von Neumann Stability Analysis of Braginskii MHD: Thermal Conduction, Viscosity, Magnetic Diffusion, and Thermoelectric Effects; and an Implementation in Athena++

arXiv:2611.05086 · astro-ph.IM · Submitted 2026-11-03 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Astrophysics. The day in astrophysics: observations, models and what the data forces.

Vera: I'm Vera.

Jocelyn: And I'm Jocelyn, here with Subrahmanyan.

Vera: We're looking at the paper: "von Neumann Stability Analysis of Braginskii MHD".

Jocelyn: Here's the gist of the paper. Athena++ is a finite-volume astrophysical magnetohydrodynamics (MHD) code that incorporates several complex physical effects, including fully anisotropic thermal conduction (parallel, perpendicular, and the Righi–Leduc effect), anisotropic viscosity (all five Braginskii viscosity tensors), the Hall effect, and multiple thermoelectric effects such as Peltier, Seebeck, Nernst, and Ettingshausen.

Vera: These effects can be modulated by the ionization fraction in a partially ionized gas.

Jocelyn: The code implements super-time-stepping support for dissipative terms while combining non-dissipative terms with the ideal MHD flux.

Vera: A comprehensive CFL condition is derived and implemented for all terms using von Neumann stability analysis.

Title and authors: Vera: So, what are we actually looking at with this paper, "von Neumann Stability Analysis of Braginskii MHD: Thermal Conduction, Viscosity, Magnetic Diffusion, and Thermoelectric Effects; and an Implementation in Athena++"? It’s essentially a deep dive into the stability of Athena++ when you add all these extra physics.

Jocelyn: The title tells us they are analyzing the stability across multiple physical processes: thermal conduction, viscosity, magnetic diffusion, and those thermoelectric effects we talked about before.

Subrahmanyan: For someone looking at cosmology or structure formation, this is important because it tells us how much numerical noise or instability we have when trying to model systems where heat and internal forces are playing a big role in the plasma.

Vera: It’s about checking if the numerical scheme blows up when you introduce these specific terms, like the anisotropic viscosity or those thermoelectric drifts. They’re using von Neumann stability analysis to derive conditions for stability across different time integration methods.

Jocelyn: The implication is that they are providing a rigorous way to understand which physical settings and which numerical methods actually work together without causing the simulation results to become totally unstable or noisy.

Subrahmanyan: It’s about ensuring that when we simulate these complex astrophysical environments, like accretion disks or stellar interiors, the code doesn't just run forever in a chaotic state because of these physics interacting poorly with the math.

The paper's summary: Vera: Okay, so what does the actual summary of this paper say about their findings? It seems they’ve mapped out how each component contributes to stability separately.

Jocelyn: They show that dissipative terms, like the viscosity and conduction terms, are stable under first-order time integration methods. That means for those specific parts, you can use simpler time stepping schemes.

Subrahmanyan: But then they point out that the non-dissipative terms—the stuff like Hall effects and thermoelectric drifts—require a much broader stability region in the complex plane, similar to what you see with some Runge-Kutta methods.

Vera: That’s a key distinction. So, for the dissipative parts, we only need them stable on the negative real axis; for everything else, we need that part of the imaginary axis to be included in our stability analysis.

Jocelyn: And they also found that the Nernst and Ettingshausen thermoelectric system is actually ill-posed unless you stabilize it with both perpendicular thermal conduction and magnetic diffusion.

Subrahmanyan: That leads them to a derived well-posedness criterion, which basically tells you exactly what physical parameters need to be present—like those specific conduction coefficients and magnetic diffusion rates—to keep the system mathematically sound.

The paper's improvements: Vera: Now, the authors aren't just doing the analysis; they’re suggesting concrete improvements for Athena++. What are these suggested upgrades they propose?

Jocelyn: First, they suggest implementing anisotropic transport by explicitly calculating fluxes at cell faces for momentum, energy, and magnetic fields. This is how you get those parallel and perpendicular terms like the Righi–Leduc effect correctly represented.

Subrahmanyan: That’s important because it moves them away from just having some bulk diffusion approximations toward a more accurate representation of how heat and momentum move across different directions in the plasma.

Vera: Second, they propose integrating the von Neumann stability analysis results directly into the code's timestep selection logic. They want to use those derived CFL conditions for each term to automatically choose the right time integration method.

Jocelyn: And they suggest a modified HLLE Riemann solver that incorporates new signal speeds for all non-dissipative terms, using the largest of the ideal and Hall phase speeds to stop those grid-scale non-linear instabilities.

Subrahmanyan: That addresses a known issue where standard solvers fail when dealing with these non-dissipative terms; they are trying to suppress those instabilities before they even get a chance to build up in the simulation.

Conclusion: Vera: So, wrapping this up, what’s the big picture here for us? This paper shows how detailed your numerical model needs to be when you include all these non-ideal physics.

Jocelyn: Essentially, it gives us a roadmap for Athena++ to ensure stability by tailoring the time integration scheme based on whether we are dealing with dissipative or non-dissipative terms.

Subrahmanyan: For the broader field, it shows us the specific physical requirements—like those conduction coefficients and diffusion rates—that are necessary to keep thermoelectric effects like Nernst from becoming mathematically ill-posed.

Vera: The authors conclude by providing specific timestep limits for every single term, based on which Runge-Kutta integrator you choose, so you know exactly what your constraints are.

Jocelyn: They also quantify the cost of strong non-dissipative terms, noting that excess numerical dissipation can reduce convergence order down to between one point one and one point nine.

Subrahmanyan: That cost is real; it means if you use too much of those complex physics, your simulation might not be as accurate as you hoped because the solver starts dumping artificial smoothing.

Vera: So, the paper "von Neumann Stability Analysis of Braginskii MHD: Thermal Conduction, Viscosity, Magnetic Diffusion, and Thermoelectric Effects; and an Implementation in Athena++" is a detailed stability study showing exactly how to handle these tricky physical effects computationally.

Jocelyn: It gives us concrete rules for choosing integrators and fixing the Riemann solver when modeling complex astrophysical plasmas.

Subrahmanyan: It sets a baseline for what’s required—the well-posedness criterion—before we can trust our simulations of those highly magnetized, partially ionized environments.

School of Physics and Astronomy, University of Leeds

astro-ph.IM

Submitted: 2026-11-03

Updated: 2026-11-05

Comments: v2 typos corrected, Submitted to the Journal of Computational Physics, 64 Pages, 26 Figures

Code: https://github.com/AlexandraAlbery/Athena-Aniso

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 92/100

The gist: Athena++ is a finite-volume astrophysical magnetohydrodynamics (MHD) code that incorporates several complex physical effects, including fully anisotropic thermal conduction (parallel…

Key concepts

Anisotropic Thermal Conduction
This describes how heat moves differently in different directions within a gas. The paper studies parallel, perpendicular, and Righi–Leduc effects, which are crucial for accurately modeling energy transport in magnetized plasmas.
Braginskii Viscosity Tensors
These tensors represent the resistance of the plasma to flow and shear. The code uses all five Braginskii viscosity tensors to capture complex momentum transport effects that occur due to magnetic fields and anisotropic pressures.
Von Neumann Stability Analysis
This mathematical technique is used to determine if a numerical scheme will remain stable when solving differential equations. It reveals that dissipative terms are stable with simple methods, while non-dissipative terms need specialized solvers like RK3.
Well-posedness Criterion
This criterion checks if the system describing thermoelectric effects is mathematically solvable and physically meaningful. The paper derives a condition involving thermal conduction and magnetic diffusion to ensure these effects are handled correctly.

Terminology

Summary

Athena++ is a finite-volume astrophysical magnetohydrodynamics (MHD) code that incorporates several complex physical effects, including fully anisotropic thermal conduction (parallel, perpendicular, and the Righi–Leduc effect), anisotropic viscosity (all five Braginskii viscosity tensors), the Hall effect, and multiple thermoelectric effects such as Peltier, Seebeck, Nernst, and Ettingshausen. p. 0

Where the paper says it

Here we present a fork of Athena++, “Athena-Aniso”. It implements fully anisotropic thermal conduction (parallel, perpendicular, and the Righi–Leduc effect); anisotropic viscosity (all five Braginskii viscosity tensors); the Hall effect; and several thermoelectric effects (Peltier, Seebeck, Nernst, and Ettingshausen).

These effects can be modulated by the ionization fraction in a partially ionized gas. p. 0
Where the paper says it

It implements fully anisotropic thermal conduction (parallel, perpendicular, and the Righi–Leduc effect); anisotropic viscosity (all five Braginskii viscosity tensors); the Hall effect; and several thermoelectric effects (Peltier, Seebeck, Nernst, and Ettingshausen). These effects can be weighted by ionisation fraction in a partially ionised gas, with the neutral part diffusing isotropically.

The code implements super-time-stepping support for dissipative terms while combining non-dissipative terms with the ideal MHD flux. p. 0
Where the paper says it

The Biermann battery is not implemented here. The dissipative terms have super-time-stepping support, while the non-dissipative terms are combined with the ideal MHD flux.

A comprehensive CFL condition is derived and implemented for all terms using von Neumann stability analysis. p. 0
Where the paper says it

The dissipative terms have super-time-stepping support, while the non-dissipative terms are combined with the ideal MHD flux. A CFL condition for all terms is derived and implemented using von Neumann stability analysis.

The analysis reveals that dissipative terms are stable under first-order time integration, whereas the non-dissipative terms necessitate a stability region that includes part of the imaginary axis, such as that of RK3. p. 0
Where the paper says it

A CFL condition for all terms is derived and implemented using von Neumann stability analysis. The dissipative terms are found to be stable under first-order time integration.

Furthermore, new signal speeds for a modified HLLE Riemann solver are derived to mitigate grid-scale non-linear instabilities arising from the non-dissipative terms, ensuring runs with grid-scale noise remain bounded where standard RK3 solvers fail. p. 0
Where the paper says it

We additionally derive new signal speeds for a modified HLLE Riemann solver. These suppress grid-scale non-linear instabilities still present in the non-dissipative terms.

The stability requirements dictate that dissipative terms only need the negative real axis for stability (forward Euler suffices), while non-dissipative terms require a broader region encompassing part of the imaginary axis. p. 1
Where the paper says it

Because the RKL stability regions are built along the negative real axis, they contain no part of the imaginary axis: like RK1 and RK2, RKL1 and RKL2 amplify every purely imaginary mode. The non-dissipative terms, whose eigenvalues lie on the imaginary axis, therefore gain nothing from STS and must be integrated with the main MHD update.

The Nernst/Ettingshausen system is found to be ill-posed unless stabilized by both perpendicular thermal conduction and magnetic diffusion, leading to a derived well-posedness criterion. p. 0
Where the paper says it

The Nernst/Ettingshausen system is found to be ill-posed unless it is stabilised by both perpendicular thermal conduction and perpendicular magnetic diffusion. We derive a well-posedness criterion for this.

Numerical tests against analytical solutions confirm the stability of each implemented effect. p. 0
Where the paper says it

We provide numerical tests of each implemented effect against derived analytical solutions. Keywords: Braginskii MHD, Anisotropic transport, von Neumann stability analysis, Thermoelectric effects, Finite volume methods ∗Corresponding author Email address: pybla@leeds.

The paper also derives specific timestep limits for every term, valid for any Runge–Kutta integrator based on its stability region along the real and imaginary axes. p. 4
Where the paper says it

These require ∆t ≤ rR max(|λ(k, u)|) (dissipative), ∆t ≤ rI max(|λ(k, u)|) (non-dissipative), (11) where rR and rI are the extents of the stability region along the negative real and imaginary axes respectively. Integrator R(Z) rR rI RK1 1 + Z 2 0 RK2, VL2 1 + Z + Z 2/2 2 0 RK3 1 + Z + Z 2/2 + Z 3/6 ≈ 2.

For instance, dissipative terms only require the negative real axis, while non-dissipative terms demand a region including part of the imaginary axis like RK3's. The cost associated with strong non-dissipative terms is excess numerical dissipation, which can reduce convergence order to between 1. p. 1
Where the paper says it

Explicit integration of dissipative terms is restricted by the timestep ∆tpar ∝ ∆x 2/D, where D is the diffusion coefficient. This becomes prohibitively restrictive at high resolution and/or when diffusion is strong.

1 and 1.9. Specific stability criteria are detailed for various physical parameters, such as thermal conduction (kappa n, kappa, kappa) and viscosity (nu, nu 1, nu 2), alongside magnetic diffusion (eta and eta AD). p. 0
Where the paper says it

von Neumann Stability Analysis of Braginskii MHD: Thermal Conduction, Viscosity, Magnetic Diffusion, and Thermoelectric Effects; and an Implementation in Athena++ Alexandra F. Alberya,∗ , Julian M.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Implement anisotropic transport by explicitly calculating fluxes at cell faces for momentum, energy, and magnetic fields as described in Section 5.2 and 5.4 to correctly account for parallel, perpendicular, and wedge terms like Righi-Leduc effect and Braginskii viscosity tensors.

  2. Integrate the von Neumann stability analysis results directly into the code's timestep selection logic by using the derived CFL conditions for each term—dissipative terms on the negative real axis and non-dissipative terms requiring a stability region including part of the imaginary axis—to ensure correct time integration methods are chosen.

  3. Develop a modified HLLE Riemann solver that incorporates new signal speeds for all non-dissipative terms, specifically using the largest of the ideal and Hall phase speeds to suppress grid-scale non-linear instabilities that plague current implementations.

  4. Create a well-posedness criterion check for thermoelectric effects by comparing the actual system eigenvalues against the derived criterion: (gamma−1) ((1 − xion)κn + xionκ⊥) (η+ηAD) ≥ x squared ion κ 2 J∧ / εint.

  5. Integrate a mechanism, such as HLLE-Mod or HLLD, to ensure stability for non-dissipative terms by using the derived signal speeds in the Riemann solver fluxes and ensuring the integration method (like RK3) is appropriate for modes requiring imaginary axis stability.

  6. Implement a conservative unsplit timestep calculation using inverse sums of each subsystem's timestep, such as ∆tunsplit = 1/∆tideal + max(1/∆tζ,ν∥,⊥ + 1/∆tκ,∥,⊥ + 1/∆tκ,∧ + 1/∆tη + 1/∆tT E)−1 to bound the non-ideal evolution when super-time-stepping is disabled.

  7. Modify the internal energy flux computation to split it into internal, kinetic, and magnetic parts using HLLE fluxes for each component (Fint = Fint internal + Fint kinetic + Fint magnetic) to ensure consistent numerical dissipation treatment across all energy variables.

Abstract

Athena++ is a finite-volume astrophysical magnetohydrodynamics (MHD) code. As of writing, isotropic thermal conduction and kinematic viscosity, Ohmic resistivity, and ambipolar diffusion are present in the main branch. Here we present a fork of Athena++, ``Athena-Aniso''. It implements fully anisotropic thermal conduction (parallel, perpendicular, and the Righi--Leduc effect); anisotropic viscosity (all five Braginskii viscosity tensors); the Hall effect; and several thermoelectric effects (Peltier, Seebeck, Nernst, and Ettingshausen). These effects can be weighted by ionisation fraction in a partially ionised gas, with the neutral part diffusing isotropically. We have also added isotropic bulk viscosity. The Biermann battery is not implemented here. The dissipative terms have super-time-stepping support, while the non-dissipative terms are combined with the ideal MHD flux. A CFL condition for all terms is derived and implemented using von Neumann stability analysis. The dissipative terms are found to be stable under first-order time integration. The non-dissipative terms are found to be stable under third-order time integration. We additionally derive new signal speeds for a modified HLLE Riemann solver. These suppress grid-scale non-linear instabilities still present in the non-dissipative terms. The Nernst/Ettingshausen system is found to be ill-posed unless it is stabilised by both perpendicular thermal conduction and perpendicular magnetic diffusion. We derive a well-posedness criterion for this. We provide numerical tests of each implemented effect against derived analytical solutions.

Sources

Related papers