which magnetic pressure—typically from toroidal (azimuthal) fields—contributes a significant fraction (often the majority) of the total pressure in the disk, thereby modifying the classical relations between heating, cooling, and instability. These disks contrast with traditional gas- or radiation-pressure supported geometrically thin disks, in that their stability, structure, and variability properties are fundamentally altered by magnetic support and the associated dynamics.

1. Theoretical Framework and Governing Equations

Thermally stable magnetically dominated disks are governed by the full set of ideal magnetohydrodynamic (MHD) equations, sometimes including general relativistic and radiative transport effects, depending on context and domain (e.g., X-ray binaries, AGN, TDEs). The central physical variable is the midplane plasma beta,

$$\beta \equiv \frac{P_{\rm gas}+P_{\rm rad}}{P_{\rm mag}},$$

where $P_{\rm mag} = B^2/8\pi$ is the magnetic pressure, and $P_{\rm gas}, P_{\rm rad}$ are gas and radiation pressures, respectively. A key regime of interest is $\beta \lesssim 1$, and often $\beta \ll 1$ in “hyper-magnetized” disks (Hopkins et al., 2023, Hopkins et al., 2023, Hopkins, 2024, Hopkins et al., 7 Feb 2025).

The equations of mass, momentum, induction, and energy (including radiative transfer) can be summarized, in conservative form, as:

• Continuity:

$$\frac{\partial \rho}{\partial t} + \nabla\cdot(\rho \mathbf{v}) = 0$$

• Momentum (including magnetic stress):

$$\frac{\partial (\rho\mathbf{v})}{\partial t} + \nabla\cdot\left[ \rho\mathbf{v}\mathbf{v} + P_{\rm tot}\mathbf{I} - \frac{\mathbf{B}\mathbf{B}}{4\pi} \right] = -\rho\nabla\Phi$$

• Induction:

$$\frac{\partial \mathbf{B}}{\partial t} = \nabla \times (\mathbf{v} \times \mathbf{B})$$

• Energy (if included):

$$\frac{\partial E}{\partial t} + \nabla\cdot\left[ (E+P_{\rm tot})\mathbf{v} - \frac{(\mathbf{B}\cdot\mathbf{v})}{4\pi}\mathbf{B} + \mathbf{F}_{\rm rad} \right] = -\rho \mathbf{v}\cdot\nabla\Phi$$

Here, $E$ includes all relevant energy densities, $$P_{\rm tot} = P_{\rm gas} + P_{\rm rad} + P_{\rm mag}$$.

Angular momentum transport is set by the Maxwell and Reynolds stresses, not a prescribed $\alpha$-viscosity. In regions where magnetic pressure dominates, $P_{\rm tot} \approx P_{\rm mag}$, and vertical support becomes

$H \sim \frac{v_A}{\Omega}$

with $v_A = B/\sqrt{4\pi \rho}$ the Alfvén speed.

These equations can be closed with an appropriate energy equation for the disk midplane, or in some idealized models, a locally isothermal equation of state with fixed thermal scale height (used in convergence tests of MRI-sustained turbulence) (Guo et al., 19 May 2025).

2. Magnetically-Induced Thermal Stabilization: Physical Mechanisms

In classical radiation-pressure dominated Shakura–Sunyaev disks ($P_{\rm rad} \gg P_{\rm gas}$, $P_{\rm mag} \ll P_{\rm rad}$), the local viscous heating $Q^+ \propto P_{\rm rad}^2$ rises more steeply with temperature than the radiative cooling $Q^- \propto P_{\rm rad} \propto T^4$, leading to the well-known thermal instability (Mishra et al., 2022, Sadowski, 2016). When magnetic pressure becomes non-negligible, this scaling flattens: $$\frac{\partial \ln Q^+}{\partial \ln P_{\rm rad}} = \frac{2 - \beta_r^{-1}}{1 - \beta_r^{-1}}$$ with $\beta_r^{-1}=P_{\rm mag}/(P_{\rm rad}+P_{\rm mag})$ (Mishra et al., 2022). Therefore, for $P_{\rm mag} \gtrsim 0.5 P_{\rm tot}$ (i.e., $\beta_r^{-1} \gtrsim 0.5$), the heating response falls below the cooling slope, and the disk is stabilized against the classical thermal runaway.

Similarly, in optically thin, cooling-dominated two-temperature disks, linear analysis shows the critical magnetic pressure fraction is $\beta_m \equiv P_{\rm mag}/P \gtrsim 0.7$ for unconditional thermal stability (Yu et al., 2015). The stabilizing influence is further enhanced by energy extraction associated with magnetically driven winds and X-ray coronae (Tajmohamadi et al., 27 Jan 2025, Zhao et al., 2023, Li et al., 2014).

The essential stabilizing features are:

• Maxwell stress replaces $\alpha$-prescriptions, with $Q^+ \propto P_{\rm mag}$.
• Magnetic pressure is less sensitive to temperature than radiation pressure.
• Additional angular momentum and energy extraction via winds/coronae increases the net cooling slope (Tajmohamadi et al., 27 Jan 2025, Zhao et al., 2023).

3. Numerical and Analytic Models: Global Simulations and Similarity Solutions

Thermally stable, magnetically dominated disks have been realized in both semi-analytic steady-state solutions and global GR(M)HD simulations (Hopkins et al., 2023, Hopkins et al., 2023, Mishra et al., 2022, Sadowski, 2016, Hopkins et al., 7 Feb 2025, Guo et al., 19 May 2025). Their main features are summarized in the table:

Model Type	Pressure Dominance	Key Diagnostic $\beta$	Typical $H/R$
Shakura–Sunyaev	$P_{\rm rad}$, $P_{\rm gas}$	$\gg 1$	$0.01-0.03$
Magnetically dominated (“Flux-frozen”)	$P_{\rm mag}$	$10^{-4} - 1$	$0.03-0.3$
Puffy disk	$P_{\rm mag} \sim P_{\rm rad}$	$\sim 1$	$0.1-1$

In the “hyper-magnetized” or “flux-frozen” regime, radial accretion brings ISM magnetic flux into the disk, amplified to $B_\phi$ fields sufficient for $P_{\rm mag} \gg P_{\rm gas}, P_{\rm rad}$ across most radii (Hopkins et al., 2023, Hopkins, 2024). Turbulence is trans-Alfvénic ($v_{\rm turb} \sim v_A$), and the turbulent Mach number $\mathcal{M}_s \gg 1$ (Hopkins et al., 7 Feb 2025).

Key simulation results include:

• Magnetically dominated disks remain thin ($H/R \sim 0.1$) but are “puffed-up” compared to pure $\alpha$-disks due to magnetic support.
• Maxwell stresses drive accretion at $\alpha \sim 0.1$.
• Surface density, temperature, $B$-field, and other quantities follow analytically predicted scaling laws (e.g., $B_\phi^2/8\pi \propto R^{-2/3}$) (Hopkins et al., 2023, Hopkins, 2024).
• Thermal timescale $t_{\rm th} \ll t_{\rm visc}$; $Q^+ \approx Q^-$ in global energy balance.
• Magnetic field topology is crucial: net vertical or quadrupolar fields are required for robust, sustained magnetic pressure; pure dipole or small-scale-loops are less effective (Mishra et al., 2022).

The global stability properties are robust over $0.01\lesssim\dot M/\dot M_{\rm Edd}\lesssim1000$ and black hole masses $10 - 10^{10}M_\odot$ (Hopkins et al., 2023, Hopkins, 2024, Winter-Granic et al., 9 Dec 2025).

4. Effects of Magnetized Winds and Coronae

Magnetically driven winds and coronae play a dual role: they extract angular momentum and mass, and carry off a fraction of the dissipated energy, further steepening the effective cooling curve in the $Q^+$-$Q^-$ plane.

• The mass-loss rate and the wind angular momentum parameter (mass-loading $\mu$) are critical in one-dimensional simulations; stability is achieved when the wind cooling term $Q_{\rm w}^-$ grows rapidly enough with $T$ (Zhao et al., 2023, Li et al., 2014).
• The coronal energy loss fraction $f_{\rm cor}$ acts analogously: as $f_{\rm cor}$ increases, the stabilizing effect is enhanced (Tajmohamadi et al., 27 Jan 2025).
• Stability boundaries in the $f_{\rm cor}$–$s$ (wind index) plane are quantified: typically, $f_{\rm cor}\gtrsim0.2$, $s\gtrsim0.2$, and $\beta_{\rm mag}\gtrsim0.3$ suffice to stabilize the hot, magnetically dominated branch (Tajmohamadi et al., 27 Jan 2025).

This mechanism provides a natural explanation for the absence of radiation-pressure-driven variability (“limit cycles”) in luminous X-ray binaries and AGN (Li et al., 2014, Zhao et al., 2023).

5. Radial and Vertical Structure

On the magnetically dominated branch:

• Vertical support: $H/R \sim v_A/v_K \sim 0.03$–$0.3$, regulated by $P_{\rm mag}$ (Hopkins et al., 2023, Hopkins, 2024).
• Surface density: $\Sigma \propto R^{-5/6}$ (Hopkins et al., 2023).
• Temperature: midplane $T$ grows inward but is regulated by rapid cooling; profiles remain shallow compared to radiative-pressure-dominated disks.
• Plasma $\beta$ never rises above unity except at the innermost radii or in the midplane after flux escapes (Guo et al., 19 May 2025).
• Turbulence is trans- to super-Alfvénic, with high Reynolds and Maxwell stresses ($\alpha \sim 0.1$).
• At high $\dot{m}$, the disk remains magnetically supported and thermally stable; radiation pressure saturates at $< P_{\rm mag}$ due to efficient vertical mixing (Hopkins et al., 7 Feb 2025).

6. Observational Implications and Applications

Thermally stable, magnetically dominated disks predict and are consistent with a variety of observed phenomena:

• Absence of thermal/viscous limit cycles in X-ray binary high/soft states even at $L\sim0.1-0.5\,L_{\rm Edd}$ (Mishra et al., 2022, Li et al., 2014, Sadowski, 2016, Winter-Granic et al., 9 Dec 2025).
• Predicts broad UV/optical plateaus in TDEs and LFBOTs, with luminosities and durations explained by stable magnetically dominated disk spreading, not by classical gas- or radiation-pressure disks (Alush et al., 5 Mar 2025, Winter-Granic et al., 9 Dec 2025).
• Provides a physical mechanism for the broad-line region, warm corona, and flared torus in AGN, all as different radial (and vertical) phases in a unified “multi-phase” magnetically dominated accretion structure (Hopkins, 2024, Hopkins et al., 2023, Hopkins et al., 7 Feb 2025).
• Jet powers via Blandford-Znajek-like mechanisms are set by the magnetic flux regulated in these disks, producing both radio-quiet and jetted TDEs within the observed range (Kaur et al., 2022).
• Magnetically dominated disks remain optically thick and can reproduce the spectral hardening seen in AGN and TDEs, with characteristic color correction factors $f_{\rm col}\sim 1.4-2$ (Kaur et al., 2022).

7. Methodological and Model-Building Considerations

Construction and diagnosis of thermally stable, magnetically dominated disks require:

• Seeding the disk with sufficient net vertical (or large-scale quadrupolar) magnetic flux to enable MRI amplification to $P_{\rm mag}/P_{\rm tot} \gtrsim 0.5$ (Mishra et al., 2022, Sadowski, 2016).
• Sufficient vertical resolution in simulations ($N_H \gtrsim 4$ cells per scale height) to capture MRI, Parker & buoyancy instabilities, and vertical escape (Guo et al., 19 May 2025).
• Accounting for flux escape, local dynamo, and boundary-driven advection for field replenishment—radial advection alone is insufficient inside the circularization radius, requiring vertical flux for sustained magnetization (Guo et al., 19 May 2025).
• Including energy loss via magnetically powered winds and/or coronae, with appropriate parameterizations for $f_{\rm wind}, f_{\rm cor}$, and wind/orbital parameters (Tajmohamadi et al., 27 Jan 2025, Zhao et al., 2023, Li et al., 2014).

Stability is robust over a wide parameter space: $P_{\rm mag}\gtrsim 0.3\,P_{\rm tot}$, $f_{\rm wind}+f_{\rm cor}\gtrsim 0.1$, and moderate to strong net flux topologies guarantee stability across physically realistic conditions for AGN, XRBs, TDEs, and beyond (Tajmohamadi et al., 27 Jan 2025, Kaur et al., 2022, Hopkins et al., 2023).

---

References:

• "The Role of Strong Magnetic Fields in Stabilizing Highly Luminous, Thin Disks" (Mishra et al., 2022)
• "Thin accretion disks are stabilized by a strong magnetic field" (Sadowski, 2016)
• "Idealized Global Models of Accretion Disks with Strong Toroidal Magnetic Fields" (Guo et al., 19 May 2025)
• "Thermal instability of thin disk in the presence of wind and corona" (Tajmohamadi et al., 27 Jan 2025)
• "Zooming In On The Multi-Phase Structure of Magnetically-Dominated Quasar Disks" (Hopkins et al., 7 Feb 2025)
• "Multi-Phase Thermal Structure & The Origin of the Broad-Line Region, Torus, and Corona in Magnetically-Dominated Accretion Disks" (Hopkins, 2024)
• "FORGE'd in FIRE II: The Formation of Magnetically-Dominated Quasar Accretion Disks from Cosmological Initial Conditions" (Hopkins et al., 2023)
• "An Analytic Model For Magnetically-Dominated Accretion Disks" (Hopkins et al., 2023)
• "Magnetically Dominated Disks in Tidal Disruption Events and Quasi-Periodic Eruptions" (Kaur et al., 2022)
• "Time-dependent global simulations of a thin accretion disc: the effects of magnetically-driven winds on thermal instability" (Zhao et al., 2023)
• "Thermal stability of thin disk with magnetically driven winds" (Li et al., 2014)
• "Thermal Stability of Magnetized, Optically Thin, Radiative Cooling-dominated Accretion Disks" (Yu et al., 2015)
• "Puffy accretion disks: sub-Eddington, optically thick, and stable" (Lančová et al., 2019)
• "Late-Time Evolution of Magnetized Disks in Tidal Disruption Events" (Alush et al., 5 Mar 2025)
• "Viscously Spreading Accretion Disks around Black Holes: Implications for TDEs, LFBOTs and other Transients" (Winter-Granic et al., 9 Dec 2025)