April 27, 2016
Paul Ennever, as part of his thesis research, has spent more than a year diluting plasma, the hot fuel that powers MIT’s fusion-based experimental tokamak reactor, Alcator C-Mod. Although it may seem paradoxical, diluting plasma appears to stabilize it, decreasing its turbulence and allowing it to reach higher temperatures — essential goals of fusion research. Ennever recently defended his PhD thesis, which details his controlled dilution experiments at the Plasma Science and Fusion Center (PSFC).
Ennever’s interest in fusion took hold at Columbia University, where, enrolled in the School of Engineering and Applied Science, he met Professor Thomas Sunn Pedersen, an MIT graduate who had done his thesis work on Alcator C-Mod. At Columbia Pedersen had built another kind of controlled fusion device, a simple stellarator, to confine and study electrically non-neutral plasmas. It was on this Columbia Non-neutral Torus (CNT) that Ennever performed his first plasma research.
“Prof. Pedersen helped me get started in research when I didn’t know how to do it at all,” he reveals.
The experiment, which involved a non-neutral plasma consisting of only electrons, was being contaminated by ions, which were present in small amounts in the CNT vacuum vessel. Ennever’s job was to build a computer model to track an ion as it moved through the plasma, to see where it would end up. This eventually allowed him to figure out how the ions were influencing the total electric potential of the otherwise pure-electron plasma.
“Learning about all you can do with plasmas, and the promise of fusion, made me interested in working in the field,” says Ennever. He followed that interest to MIT, and to the same tokamak his mentor at Columbia had used for his own dissertation.
As a physics major under the guidance of then PSFC Director Miklos Porkolab, Ennever became absorbed in evaluating turbulence in plasma. Understanding and controlling turbulence is important if plasmas are to reach the temperatures and densities necessary to make fusion effective. And Porkolab’s Physics Division of the PSFC provided a means of measuring turbulence with phase contrast imaging (PCI).
The graduate student was specifically interested in examining the discrepancy between observations of turbulence and related transport of energy in simulations versus measurements from actual experiments.
“Through a simulation you can predict how much turbulence and transport should be present in the plasma,” he says, “the rate at which turbulence is causing the plasma to lose energy. But in Alcator C-Mod, the turbulence and heat loss in the simulation was much larger than that measured in the actual experiment.”
This discrepancy had been observed a few years earlier by Liang Lin, another graduate student under Prof. Porkolab’s supervision. To understand it, every possible parameter of the simulation was changed to see what would make the modeled plasma react like the actual plasma. It was a revelation: diluting the fuel ions actually caused a 50% reduction in the amount of turbulence and even more reduction in the heat loss through ion energy transport.
In order to better comprehend these results, for his thesis research Ennever proposed doing controlled impurity injection experiments on the PSFC’s C-Mod tokamak that would dilute the plasma by known amounts. As in his first research at Columbia, he was again investigating ions contaminating a plasma.
He chose to seed the plasma with nitrogen, puffing the gas into the edge to dilute the fuel. The cryopump would keep the total density of the plasma constant, while nitrogen ions were displacing fuel ions. With the PCI diagnostic he was able to confirm that the energy losses from turbulence decreased, as did the density fluctuations, giving him and his colleagues confidence that the simulations were accurate. In the process he was also able to make the first quantitative measurements of the impurities in the C-Mod plasma.
His simulations, which used state-of-the-art codes developed by General Atomics, supported the findings of a number of fusion experiments performed in the late 1990s, before models were sophisticated enough to make such predictions. They also helped explain in part why tokamak performance decreased when their interior walls were changed from carbon or graphite to metals such as tungsten. It turns out that carbon and graphite produced impurities in the plasma that acted as a dilutant in much the same way as nitrogen, consequently reducing turbulence and related heat loss.
Ennever muses, “Impurities were always considered a negative because they don’t produce fusion and they irradiate more power. But they can help if they are the right kind of impurities.”
He suggests that ITER, the next-step fusion device being built in southern France with US participation, will need to consider similar seeding (perhaps neon) to make sure its metal walls do not get too hot. He notes that, as a by-product of fusion reactions, helium ash will also act as a natural dilutant, further reducing turbulence and improving net energy gain.
As Ennever awaits graduation and writes up his results for publication, he continues to work on Alcator C-Mod, running PCI for the final campaign.
“I feel tremendously privileged to have had the opportunity to work on Alcator C-Mod. I was able to make a tangible contribution to nuclear fusion energy as a graduate student, and there is really no other place where that is true. When the history of fusion energy is written, C-Mod will uniquely stand out as an early contributor of both influential fusion science and influential fusion scientists.”