Lehigh University By: Kurt Pfitzer | February 09, 2017
Serious discussions of the long-term future of energy use, says Arindam Banerjee, are turning more and more to nuclear fusion, a potential power source that not long ago was considered a distant dream.
Fusion is the process by which the sun generates heat and light. It occurs when two isotopes of hydrogen—deuterium and tritium—collide at great speeds under extreme heat and pressure and fuse to form helium, losing a small amount of mass and releasing huge quantities of energy.
Unlike fossil fuel-fired power plants, fusion reactors emit no greenhouse gases such as carbon dioxide or toxic pollutants such as mercury or sulfur dioxide. Unlike nuclear power plants, which generate energy from nuclear fission, or the splitting of the atom, fusion reactors produce no long-lasting radioactive waste. And the fusion process itself is self-sustaining and highly efficient.
Scientists have exercised caution when predicting the future of fusion, saying it will take at least several decades of further research for commercial fusion to become a viable, cost-effective alternative to existing sources of energy.
Banerjee, an associate professor of mechanical engineering and mechanics and director of the Turbulent Flow Design Group, says advances in supercomputing and the development of novel experimental techniques give cause for greater optimism.
“In the last decade,” he says, “advances in supercomputers have given us enormous power to numerically model complex phenomena like nuclear fusion. This has resulted in quite a leap in understanding the fusion process.”
Half a dozen process occur simultaneously inside a nuclear fusion reactor, says Banerjee. They include nuclear reactions, combustion, radiative heat and shock waves. Each process interacts with and influences the other processes. For fusion research to advance, all the processes and interactions must all be numerically modeled and understood—a task for which today’s supercomputers are ideally suited.
“The new tools enable us to do large-scale simulations. How large? Ten years ago, I would have needed a supercomputer to run simulations that I can now run on my laptop. And today, the largest supercomputer can model the entire nuclear fusion process, including the equations that represent the underlying coupling between all of the physical phenomena.”
The second reason for Banerjee’s optimism is a unique experiment that he and his students are conducting on a two-wheel, high-acceleration facility in the basement of Packard Laboratory. The test replicates the turbulence that occurs inside the fusion reactor and sheds light on the hydrodynamic instabilities that limit the efficiency of the reactor.
Inside the fusion reactor
The simultaneous phenomena that make nuclear fusion so complex occur inside the fusion reactor under extremely high pressure with temperatures reaching tens of millions of degrees Celsius. The heat sustains the fusion reaction, while the pressure confines the hydrogen isotopes, allowing them to fuse.
Researchers today are testing two different approaches to achieve fusion—magnetic confinement and inertial confinement. Banerjee studies the second method, in which high-powered x-rays and lasers heat metal shells containing hydrogen isotopes as they pass through the reactor. The goal is for the isotopes to fuse and release energy before the heat from the lasers and x-rays melts the shells.
During this process, says Banerjee, a multifluid surface forms between the melting metal and the hydrogen gas, creating a phenomenon called Rayleigh-Taylor Instability, which occurs at the interface of two fluids of differing densities.
“Rayleigh-Taylor Instability causes the metal shell to rupture and the gas inside to escape,” says Banerjee. “The result is a reduced yield of inertial confinement fusion process.”
Banerjee and his students want to learn how Rayleigh-Taylor Instability evolves over time inside the inertial confinement capsule and how it can be suppressed or mitigated. They are contributing their results to the “larger picture” being developed by physicists at Los Alamos National Laboratories in New Mexico.
“Our goal at Lehigh is to better understand the hydrodynamic instability of the mixing of the hydrogen gas with the molten metal. But hydrodynamics is just one feature of what is happening inside the fusion reactor. So we provide Los Alamos with data from our model that helps their physicists, who are developing more elaborate and detailed models of all the phenomena occurring inside the reactor.”
Banerjee’s group at Lehigh is carrying out several sets of tests. In one, they are using two real-world fluids—mayonnaise and air—to mimic the behavior and interactions of the molten metal and hydrogen inside the fusion reactor.
“We use mayonnaise to represent the molten metal,” says Banerjee, “because it has properties that, like molten metal, are rate-dependent. If you change the shear or strain rate, in other words, the properties of the mayonnaise change too. This mimics what you have in an inertial confinement fusion capsule, in which one of the two fluids—the molten metal—has rate-dependent behavior.”
A second experiment is conducted on the two-wheel, high-acceleration testing facility in Packard Lab. Banerjee and his students fill a container with two fluids of different densities, place it on the track of one of the two wheels of the test facility and turn on the track so the container circles it at a predetermined speed. When the track is “flipped,” the container switches from one wheel to the other, changing its direction of gravity in the process.
“On one wheel of the track,” says Banerjee, “the container is on a stable stratification. When it flips to the other side, it becomes unstable. It rotates like a centrifuge so the heavy fluid is outside and the lighter fluid is inside. The moment the track is flipped, the stratification of the container is reversed, causing the lighter fluid to move to the outside and the heavier fluid to the inside.”
The change in orientation of the two fluids, says Banerjee, causes mixing, which is an undesired outcome inside the fusion reactor as it allows hydrogen isotopes to escape from the inertial confinement capsule before fusing.
The purpose of this experiment, says Banerjee, is to determine the Reynolds threshold number at which mixing occurs and then go a step further.
“The challenge is that no one has been able to run experiments beyond that threshold Reynolds number. The facility we’ve built actually allows us to run experiments at Reynolds numbers beyond that threshold number. Beyond that number, mixing becomes universal.”
The two-wheel, high-acceleration facility, says Banerjee, is the “only one of its kind in the world. “We are, in terms of Reynolds capability, several orders of magnitude improved over what any other experiment can do right now.”
A critical advantage of the facility, says Banerjee, is that it enables researchers to conduct experiments at a relatively low Reynolds number and predict the results of experiments conducted at much higher values, thus allowing for “order-of-magnitude analysis.”
Banerjee has conducted fusion-related research for 12 years. His work is supported by a five-year CAREER Award from the National Science Foundation and by successive grants from the U.S. Department of Energy. He has published his results in some of the top journals in his field, including the Journal of Fluid Mechanics and Physical Review E.
He is confident about fusion’s future.
“In the last 20 years we have made significant progress. In the next decade or two, I think we can attain nuclear fusion—if we continue to get funding and if we get the coming generation of researchers excited.”