HomeArticlesElectromagnetism

Plasma Waves and Landau Damping: A Wave That Dies Without Friction

In an ionised gas, electrons can slosh back and forth as a wave with no collisions at all - and that same collisionless plasma can also make the wave vanish, by quietly handing its energy to the particles riding its crest.

mysimulator teamUpdated June 2026≈ 9 min read▶ Open the simulation

The simplest wave in a plasma

Displace a slab of electrons in an ionised gas away from the much heavier, nearly stationary ions, and the resulting charge separation creates an electric field that pulls the electrons straight back - and, having built up momentum, they overshoot and oscillate. This is the Langmuir wave, or electron plasma oscillation, and unlike a sound wave it needs no collisions between particles to work; the restoring force is purely electrostatic. Its natural frequency, the plasma frequency, depends only on the electron density.

Adding thermal motion: the Bohm-Gross dispersion

In a perfectly cold plasma, every electron would oscillate in place at exactly the plasma frequency regardless of wavelength - a standing pulsation, not a travelling wave. Real plasmas are warm, and that thermal spread of electron velocities lets the disturbance actually propagate, with faster-moving thermal electrons carrying the oscillation forward slightly ahead of the bulk. The resulting dispersion relation, connecting frequency to wavenumber, is named for David Bohm and Eugene Gross:

ωpe = sqrt(ne · e² / (ε0 · me))       electron plasma frequency

ω² = ωpe² + 3 · k² · vth²             Bohm-Gross dispersion relation

ne  = electron number density        vth = electron thermal velocity
k   = wavenumber (2π / wavelength)   ω   = wave frequency
live demo · an electron density wave riding through a plasma● LIVE

Landau damping: no collisions required

In 1946, Lev Landau discovered something the fluid picture above completely misses: analysing the plasma's full velocity-distribution physics through the Vlasov equation, he showed that a plasma wave can lose energy even with zero collisions. The mechanism is resonance. Electrons moving at close to the wave's phase velocity ω/k stay in step with it for a long time and exchange energy efficiently, like a surfer riding a swell. For the typical Maxwellian (thermal) velocity distribution, there are slightly more electrons moving just slower than the phase velocity than just faster than it. The wave accelerates that slight excess of slower electrons up to its own speed, paying for the acceleration out of its own energy - and because there are more of them being sped up than being slowed down, the net transfer drains the wave. The wave's amplitude decays exponentially, purely from this kinetic resonance, with no friction or particle collisions anywhere in the picture.

Instability: when the process runs backward

The sign of the effect is set entirely by the slope of the velocity distribution at the wave's phase velocity, not by its absolute height. A normal thermal distribution slopes downward there (fewer fast particles than slow ones), which damps the wave. But inject a beam of fast particles into the plasma and the distribution can develop a "bump on the tail" - a region where there are more particles just faster than the phase velocity than just slower. There, the wave now extracts more energy from the excess of fast particles being slowed down than it gives up to slow ones being sped up, so the wave grows instead of decaying. This inverse Landau damping, better known as the beam-plasma instability, is how particle beams in astrophysical plasmas and some fusion devices can spontaneously excite large-amplitude waves.

Frequently asked questions

How can a wave lose energy in a plasma with no collisions at all?

Through a purely kinetic resonance - particles moving at nearly the wave's phase velocity exchange energy with it as they surf alongside it. For a typical velocity distribution there are more particles just below that speed, which the wave accelerates at its own expense, than just above it, so the net effect drains the wave's energy without any particle-particle collisions.

What decides whether a plasma wave damps or grows?

The slope of the particle velocity distribution at the wave's phase velocity. A negative slope, as in a normal thermal distribution, damps the wave. A positive slope, as in a beam of fast particles injected into the plasma, makes the wave grow instead, known as inverse Landau damping or a beam-plasma instability.

Is Landau damping the same as ordinary friction or resistive damping?

No. Resistive damping comes from collisions dissipating energy as heat through random scattering, while Landau damping is a reversible, collisionless kinetic effect derived directly from the Vlasov equation. It still shows up as an exponential decay of the wave's amplitude, but the mechanism is resonant energy transfer, not friction.

Try it live

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

▶ Open Plasma Wave simulation

What did you find?

Add reproduction steps (optional)