Alfvén eigenmode is a discrete, weakly damped oscillation of a magnetically confined 📝plasma that lives inside a gap in the shear Alfvén wave spectrum and grows when energetic particles resonate with it.
A magnetic field line under tension behaves like a plucked string: disturb it and a shear Alfvén wave runs along it at the Alfvén speed, the field strength divided by the square root of the plasma mass density. The waves are named for Hannes Alfvén, who predicted them in 1942 and received the 1970 Nobel Prize in Physics for the 📝magnetohydrodynamics that describes them. In a straight cylinder every 📝magnetic flux surface carries its own Alfvén frequency, forming a continuum in which a global oscillation is absorbed almost as fast as it forms. Bending the field into a torus breaks that continuum: toroidal geometry couples neighbouring poloidal harmonics and opens frequency gaps, in close analogy to the band gaps a crystal lattice opens for electrons. A mode whose frequency falls inside a gap escapes continuum damping and survives as a discrete, global eigenmode. The commonest is the toroidicity-induced Alfvén eigenmode, or TAE; plasma shaping opens further gaps hosting ellipticity- and triangularity-induced modes, and reversed 📝magnetic shear supports the family known as reversed-shear modes or Alfvén cascades.
Because a gap mode is only weakly damped, a small drive destabilises it — and in a 📝burning plasma the drive is the 📝fusion product itself. When a fast ion's velocity along the field approaches the Alfvén speed it resonates with the wave, and the radial gradient of the fast-ion pressure inverts ordinary Landau damping, so the particles feed the mode instead of absorbing it. 📝Alpha particles born at 3.5 MeV supply exactly that population. An unstable Alfvén eigenmode carries them outward before they slow down and thermalise, degrading 📝alpha heating and depositing energetic ions on 📝plasma-facing components. That is why the mode is a burning-plasma problem in a way it is not in a beam-heated experiment: the particles driving the instability are the ones meant to sustain the reaction.
📝Steve Scott and co-authors argued in "Fast-ion physics in SPARC" (2020) that because classical ripple losses in 📝SPARC are computed to be small, the machine should be able to isolate fast-ion redistribution caused by MHD — sawteeth and Alfvén eigenmodes — and that SPARC's eigenmode parameter space even at moderate 📝Q (fusion energy gain factor) overlaps that of 📝ARC, so the measurements would transfer. Those are projections for a tokamak that has not operated. The measurement that does exist was made on 📝JET: a 2025 Physical Review Letters paper (DOI 10.1103/PhysRevLett.134.095103) co-authored by 📝Jon Hillesheim reports the first experimental detection of a zero-frequency fluctuation pumped by Alfvén modes, generated by three-wave coupling with core-localised modes inside the toroidicity-induced gap and its harmonics. On JET the fluctuation was correlated with higher deep-core 📝ion temperature and temperature gradient, a higher H89,P confinement factor and reduced main-ion heat diffusivity — so despite the energetic-particle transport the eigenmodes cause, the zonal fluctuation they pump suppresses turbulence and confinement improves overall.