The self-heating of a fusion plasma by the 3.5 MeV helium-4 nuclei produced in deuterium-tritium reactions — the defining mechanism of a burning plasma and the key to net energy gain.
In a deuterium-tritium (D-T) fusion reaction, each fusion event produces a 14.1 MeV neutron and a 3.5 MeV alpha particle (helium-4 nucleus). The neutron, carrying no electric charge, escapes the magnetic confinement and deposits its energy in the surrounding blanket structure. The alpha particle, however, is a charged particle and remains trapped by the magnetic field. As it slows down through Coulomb collisions with the background plasma electrons and ions, it transfers its 3.5 MeV of kinetic energy to the plasma itself — heating it from within. This process is called alpha particle heating, or simply "alpha heating."[1]
Alpha heating is the mechanism that makes self-sustaining fusion possible. In a plasma with sufficient temperature, density, and energy confinement time (meeting the Lawson criterion), the power deposited by alpha particles can exceed all power losses, meaning the plasma sustains its own temperature without external heating. This condition — called ignition — is the ultimate goal of fusion energy research.
For alpha heating to work effectively, the alpha particles must remain confined long enough to transfer their energy to the bulk plasma. The slowing-down time for a 3.5 MeV alpha in a 10 keV plasma is on the order of 0.5–1 second. During this time, the alpha particle follows a complex orbit determined by the magnetic geometry. In a tokamak, alphas born on passing orbits are generally well confined, but those on trapped (banana) orbits can be lost if the plasma current is too low or if the magnetic field has excessive ripple.[3]
Toroidal field ripple, MHD instabilities (particularly Alfvén eigenmodes driven by the alpha particle population itself), and orbit losses near the plasma edge can all degrade alpha confinement and reduce the self-heating efficiency. Understanding and controlling these loss channels is one of the central physics challenges for ITER and future burning-plasma devices.
A population of super-Alfvénic alpha particles (those traveling faster than the Alfvén speed) can resonantly excite toroidal Alfvén eigenmodes (TAEs) and other energetic-particle-driven modes. If these instabilities grow to large amplitude, they can expel alpha particles before they thermalize, reducing the heating efficiency and potentially damaging plasma-facing components. Predicting and mitigating these instabilities is an active area of theory, simulation, and experimental research on present-day devices using fast-ion surrogates from NBI and ICRH.[3]
Direct alpha heating was first unambiguously observed in the D-T campaigns on TFTR (1994) and JET (1997), where electron temperature increases consistent with alpha power deposition were measured. JET's subsequent 2021–2024 D-T campaigns provided further data on alpha heating in plasmas approaching reactor-relevant conditions, including new measurements of alpha-driven instabilities and their interaction with the thermal plasma.[1]