FES Science Highlights | U.S. DOE Office of Science (SC)

Illustration of a tokamak, a donut-shaped device with a metal column in the middle and plasma around it

Inverted Plasma Shape Shows Promise for Future Fusion Power Plant Design

Negative triangularity exhibits high core fusion performance and good power handling, pointing to a compelling approach for future fusion pilot plants.

Series of five different line graphs indicating how the code matched the experimental data fairly well

Improving Predictions for Fusion Device Transport

Researchers validate a new workflow for plasma transport models, aiding future fusion device design.

Left: geometry of the SPARC tokamak showing the pump volume and the “louver” actuator that restricts gas flow between the plasma and pumping chambers. Right: neutral gas pressure increases strongly with the louver closed.

“Louvers” on the SPARC Fusion Device Should Exhaust Gases as Hot as a Star

Public researchers partner with a private company to improve simulations key to controlling plasma heat in a fusion energy power plant.

(a) Control scheme used for reinforcement learning (RL). (b) RL control successfully drives plasma through the valley of tearability, avoiding instabilities.

AI Tackles Disruptive Tearing Instability in Fusion Plasma

Researchers trained a deep reinforcement learning algorithm to adjust magnetic confinement fields in real time to maintain plasma stability.

(a) Architecture of real-time field control on the DIII-D and KSTAR tokamaks. (b) This real-time 3D control drives plasma to the highest performance state for normalized fusion gain (energy out vs energy in, <em>G</em>) and confinement quality (<em>H</em><sub>89</sub>) without edge bursts.

Controller with Integrated Machine Learning Tweaks Fusion Plasmas in Real Time

Integrating machine learning with real-time adaptive control produces high-performance plasmas without edge instabilities, a key for future fusion reactors.

(Left) Measured and simulated electron temperature curves during an NBI pulse showing the outcome of the competing effects between neutral particle-derived fast ions and cold electrons. (Right) One of DIII-D’s four neutral beams.

For Heating Plasma in Fusion Devices, Researchers Unravel How Electrons Respond to Neutral Beam Injection

Study finds that neutral beam performance can be experimentally deduced from electron temperature evolution during neutral beam injection.

Contour plot of electron temperature (left) and radial profiles of electron temperature (center) and ion temperature (right) across an island, with ion temperature showing a gradient observed experimen-tally (blue) and reproduced in simulations (red).

In a Fusion Device Plasma, a Steep Ion Temperature Gradient Slows the Growth of Magnetic Islands

The first measurement of ion temperature in magnetic islands identified a steep gradient, providing insights for improving plasma confinement in tokamaks.

For the first time, researchers simultaneously achieved density above a historical empirical limit (>1.0) and very good confinement (~1.5) in a tokamak device.

Ground-Breaking Efforts Overcome an Operational Limit of Tokamaks, Advancing Efforts to Achieve Fusion Energy

By achieving very high density and confinement quality at the same time, researchers make new strides toward fusion energy.

The weight of antihydrogen atoms (a positively charged positron or anti-electron orbiting a negatively charged antiproton) balances the weight of hydrogen atoms (a negatively charged electron orbiting a positively charged proton).

Antihydrogen Falls Downward!

Settling a long-standing question, scientists have proven that antihydrogen falls downward in a first-ever direct experiment.

Negative triangularity shaping in the DIII-D tokamak (left) produced plasmas with no observed instabilities for triangularities less than approximately -0.15, even at high heating power and core performance (right).

Inverting Fusion Plasmas Improves Performance

Plasmas with negative triangularity show reduced gradients that develop into instabilities, including under conditions relevant to fusion power plants.

Using light from the edge of a plasma in a tokamak (interior view at left), a Physics Informed Neural Network reconstructs the turbulent fluctuations in plasma density and temperature and the distribution of a probing helium gas puff (right).

Bridging Theory and Fusion Experiments through Physics-Informed Deep Learning

Neural networks guided by physics are creating new ways to observe the complexities of plasmas.

Opening the Magnetic Bottle of a Tokamak Causes Particles to Rush Inward

Perturbing the edge magnetic field of a tokamak produces a counterintuitive response: particles entering the confined region rather than escaping it.

Science Highlights