Magnetic Levitation: How Superconductors Cheat a 19th-Century No-Go Theorem

A theorem that says stable levitation is impossible

In 1842 Samuel Earnshaw proved a deceptively simple result: a collection of point charges (or magnetic dipoles) interacting only through inverse-square forces cannot be held in stable static equilibrium by those forces alone. Mathematically it follows from the fact that a magnetic or electric potential energy obeys Laplace's equation in free space, which forbids a true local minimum in three dimensions — any equilibrium point is a saddle, stable in some directions and unstable in others. That is why you cannot balance a permanent magnet motionless above another one by hand; nudge it slightly and it either snaps to the other magnet or flips and flies away.

Every real levitation trick therefore has to break one of Earnshaw's assumptions — using time-varying fields, feedback, diamagnetism, or genuinely quantum physics rather than simple static dipole attraction.

The Meissner effect: a superconductor that refuses a field

A type-I superconductor below its critical temperature does something no ordinary conductor does: it actively expels magnetic field from its interior, the Meissner effect. Surface supercurrents spontaneously arise to cancel the field inside, and by Lenz's-law-like reasoning those same supercurrents push back against any magnet brought close — the superconductor behaves as a nearly perfect diamagnet. Bring a strong permanent magnet near a superconducting disc and the repulsive force can be large enough to support the magnet's weight: stable, silent levitation with essentially zero friction, because the supercurrents flow without resistance.

B_inside ≈ 0                       (Meissner effect, ideal type-I superconductor)
below T_c and below H_c1/H_c2:     field is expelled or only partially penetrates
force ~ −∇(interaction energy) is now genuinely repulsive and restoring in ALL directions
    → this sidesteps Earnshaw because the superconductor is not a fixed dipole,
      it is an induced response that always opposes the change in field

Type-II superconductors and flux pinning: why it locks in place

Pure Meissner levitation from a type-I superconductor is real but delicate — it is stable to small perturbations only because the induced supercurrents actively resist any change, not because of a true energy minimum in the Earnshaw sense. Most practical levitation demonstrations instead use type-II superconductors (like YBCO), which behave differently above a lower critical field: rather than expelling flux completely, they let magnetic field penetrate in discrete quantised tubes called Abrikosov vortices, each carrying one flux quantum Φ₀ = h/2e.

In a real, imperfect crystal these vortices get pinned at defects, grain boundaries and impurities — they cannot move freely without energy cost. That pinning is what makes type-II superconducting levitation so strikingly stable: the material doesn't just repel the magnet, it locks onto the exact field configuration it was cooled in, so it can hover, hang beneath a magnetic track, or stay fixed in mid-air at an odd angle, resisting displacement in every direction because moving would require dragging pinned vortices along with it.

Other ways around Earnshaw's theorem

Static superconductors are not the only trick. Electrodynamic levitation (as in some maglev trains and the classic "Levitron" spinning-top toy) uses time-varying or velocity-dependent forces — induced eddy currents that only appear when there is relative motion, or gyroscopic precession that converts an unstable equilibrium into a dynamically stable one — both explicitly outside Earnshaw's static-force assumption. Active feedback levitation (used in maglev trains like the German Transrapid and in magnetic bearings) simply measures the gap continuously and adjusts electromagnet current in real time, engineering stability with control theory rather than finding it in a static field configuration. Ordinary diamagnetic levitation of non-superconducting materials — famously, a live frog in a sufficiently strong magnet — works because any material has a weak induced diamagnetic response that, in a strong enough field gradient, can exceed gravity; it obeys the same induced-response loophole as the Meissner effect, just far more weakly.

What actually holds the object up

In every one of these cases, the force that appears to defy Earnshaw's theorem is not a static dipole-dipole attraction at all — it is a response that only exists because the system is opposing change: induced supercurrents opposing flux change, eddy currents opposing relative motion, or a feedback controller opposing displacement. Earnshaw's proof applies strictly to fixed, static configurations of simple dipoles; anything that actively reacts to being disturbed is exempt, and that is the loophole every levitation demonstration in this simulation exploits.

Frequently asked questions

Try it live

Everything above runs in your browser — open Magnetic Levitation and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.