Fusion systems need inner walls that are able to withstand extreme heat. One promising solution uses liquid lithium to protect the walls, held like water in a sponge made of the exceptionally strong metal tungsten. An advanced manufacturing process can be used to make tungsten into sponge-like wall tiles with lots of pores for flowing liquid lithium. But this process also leaves the tungsten contaminated with other materials, such as carbon, oxygen or nitrogen. When exposed to the liquid lithium, the carbon and oxygen react to form solids that can plug the holes in the tungsten, preventing the lithium from flowing properly. Even if the tungsten was cleaned at the end of the manufacturing process, it would only become recontaminated when the tiles are exposed to air during installation.
Now researchers from the U.S. Department of Energy's Princeton Plasma Physics Laboratory (PPPL), Princeton University and Pennsylvania State University have found a clever way to clean those tiles after they are installed and sealed inside the fusion system - in vacuum chambers that have removed the air. The advance, which uses heat combined with particles from a neon plasma to knock contaminants out of tiles, could help future fusion systems run better with liquid lithium. The results appear in the journal Nuclear Materials and Energy.
PPPL: A leader in liquid lithium research for fusion
In the experiments, highly contaminated porous tungsten samples were placed inside of PPPL's fusion system known as the Lithium Tokamak Experiment-Beta (LTX-β). This experimental system was designed specifically to research the effect of liquid lithium on the plasma and materials that directly face it. Once the tungsten samples were placed inside LTX-β, they were exposed to a neon glow discharge, which is a kind of low-temperature plasma, and then heated to roughly 800 degrees Celsius. Combined, the heat and the plasma proved to be an effective cleaning treatment and significantly more effective than using the plasma alone.
Camila López Pérez, who led the research, said the change was even visible to the eye, as the sample changed color from a dull dark gray to a metallic silvery color. At first, López Pérez thought something had gone wrong.
"I initially thought we were depositing material on the sample, which would have been a terrible result," López Pérez said. "Then the analysis showed that wasn't the case. It was just being cleaned very effectively. It was very surprising and very exciting to see."
In fact, the fraction of clean, metallic tungsten at the surface of the sample jumped from zero to 87%, and the fraction of carbon dropped sharply.
"This research serves as a framework for cleaning these porous tungsten parts while they are inside the fusion system so that material contamination is not what's holding you back," she said. López Pérez first came to PPPL as part of the Dr. Robert A. Ellis Fellowship in 2024 from the Pennsylvania State University and will return to the Lab this fall as an associate research physicist under the Strategic Science Initiative Fellowship.
Cleaning with neon
Glow discharge cleaning has been used for decades. But no one had documented its use to clean materials made from powder, like the porous tungsten samples, while in the fusion system and then studied what happened without ever exposing the sample to air. Neon was chosen over other gases, such as helium, because its charged particles are heavier and therefore better at knocking off carbon and oxygen impurities.
López Pérez also led the project that significantly upgraded the probe that held and measured the sample. This new probe is able to hold the thicker samples required for the research and also refines the heating process so that more of the sample and less of the holder heats up, thereby localizing the heating results in more precise measurements.
VIDEO: Learn more about Camila López Pérez
Coauthors of the research include Anurag Maan, Shota Abe, Dennis Boyle, Dick Majeski and Tosh Le of PPPL; Ama Dahanayake and Martin Nieto-Perez of Pennsylvania State University; Camilo Jaramillo-Correa and Bruce Koel of Princeton University; and Alexandru Marin, affiliated with Pennsylvania State University and the Institute for Nuclear Research in Pitești, Romania. The work was supported by the U.S. Department of Energy under contract DE-AC02-09CH11466 and Fusion Energy Sciences awards DE-SC0021119 and DE-SC0021331.
