The goal of this project is development of an algorithm to model disruption radiation in three dimensions (3D), to measure and predict disruption mitigation efficacy.
One of the most promising approaches to achieving net fusion energy is the Tokamak. Tokamaks use strong magnets to confine a plasma and heat it to fusion temperatures. Plasmas occasionally escape confinement, through a variety of plasma instabilities. A major loss of confinement is called a disruption. In a disruption, the released plasma energy can damage the tokamak, melting or warping key components.
One way to prevent or reduce damage from disruptions is to inject neutral atoms, often noble gasses like neon or argon, into the plasma before it disrupts. Neutrals absorb energy from the plasma and release it as less damaging visible or ultraviolet radiation. This radiation can be detected and measured by sensors called bolometers. Bolometers detect broad spectrum radiation along a linear chord of the plasma. To determine the total energy released through mitigation, it is crucial to model radiation emitted throughout the plasma using many bolometers. Emis3D is a software analysis tool designed at the core of this project which uses reduced \(\chi^2\) statistics and a modified guess-and-check algorithm to match a synthetic radiation structure (image right) to experimental disruption radiation (image left). The images from the figure are from a mitigated disruption through pellet injection during a JET tokamak discharge.
JET holds many of the records for fusion energy generation. Net energy tokamaks like ITER and SPARC rely on data from JET to inform their design and operation. Emis3D can be used to determine radiation structures and measure radiated energy on JET, and as a validation tool for comparing mitigation simulations to JET experiments. Emis3D was also used as a control room tool in recent JET experimental mitigation campaigns. The next step is to adapt Emis3D for an investigation into the relationship between the thermal energy fraction of the plasma and mitigation efficacy.
SPARC is a net energy tokamak experiment planned to begin operation in 2025. Emis3D workflow is currently being adapted for synthetic SPARC bolometer layouts and SPARC mitigation simulations. Thanks to this workflow, it was demonstrated that the currently planned SPARC bolometer layout is sufficient to accurately measure radiated energy during several stages of a full power SPARC disruption to within the desired uncertainty of 10%. The team will expand this analysis to assess the bolometer layout performance for complete SPARC disruptions, and to evaluate its performance across a broader range of disruption scenarios. Emis3D is also under consideration as as a potential control room tool for SPARC operation.
The Emis3D team is led by graduate student Ben Stein-Lubrano (MIT Physics), with the close collaboration of Ryan Sweeney, Disruption Scientist at Commonwealth Fusion Systems (CFS); Rebecca Li, Bolometer Diagnostics Lead at CFS; and Jacob Rabinowitz, a student at Columbia University. Ben’s supervisors are Dr. Robert Granetz and Dr. Earl Marmar.