Fusion Science and Technology / Volume 42 / Number 2,3 / September 2002 / Pages 298-314
Intensive studies on the physics of disruptions and developments of avoidance/mitigation methods of disruption-related phenomena have being carried out in JT-60U. The characteristics of the disruption sequence were well understood from the observation of the relationship between the heat pulse onto divertor plates during thermal quench and the impurity influx into the plasma, which determined the speed of the following current quench. A fast shutdown was first demonstrated by injecting impurity ice pellets to the plasma and intensively reducing the heat flux on first wall. The halo current and its toroidal asymmetry were precisely measured, and the halo current database was made for ITER in a wide parameter range. It was found that TPF × Ih/Ip0 was 0.52 at the maximum in a large tokamak like the JT-60U, whereas the higher factor of 0.75 had been observed in medium-sized tokamaks such as Alcator C-Mod and ASDEX-Upgrade. The vertical displacement event (VDE) at the start of the current quench was carefully investigated, and the neutral point where the VDE hardly occurs was discovered. MHD simulations clarified the onset mechanisms of the VDE, in which the eddy current effect of the up-down asymmetric resistive shell was essential. The real-time Zj measurement was improved for avoiding VDEs during slow current quench, and plasma-wall interaction was avoided by a well-optimized plasma equilibrium control. Magnetic fluctuations that were spontaneously generated at the disruption and/or enhanced by the externally applied helical field have been shown to avoid the generation of runaway electrons. Numerical analysis clarified an adequate rate of collisionless loss of runaway electrons in turbulent magnetic fields, which was consistent with the avoidance of runaway electron generation by magnetic fluctuations observed in JT-60U. Once generated, runaway electrons were suppressed when the safety factor at the plasma surface was reduced to 3 or 2.