In April 1951, crews began building a plant in Albany, Oregon, to separate zirconium and hafnium for the Atomic Energy Commission. A 1959 U.S. Bureau of Mines report by William A. Stickney says it was “patterned after a plant operated by the Carbide and Carbon Chemicals Corp. at Oak Ridge, Tenn.”
Seventy-five years later, chemists at Oregon State University in nearby Corvallis have patented a way to do that job in water, with no organic solvent. Their best run reached a separation factor of 33. The industry standard is 6 to 7.
It started with a suspected misprint
The Albany plant, Stickney wrote, “was the result of a chain of events that started with studies by A. R. Kaufmann at the Massachusetts Institute of Technology. Variations in the listed neutron-absorption cross-section of zirconium caused him to suspect a misprint.”
That cross-section tells physicists how readily an atom swallows a passing neutron. Oak Ridge scientists then measured “a value of 0.4 barn or lower,” the report says. “This meant that the potential of zirconium for reactor application was greatly enhanced.”
Hafnium spoiled the plan. “Since hafnium is invariably associated with zirconium and has a high neutron cross-section, it was necessary that an efficient hafnium-separation method be developed before the superiority of zirconium as a structural material in nuclear reactors could be realized,” Stickney wrote.
The Albany plant used hexone, also called methyl isobutyl ketone. In it, “hafnium was to be extracted preferentially as hafnium thiocyanate from an aqueous mixture of zirconyl and hafnyl chlorides,” by “a liquid-liquid countercurrent system.”
Two metals that come out of the ground together
The U.S. Geological Survey puts it simply: “Zircon is the primary source of all hafnium. Zirconium and hafnium are contained in zircon at a ratio of about 50 to 1.”
Zircon grains last. One from Western Australia’s Jack Hills is 4.4 billion years old. Zircon itself is “a coproduct or byproduct of the mining and processing of heavy-mineral sands for the titanium minerals, ilmenite and rutile, or tin minerals,” the USGS says.
The agency lists nuclear fuel cladding among zirconium’s uses. For hafnium, it lists “nuclear control rods, nickel-based superalloys, nozzles for plasma arc metal cutting, and high-temperature ceramics.”
Oregon State calls this “one of the most difficult separations on the periodic table, made challenging by the elements’ similarity.” The two “are so closely related” that zircon “almost always contains trace amounts of hafnium,” the university said.
“Naturally occurring Zr with ∼2% Hf must be separated to high levels of purity for both nuclear and semiconductor industries,” the study’s authors write. Zr and Hf are the chemical symbols for zirconium and hafnium.
The solvent problem in Albany and Ogden
Only two U.S. sites do the separation at industrial scale, according to Oregon State: ATI Specialty Alloys & Components in Albany and Westinghouse Electric in Ogden, Utah. Each uses “millions of pounds of flammable organic solvent annually.”
“Since the dawn of the nuclear age, this process is executed on an industrial scale by selective Hf-isothiocyanate (N═C═S) extraction into methyl isobutyl ketone (MIBK),” the authors write. “Yet SCN-MIBK mixtures yield toxic, noxious byproducts, also giving rise to performance variability.”
The researchers describe the plants’ liquid-liquid extraction as energy intensive. They note that it “results in roughly 4% of the solvent being lost to the air as noxious pollution.”
A food additive does the sorting
The Oregon State method uses precipitation instead, which the university describes as “dissolved ions with opposite charges attracting each other to form an insoluble solid.”
The mix is a water-based solution of natural zirconium, “with its few percentage points of hafnium impurity.” Thiocyanate ligands bind to the metal ions. Then comes choline, “an inexpensive, nontoxic chemical commonly used as a food additive.”
“Here, we eliminate the MIBK and instead precipitate Hf-rich species via inclusion of choline, a common food additive, in the aqueous phase,” the authors write.
What drops out is a hafnium-rich solid. Under a microscope, those precipitates grow as crystals of several shapes, some flat and square, some in rough clumps.
Graduate research assistant Alex Roseborough and May Nyman, the Terence Bradshaw Chemistry Professor at Oregon State, led the study. Roseborough now works as a metallurgist at the Department of Energy’s National Energy Technology Laboratory.
Nyman’s group makes metal-oxo clusters, small bundles of metal and oxygen atoms. “Besides being beautiful, they are useful for advancing science and technology,” her faculty page at Oregon State says.
Doctor Stephen, a Ph.D. student in materials chemistry, was a co-author, with Jack McLaughlin, Lev Zakharov, George Donkor-Gyami and Pere Miró.
Why 33 beats the industry’s 6 or 7
A separation factor rates how well a process splits two parts of a mixture. “A value greater than one denotes that separation is possible; the higher the number, the better,” Oregon State explains.
“The industry standard for zirconium-hafnium separation is between six and seven, and the OSU process produced a top score of 33,” the university said.
The team got there by adding sulfate, acid and ammonium. In the authors’ words, that “finely tunes Zr/Hf nuclearity and coordination,” where nuclearity means how many metal atoms sit in one molecule.
More acid made the molecules smaller. The study reports that it “decreases nuclearity from tetrameric to trimeric to dimeric to monomeric Zr/Hf complexes,” going from four metal atoms to one.
“In agreement with experiment, computation predicts increasing thermodynamic favorability for lower-nuclearity Hf species (monomer) compared to Zr species,” the authors write.
“We describe in atomic-level detail how the separation works and how precipitation-based separations can compete with solvent extraction,” Nyman said.
That is the paper’s main point. Precipitation-based separations “have been notoriously inferior to their solvent extraction counterparts,” the authors write, and this one is competitive.
It builds on a February 2025 paper in Angewandte Chemie International Edition by Roseborough, Nyman and colleagues. “Hf-selective precipitation from water introduces a ‘greener’ method for this Zr/Hf separation that is crucial for the nuclear industry,” that paper’s summary says.
Hafnium already sits inside computer chips
On Jan. 27, 2007, Intel announced new 45-nanometer transistors. About 400 of them “could fit on the surface of a single human red blood cell,” the company said.
Intel had shrunk the silicon dioxide insulator in its transistors “to as little as 1.2nm thick – equal to five atomic layers,” and current leaked through it. So “Intel replaced the silicon dioxide with a thicker hafnium-based high-k material in the gate dielectric, reducing leakage by more than 10 times.”
“The implementation of high-k and metal materials marks the biggest change in transistor technology since the introduction of polysilicon gate MOS transistors in the late 1960s,” Intel co-founder Gordon Moore said then.
Labs are now testing chip materials only three atoms thick. Oregon State says hafnium in its pure form is “crucial for both nuclear energy and semiconductor manufacturing.”
What still has to be proven
“We still have questions to answer and milestones to achieve, but these findings are really exciting and impactful, especially as society must move toward more carbon-free and high-density electricity generation, including nuclear energy,” Nyman said.
The university calls the process patented and low energy. It has not announced a pilot plant, an industrial partner, or measured savings in energy, cost or emissions.
A score of 33 is not a purity figure either. Reactor and chip makers need very pure metal in large amounts, and the announcement does not say how many rounds of precipitation that would take.
The method drops MIBK but still uses thiocyanate, sulfate, acid and ammonium, and that waste needs its own count. Getting valuable metals out cheaply and at scale is the hard part in other fields too, as with lithium and magnesium in seawater.
The full study was published in the Journal of the American Chemical Society.
The U.S. Department of Energy and the Murdock Charitable Trust supported the research.
Photo: Oregon State University (CC BY-SA 4.0)
