Scientists have struggled for years to explain the strange patterns inside tokamaks, donut-shaped machines designed to one day fuse atoms together to create electricity. Inside these devices, a superheated plasma is held in place by a magnetic field. Some of these particles eventually escape from the core and travel toward an exhaust system called a diverter.
When the particles reach the divertor, they collide with a metal plate, where they are cooled and repelled. (The returning atoms help fuel the fusion reaction.) But experiments always revealed unexpected imbalances. Far more particles hit the inner divertor target than the outer divertor target.
This uneven distribution is more than just a curiosity. It will have a major impact on future fusion reactors. Engineers need to know exactly where particles end up in order to design diverters that can withstand extreme heat and stress. The main explanations so far have focused on cross-field drift, which explains how particles move laterally across the magnetic field lines within the divertor. However, simulations that included only this effect failed to reproduce what was shown in experiments, raising questions about whether the models could reliably guide reactor design.
Plasma rotation emerges as a missing element
New research reveals an important piece of the puzzle. Scientists have discovered that toroidal rotation, the motion of the plasma as it orbits around the tokamak, has a strong influence on where particles in the exhaust system end up.
The researchers used the SOLPS-ITER modeling code to simulate the behavior of the particles under different conditions. The result is physical review lettershowed that simulations match real-world measurements only if plasma rotation is included along with cross-field drift. This coordination between model and experiment is essential for designing fusion systems that can operate reliably outside the laboratory.
“There are two components that flow through the plasma,” said Eric M.D., associate research physicist at the U.S. Department of Energy’s (DOE) Princeton Plasma Physics Laboratory (PPPL) and lead author of the study. “There is cross-field flow, where particles move sideways across the magnetic field lines, and parallel flow, where particles move along the magnetic field lines. Many have said that cross-field flow is responsible for creating the asymmetry. What this paper shows is that parallel flow, driven by the rotating core, is just as important.”
Simulation finally matches reality
To test their idea, the team modeled the behavior of plasma in the DIII-D tokamak in California. They ran four different scenarios in which cross-field drift and plasma rotation were turned on and off. The results were clear. None of the simulations matched the experimental data until one important factor was added. It has a measured core rotational speed of 88.4 kilometers per second.
Including both effects, the model closely reproduced the nonuniform particle distribution seen in real experiments. The combined effect of lateral drift and rotation was found to be much stronger than either factor alone.
Designing a convergence system for real-world situations
This finding highlights the important relationship between the rotating plasma core and the behavior of particles at the edges of the system. Accurately capturing this relationship is essential for predicting how exhaust particles will move in future nuclear reactors.
Better predictions mean better engineering. With a clearer understanding of where heat and particles are concentrated, designers can build diverters that are more resilient and better suited to real-world operating conditions.
In addition to Emdee, the research team included PPPL’s Laszlo Horvath, Alessandro Bortolon, George Wilkie, and Shaun Haskey. Raúl Gérou Miguelañez of the Massachusetts Institute of Technology. and Florian Ragner of North Carolina State University.
This research used the DIII-D National Fusion Facility, a DOE Office of Science User Facility, and was supported by DOE’s Office of Fusion Energy Sciences under awards DE-AC02-09CH11466, DE-FC02-04ER54698, DE-SC0024523, DE-SC0014264, and DE-SC0019130.
