From the cosmos to fusion plasmas, PPPL presents findings at global APS gathering
Among PPPL presenters was Seth Davidovits, a 2017 graduate of the Program in Plasma Physics in the Princeton University Department of Astrophysical Sciences, who spoke as winner of the Marshall N. Rosenbluth Outstanding Doctoral Thesis Award for his dissertation on the theory and simulation of turbulence in suppressing fluids. Davidovits is now a post-doctoral research fellow at Princeton and PPPL.
Invited talks by PPPL scientists covered topics ranging from the formation of stars and planets to the development of computer codes for predicting and avoiding disruptions of fusion plasmas. These talks included the following:
Developing a path to stable tokamak operation
Among the hurdles to capturing and controlling the power of fusion that drives the sun and stars is the risk of disruption of plasma, the hot, charged state of matter composed of free electrons and atomic nuclei that fuels fusion reactions. Disruptions can halt the reactions and damage the doughnut-shaped devices called tokamaks that confine the plasma in magnetic fields. Operators of tokamaks must therefore develop real-time control of plasma instabilities that can lead to disruptions while pushing plasma toward the best possible performance.
Physicists at the U.S. Department of Energy's (DOE) Princeton Plasma Physics Laboratory (PPPL) and Princeton University, headed by Egemen Kolemen of PPPL and Princeton, have conducted real-time analyses that predict approaching disruptions and reduce instabilities while maintaining high performance. Such performance, called "high beta," is the ratio of plasma pressure—a key ingredient in fusion reactions—to the confining magnetic field. The higher the ratio, signifying the creation of relatively high pressure with relatively low magnetic fields, the better the confinement and control of the plasma and its ability to create fusion.
The real-time analyses employed both physics-based and machine learning computer programs, or algorithms, that the researchers developed. The first type uses physics first-principles while the second uses data gleaned from previous experiments. The physicists used both types to control plasma experiments on the DIII-D National Fusion Facility, a DOE Office of Science user facility operated by General Atomics in San Diego, California.
The physics-based analysis detected growing instabilities prior to disruptions thousands of times faster than a statistical Monte Carlo approach. The analysis showed that plasma becomes "touchy" and produces minor variations in equilibrium before an instability called a "tearing mode" that can lead to disruptions sets in.
However, the physics-based algorithms could accomplish only so much. So researchers applied data-driven machine learning techniques that utilized two-to-three years of DIII-D instabilities and disruptions. The best machine learning algorithms then predicted DIII-D disruptions more than 90 percent of the time. "Taken together, the two algorithms proved that accurate prediction of instabilities could better enable the stabilization of high-performance plasmas without leading to disruptions," Kolemen said. Support for this work comes from the DOE Office of Science.
A key step toward understanding the development of heavenly bodies
The cosmos is a void dotted with stars and an ever-increasing number of newly-observed planets discovered beyond our solar system. However, the formation of these stars and planets out of clouds of interstellar dust and gas remains mysterious.
The study of black holes provides clues to the solution of this mystery. Illustrations of black holes typically depict them as vacuum cleaners sucking up all matter and light. In reality, clouds of dust and gas called accretion disks swirl around black holes, gradually moving closer and closer until they are trapped by the black holes and fall into them. Experiments led by researchers studying the Magnetorotational Instability (MRI) at PPPL help verify one of the proposed models for how this process works.
Typical orbits, such as those that planets carve around our sun, continue for billions of years because their angular momentum—the conservation of which causes ice skaters to spin faster when pulling in their arms—prevents the planets from falling into the sun. In an accretion disk, forces such as friction can cause objects to lose their angular momentum but are insufficient to explain how quickly matter falls into the body that the disk orbits. MRI can provide an explanation.
One of the experiments at PPPL simulates this process using a unique rotating water-filled device. Video is recorded of a water-filled red plastic ball as it moves away from the center of the device. A spring in the experiment connects the ball to a post to simulate magnetic forces. Position measurements of the ball indicate that the behavior of its angular momentum is consistent with the MRI predictions of developments in a real accretion disk.
Researchers are now conducting experiments using spinning liquid metals to study what happens in accretion disks with actual magnetic fields present. The experiments confirm how strongly the magnetic field affects the metal and pave the way toward a clear understanding of the role the fields play in accretion disks. The combined results mark a significant step toward a more complete explanation of the development of heavenly bodies. Support for this research comes from sources including the DOE Office of Science, the National Science Foundation, and the National Nuclear Security Administration.
Twist and turn: A new understanding of the rotation of fusion plasma
Direct measurement of the main-ion velocity in fusion plasmas provides insight into the turbulent transport of momentum and the mechanisms that generate plasma rotation. Understanding rotation of the main ions provides a key to validating models of turbulent momentum transport.
Such measurements, led recently by physicist Brian Grierson of PPPL on the DIII-D National Fusion Facility at General Atomics, are distinct from the commonly measured rotation of carbon and other impurities that swirl within the plasma. The distinction, which provides improved understanding of the ability of the plasma to generate its own "intrinsic rotation," has two principal aspects:
- First, the main-ion rotation in the outer regions of the plasma is twice the rate of the impurity rotation. This finding is consistent with the different pressure forces and the neoclassical flows between the bulk plasma and the low-concentration carbon impurity.
- Second, increasing the plasma density causes the main-ion rotation speed to evolve from a constant value across the profile, to a hollow profile, meaning that the edge of the plasma rotates faster than the center of the plasma. This difference in the shape of the rotation profile tells physicists whether the plasma is responding to a strong and large scale self-generated torque, which plays a key role in maintaining the stability of the plasma.
If only the impurities were measured, physicists might incorrectly conclude that the plasma is generating a torque that causes the plasma rotation to peak, which would not be the case. It is therefore essential to measure the bulk—or main ion —plasma rotation when studying the intrinsic rotation of fusion plasmas.
"Understanding how turbulence generates rotation in fusion reactors is important, because in future larger machines the ability to drive rotation with high power neutral beam injection will be relatively small," says Wayne Solomon, deputy director of the DIII-D Program. Strong intrinsic rotation will thus be key to stable plasmas. Support for this work comes from the DOE Office of Science.
No longer whistling in the dark
Magnetic reconnection, the snapping apart and violent reconnection of magnetic field lines in plasma, occurs throughout the universe and can whip up space storms that disrupt cell phone service and knock out power grids. Now scientists at PPPL and other laboratories, using data from a NASA four-satellite mission that is studying reconnection, have developed a method for identifying the source of waves that help satellites determine their location in space.
The team of researchers, led by PPPL physicist Jongsoo Yoo, have correlated magnetic field measurements taken by the Magnetospheric Multiscale (MMS) mission that is orbiting at the edge of the magnetic field that surrounds the Earth. The findings identified the source of the propagation of "whistler waves"—waves with whistle-like sounds that drop from high to low and stem from reconnection—whose detection orients the satellites relative to reconnection activity that can affect the Earth.
The research marks development of "a new methodology for measuring how the wave propagates in reconnection," said Yoo. The source, he said, is what are called "tail electrons"—particles with energy that is far greater than that of the bulk electrons in reconnecting field lines. "What we prove is that you couldn't have whistler waves without the active X-line"—the central reconnection region—"so whistler waves indicate that reconnection is near," Yoo said.
The team now plans to investigate the development of whistler waves near the electron diffusion region, the narrow region in the magnetosphere and laboratory experiments where electrons separate from field lines before reconnection takes place. Results could prove relevant to the MMS mission, whose goals include uncovering the role that electrons play in facilitating reconnection. Support for this work comes from the DOE Office of Science, NASA, and the National Science Foundation.
Using the right magnetic fields for the job
As it does for a spinning top, rotation helps smooth out any wobbles or instabilities in the hot, charged plasma that fuels circular fusion devices known as tokamaks. One way to control this rotation is to create asymmetric perturbations, or ripples, in the plasma with external magnetic coils. Now physicist Nik Logan of PPPL and PPPL researchers have validated predictions of the optimal ripples for their desired "neoclassical toroidal viscosity torque" (NTV)—a fancy way of saying their effect on the rotation.
Validation of these predictions on the DIII-D National Fusion Facility enables optimization of external coils to control the plasma rotation, a major factor in plasma stability. The ripples themselves are "non-resonant," which means that they impact the momentum of plasma rotation but not the plasma's density and energy. The validation allows researchers to arrange and design coils to produce the most effective 3-D perturbations from an infinite array of possibilities, which could prove beneficial to both existing and future tokamak devices. Support for this work comes from the DOE Office of Science.
For ITER: A new way to monitor the stability of fusion plasmas
Plasma, the soup of free-floating electrons and atomic nuclei that fuels fusion reactions, exhibits many types of behavior, or modes, when perturbed by magnetic forces in doughnut-shaped tokamaks that house the reactions. New findings led by physicist Zhirui Wang of PPPL clearly distinguish between modes and offer the potential for understanding and controlling the impact of perturbations on instabilities called edge localized modes (ELMs) and for the real-time monitoring of plasma stability.
"Such monitoring can serve as the key to an integrated approach for disruption prediction and avoidance in future reactors such as ITER," the international tokamak under construction in France, Wang said.
Researchers first developed a model for extracting the dominant modes that stem from the response of plasma to externally applied 3-D magnetic fields. Some modes can suppress ELMs while others can lead to disruptions, so extracting the dominant type can be crucial for predicting disruptions.
The physicists then validated their model with experiments on the DIII-D National Fusion Facility and on the Experimental Advanced Superconducting Tokamak (EAST) in China. In both cases, the model provided accurate descriptions of the development of modes and correctly extracted the dominant modes.
Going forward, the findings can enable researchers to quantitatively identify the stability of dominant modes, and to predict disruptions or optimize RMPs for suppressing ELMs. "We can monitor the stability of the mode and predict at what point it becomes unstable," Wang said. "The model has fit the experiments quite well." Support for this work comes from the DOE Office of Science.
An effective paradigm for characterizing and forecasting tokamak disruptions
High-reliability disruption prediction and avoidance are critical needs for next-step tokamaks such as ITER. PPPL scientists led by Steven Sabbagh, a senior research physicist and adjunct professor at Columbia University on long-term assignment to PPPL, have developed a unique Disruption Event Characterization and Forecasting (DECAF) code.
The code provides a unified paradigm that automates the analysis of tokamak data to determine chains of events leading to disruptions and to forecast their evolution. The approach supports a range of methods ranging from first-principles physics analysis to empirical models to provide a flexible framework for evaluating the proximity of plasma states to a disruption event.
An expanding data base of tokamak activity in the United States, Asia, and Europe continues to be collected for the code to successfully produce insights into the forecasting of disruptions. Support for this work comes from the DOE Office of Science.
Provided by Princeton Plasma Physics Laboratory