The rapid destruction of toroidal magnetic surfaces

TL;DR

This paper discusses the rapid destruction of magnetic surfaces in plasma, crucial for ITER's operation, by analyzing how ideal magnetic evolution can trigger fast magnetic reconnection leading to disruptions.

Contribution

It provides a review of the mathematical foundations of magnetic reconnection theory and explains how ideal evolution can cause reconnection to dominate.

Findings

Reconnection occurs when $\Delta_{max}/\Delta_{min}$ exceeds a critical ratio.

Traditional reconnection theory focuses on steady-state current layers.

Understanding ideal evolution is key to predicting disruptions in plasma devices.

Abstract

The operation of ITER will require reliable simulations in order to avoid major damage to the device from disruptions. Disruptions are the sudden breakup of magnetic surfaces across the plasma volume -- a fast magnetic reconnection. This reconnection can be caused by the growth of perturbations outside of the plasma core causing an ideal perturbation to the core. This causes an increasing ratio of the maximum to the minimum separation, $\Delta_{max}/\Delta_{min}$, between neighboring magnetic surfaces. Magnetic reconnection becomes a dominant process when magnetic field lines can quickly interchange connections over a spatial scale $a_r$. This occurs when $\Delta_{max}/\Delta_{min}\gtrsim a_r/\Delta_d$, where $\Delta_d$ is the scale over which non-ideal effects make magnetic field lines indistinguishable. Traditional reconnection theory is fundamentally different. It is a study of the steady-state cancellation of oppositely directed magnetic field components across a thin layer. During more than sixty years, mathematical implications of Faraday's Law have been derived that clarify and constrain the physics of fast magnetic reconnection. These are reviewed because they are not commonly known but are needed to understand and to place in context how an ideal magnetic evolution can cause reconnection to quickly become a dominant process no matter how small $\Delta_d/a_r$ may be.