Skip to content

The Virial Theorem: No Self-Confinement

Source lecture(s): PC368 Lec 14

Intuition

Could a plasma hold itself together with the magnetic field generated by its own currents — no external coils, no walls? The virial theorem says no, in complete generality, in about five lines. It is one of the few genuinely universal negative results in the subject, and it explains why every confinement device has expensive hardware on the outside.

The statement

For a static MHD equilibrium, \(\nabla p = \mathbf{J}\times\mathbf{B}\). Take the scalar product with the position vector \(\mathbf{r}\) and integrate over all space. Using the stress tensor form and the divergence theorem, the volume terms combine into

\[\int_V\left(3p + \frac{B^2}{2\mu_0}\right)dV = \oint_S\left[\left(p + \frac{B^2}{2\mu_0}\right)\mathbf{r}\cdot d\mathbf{S} - \frac{(\mathbf{B}\cdot\mathbf{r})(\mathbf{B}\cdot d\mathbf{S})}{\mu_0}\right]\]

The left-hand side is strictly positive — pressure and magnetic energy density cannot be negative. So the surface integral must be positive too, and cannot vanish.

Now let the surface go to infinity. For a localised, isolated configuration the fields must fall off at least as fast as a dipole, \(B \sim r^{-3}\), so the surface term dies as \(r^2 \cdot r^{-6} \to 0\). A positive quantity equals zero: contradiction.

Virial theorem

No finite, isolated, static plasma configuration can be confined by its own magnetic field alone. Confinement always requires currents outside the plasma — coils, conducting walls, or gravity.

What it does and does not forbid

Forbids: a self-confined plasma ball floating in vacuum. Every "magnetic bottle made of plasma" scheme is dead on arrival.

Permits:

  • External coils. The surface integral is non-zero because current-carrying conductors sit outside. Every tokamak, stellarator, mirror and pinch works this way — and pays for it, since the coils are typically the most expensive part of the machine.
  • Gravity. Add a gravitational term and the theorem is satisfied. Stars are self-confined plasmas; that is exactly why the Sun does not need magnets.
  • Dynamic equilibria. The theorem assumes static. Flows, jets and continuously driven systems evade it — astrophysical jets are collimated by rotation and external pressure, not by static self-confinement.

The engineering translation

Since the plasma cannot hold itself, the field must be made externally, and the mechanical loads land on the coil structure. The virial theorem even sets a floor on how much structural material a magnet needs: the mass scales as the stored magnetic energy divided by the working stress of the material. This is a real design constraint on reactor magnets, not a curiosity — it is why high-field magnets are so massive, and why the structural steel is often the dominant cost.

Common mistakes

  • Thinking the theorem forbids magnetic confinement. It forbids self-confinement. The distinction is where the currents live.
  • Applying it to stars and concluding they are impossible. Gravity supplies the missing term; the theorem then becomes the standard stellar virial relation.
  • Forgetting the static assumption. Time-dependent and flowing systems are outside its scope.

Knowledge graph position

Prerequisites: MHD equilibrium, magnetic stress tensor. Leads to: magnet and reactor design, astrophysical equilibria.

Quiz

Q1 (conceptual). Where exactly does the proof fail for a star?

Answer

At the step where only pressure and magnetic stress appear. A star has a gravitational potential-energy term, which is negative, so the left-hand side is no longer strictly positive and the contradiction disappears. Gravity is the external agent.

Q2 (conceptual). Why does the surface term vanish for an isolated configuration?

Answer

With no external currents the far field is at best dipolar, \(B \sim r^{-3}\), so the integrand goes as \(B^2r \sim r^{-5}\) while the area grows as \(r^2\) — the product vanishes as \(r^{-3}\). External coils keep the far field from falling off that fast.

Q3 (MCQ). The virial theorem's practical implication for fusion reactors is:

  • (a) plasma pressure must exceed magnetic pressure
  • (b) confinement requires external current-carrying structures, which must bear the load
  • (c) only toroidal geometries can work
  • (d) plasmas must always be dynamic
Answer

(b). Something outside must make the field and take the force — and the required structural mass scales with the stored magnetic energy, which is a first-order cost driver in reactor design.