Research and development work continues to improve a proven method for storing high-level radioactive waste, and explore other applications for the future.

During the Cold War, the United States produced materials for nuclear weapons by extracting uranium and plutonium from materials assemblies irradiated in government-owned nuclear production reactors. The extraction process left behind a highly radioactive liquid waste, commonly known as sludge. Savannah River Site (SRS), in Aiken, South Carolina, is home to the Defense Waste Processing Facility (DWPF), the world’s largest operation dedicated to disposing of radioactive waste in the safest known manner. “The DOE Office of Environmental Management is charged with remediating the legacy waste left from the defense nuclear mission,” said John Marra, associate laboratory director at the Savannah River National Laboratory (SRNL), the environmental office’s corporate lab that developed the processing facility’s waste remediation methods.

Approximately 37 million gallons of radioactive waste are stored in the SRS tank farms. Of that, all the sludge and the high-activity portion of the “salt” waste are sent to DWPF to be converted into a stable glass product by processing it with frit, a tailored glass crushed to the consistency of sand. Through this process, called vitrification, the facility has already poured more than 10 million pounds of glass containing more than 16 million curies since operations began in 1996.

“The waste is blended with the frit and the mixture is processed through a high temperature melter—nominally 1,150 degrees Celsius—to produce a homogeneous borosilicate [silica and boric oxide] glass,” said David Peeler, a research scientist in the lab’s glass formulation group. “It becomes molten and is then poured into stainless steel canisters.”

As the newly created mixture hardens in the two feet by ten feet canisters, it becomes a glossy black glass that has chemically bonded with the radioactive species. This finished glass product is the result of complex calculations, testing and retesting. The frit composition is at the heart of a successful effort.

The waste glass must meet a multitude of criteria. Constraints include viscosity (how thick or thin the glass is) and the “liquidus” temperature (the temperature above which the glass is completely melted). If the glass’ viscosity is too low, it can corrode parts in the melter; too high and it could slow down the melt rate and in turn slow down the mission. Viscosity also can affect the glass’ ability to pour into the canister.

Of course, not all sludge is the same. So the glass formulation team must adjust the frit composition for each batch of waste. Carol Jantzen, an internationally known ceramics expert at the laboratory, developed models that predict glass properties such as durability, liquidus temperature and viscosity based on glass composition.

Personnel at the tank farms work with the lab to determine which tanks of sludge are due for processing next. Using Jantzen’s models, the glass formulation group identifies frit candidates to combine with the next batches of sludge. “We as a lab utilize those models and, working a year or two ahead of the processing facility, evaluate different tank blending strategies for the next sludge batch,” said Peeler.

After the tank farm and lab scientists agree on a blending strategy, SRNL’s glass formulation team enters an experimental phase. To validate model predictions, the laboratory produces and characterizes nonradioactive glasses using simulated sludge, and radioactive glasses to ensure measured properties are in line with the model predictions. The team also runs melt-rate tests to see which candidate frit will convert the waste into glass fastest. Based on model predictions, glass characterization and meltrate testing, the laboratory recommends a frit suitable for processing a specific sludge batch. A separate SRNL team receives samples of radioactive sludge from the tank farms and runs that sample through the entire process to prove its viability before the sludge is moved to the waste processing facility.

The remediation mission is scheduled to run into the late 2020s. With every batch of sludge, the teams at Savannah River fine-tune the process with an eye on minimizing mission life, and in turn, minimizing costs. One popular term among the scientists and management is “waste loading,” getting as much waste into each canister as possible. “If you have 4,000 pounds of glass in a canister and you’re at 25 percent waste loading, you have approximately 1,000 pounds of waste and 3,000 pounds of frit in each canister,” said Peeler. “Targeting a higher waste loading will yield fewer canisters for a given sludge batch.”

DOE eventually will store the canisters in a national repository, where canister storage space will be limited. SRNL’s model development and the laboratory’s refinements to the glass formulation, along with other collaborations between SRNL and DWPF, have enabled the facility to process specific sludge batches with as much as 38 percent waste loading, up from 25 to 28 percent when the mission started.

Developments at SRNL have applications that extend beyond this remediation effort. “If you look at a country like France that uses nuclear power, as well as Japan and the United Kingdom, they reprocess spent nuclear fuel,” Marra said. “The technology has direct application to commercial reprocessing. Also, the vitrification process has been used for low-level nuclear waste and other difficult-to-deal-with waste.”