edge localized mode (ELM) is a repetitive eruption at the edge of a high-confinement tokamak plasma that expels short, intense bursts of particles and energy toward the machine's wall.
ELMs are the price of 📝H-mode. When a 📝Tokamak crosses into the high-confinement regime, a 📝transport barrier forms at the edge of the 📝plasma and the pressure there steepens into a 📝pedestal on which the whole core profile stacks. That steep edge is a loaded spring: when its gradient and the current flowing in it exceed the peeling-ballooning stability limit — the edge cousin of the 📝ballooning mode — the barrier partially collapses, flinging filaments of hot plasma outward before the pedestal rebuilds and the cycle repeats, often many times per second. The distinction against a 📝disruption is scale and survival: an ELM relaxes only the edge and the discharge continues, while a disruption terminates the whole plasma.
Each burst can carry several percent of the stored energy to the 📝divertor surfaces in under a millisecond. Small machines shrug this off; at reactor scale, uncontrolled large ELMs would erode 📝plasma-facing components at unacceptable rates, which is why ELM control is a design requirement rather than an operational nicety. The main tools are the 📝resonant magnetic perturbation (RMP), applied fields that hold the edge just below the eruption threshold; pellet pacing, which triggers frequent small ELMs instead of rare large ones; and regimes engineered to be ELM-free.
