The MHD Energy Principle
Source lecture(s): PC368 Lec 15
Intuition
Solving the linearised MHD equations for every possible perturbation of a realistic equilibrium is hopeless. The energy principle replaces that eigenvalue problem with a question a physicist can actually answer: does any displacement of the plasma lower its potential energy? If yes, the equilibrium is unstable — a ball on a hilltop, no differential equations required.
The statement
Displace every fluid element by \(\boldsymbol{\xi}(\mathbf{r})\). The change in potential energy is a quadratic functional \(\delta W[\boldsymbol{\xi}]\), and:
Energy principle (Bernstein et al., 1958)
An ideal-MHD equilibrium is stable if and only if $\(\delta W[\boldsymbol{\xi}] > 0 \quad\text{for every allowed } \boldsymbol{\xi}\)$ It is unstable if a single trial displacement makes \(\delta W < 0\).
The asymmetry is what makes it practical. Proving stability needs a minimisation over all \(\boldsymbol{\xi}\); proving instability needs one lucky guess. Most published results are lucky guesses.
Reading the terms
For a plasma with a vacuum region and a wall, \(\delta W\) splits into plasma, surface and vacuum contributions. The plasma term is the instructive one:
The signs organise the whole subject:
- First three terms are positive-definite — always stabilising. Bending field lines, compressing them, and compressing the plasma all cost energy. Nothing you do can make these destabilise you.
- Last two can be negative — the only two sources of instability in ideal MHD. The pressure–curvature term drives interchange and ballooning modes; the parallel-current term drives kinks.
Ideal MHD has exactly two instability drives: pressure gradients in bad curvature, and parallel current. Every named ideal instability is one of these two wearing a hat.
How it is used
The first three terms tell you how to be stable: force any dangerous perturbation to bend or compress field lines.
- Magnetic shear makes a perturbation that is resonant on one flux surface non-resonant on its neighbours, so it must bend field lines somewhere. Strongly stabilising.
- Average minimum-B arranges good curvature on average, so the pressure–curvature term integrates positive.
- A conducting wall modifies the vacuum term, raising \(\delta W\) for external modes — the basis of wall stabilisation and resistive-wall-mode feedback.
- Ballooning theory applies the principle to perturbations localised on the bad-curvature side, and yields the \(\beta\) limits that set reactor economics.
The limits of the method
The principle assumes ideal MHD — perfect conductivity, so field lines cannot break. Any resistivity, however small, opens perturbations that ideal MHD forbids: tearing modes, neoclassical tearing modes, resistive wall modes. A configuration can be ideally stable and still die resistively, only more slowly. Since laboratory plasmas are never perfectly conducting, "\(\delta W > 0\)" is necessary for a working device, not sufficient.
Common mistakes
- Concluding stability from one trial function. A single \(\boldsymbol{\xi}\) with \(\delta W > 0\) proves nothing; you must minimise over all of them. The logic only runs one way.
- Forgetting the constraints on \(\boldsymbol{\xi}\). Trial displacements must respect the boundary conditions and be physically admissible; an inadmissible one can give a spurious negative.
- Treating ideal stability as sufficient. See above — resistive modes do not care.
Related concepts
- MHD instability — the phenomena being classified
- Interchange instability · Kink instability — the two drives
- MHD equilibrium — what is being tested
- Magnetic islands — what resistive modes make instead
Knowledge graph position
Prerequisites: MHD, MHD equilibrium, magnetic stress tensor. Leads to: ballooning theory, beta limits, resistive MHD, tokamak operating limits.
Quiz
Q1 (conceptual). Why is it much easier to prove instability than stability with this method?
Answer
Instability requires finding one displacement with \(\delta W < 0\) — an existence claim, settled by a single well-chosen trial function. Stability requires \(\delta W > 0\) for every admissible displacement, which means a full minimisation over an infinite-dimensional space.
Q2 (conceptual). Which terms in \(\delta W\) can drive instability, and what does each correspond to physically?
Answer
Only two: the pressure–curvature term \(-2(\boldsymbol{\xi}\cdot\nabla p)(\boldsymbol{\kappa}\cdot\boldsymbol{\xi})\), which drives interchange and ballooning modes in bad curvature; and the parallel-current term, which drives kinks. The bending and compression terms are positive-definite and can never destabilise.
Q3 (MCQ). Magnetic shear is stabilising because it:
- (a) increases the plasma pressure
- (b) forces a perturbation resonant on one surface to bend field lines on neighbouring surfaces, adding positive \(\delta W\)
- (c) reduces the parallel current to zero
- (d) eliminates bad curvature
Answer
(b). Shear means the field-line pitch varies across surfaces, so no single perturbation can avoid bending lines everywhere — and bending is positive-definite in \(\delta W\).