Oregon State Water Process Separates Zirconium and Hafnium at 33

Oregon State University chemists have patented a water-based way to separate zirconium from hafnium with a top separation factor of 33.

The industrial standard is 6 to 7, using methyl isobutyl ketone at the only two U.S. plants that do this job at scale.

Only Two U.S. Plants Do This Job

ATI Specialty Alloys & Components in Albany, Oregon, and Westinghouse Electric’s Western Zirconium works in Ogden, Utah, are the plants Oregon State names as the U.S. industrial pair. Each, the university said, uses millions of pounds of flammable organic solvent a year, and about 4% of that solvent is lost to the air as noxious pollution.

The Geological Survey’s 2026 mineral sheet matches the map. Zirconium metal and hafnium metal were made from chemical intermediates by one producer in Oregon and one in Utah.

THE TWO U.S. SEPARATION PLANTS

Those columns are the quiet gate for nuclear fuel cladding. They are also a gate for hafnium that later shows up in control rods, nickel superalloys, and the oxide powders chip plants buy. A beaker method that skips the ketone does not automatically replace that hardware. It does show that the hardware is the constraint, not the ore.

Zircon Carries Both Metals at 50 to 1

The metals start together. Zircon, the silicate mineral ZrSiO4, is the main commercial source of both, and the Survey puts zirconium and hafnium in that mineral at a ratio of about 50 to 1. The Journal of the American Chemical Society paper on the Oregon State work states the same fact another way: natural zirconium carries about 2% hafnium.

They are hard to split because they are almost the same ion in water. Reactors need the opposite of that mixture. Zirconium cladding has to let neutrons pass; hafnium soaks them up, which is why the same element is useful in control rods and a problem inside a fuel tube.

Chip makers want the hafnium side as oxide, not as a leftover in zirconium metal. The Survey’s metal-use list is narrower. It names superalloys as the leading use of hafnium metal, with control rods, plasma-arc nozzles, and high-temperature ceramics close behind. The oxide path into gate stacks is a different product line that still depends on the same first split.

U.S. ZIRCONIUM AND HAFNIUM IN 2025

• World mine output: Zirconium mineral concentrates are estimated at 1.2 million metric tons, down from 1.3 million tons in 2024, per the Geological Survey table.
• Hafnium metal trade: Unwrought U.S. imports are estimated at 72 metric tons, with exports at 15 tons, so the country is a net importer of that metal.
• Zirconium metal trade: Unwrought imports are estimated at 530 metric tons, and the Survey says the United States is a net exporter of zirconium metal.
• Where the sand comes from: For 2021 through 2024, zirconium ore and concentrate imports were 48% from South Africa, 35% from Australia, 15% from Senegal, and 2% from other countries.

Hafnium unwrought metal still arrives mainly from Germany, China, and France. The two domestic plants do not close that gap on their own.

Precipitation Without Methyl Isobutyl Ketone

The industrial route, used since the nuclear age, pulls hafnium isothiocyanate into methyl isobutyl ketone. The Oregon State mix keeps thiocyanate, which binds both metal ions, and drops the ketone. Choline, a cheap food additive, goes into the water instead. Opposite charges then lock into a solid, and the solid that falls is rich in hafnium.

Sulfate, acid, and ammonium are the knobs. Acid cuts the clusters down from tetramers toward trimers, dimers, and monomers, swapping oxyhydroxide groups for more thiocyanate. The paper names two hafnium-rich complexes as the ones that matter, including a mixed thiocyanate-sulfate anion.

WHAT GOES INTO THE WATER MIX

• The feed: Natural zirconium, with its few percentage points of hafnium, dissolved as ions.
• The binder: Thiocyanate ligands that grab both zirconium and hafnium.
• The precipitant: Choline, described by the university as inexpensive and nontoxic.
• The tuners: Sulfate, acid, and ammonium, used to set cluster size and coordination.

Graduate researcher Alex Roseborough led the study with May Nyman, the Terence Bradshaw Chemistry Professor at Oregon State. Also on the paper were Doctor Stephen, Jack McLaughlin, Lev Zakharov, George Donkor-Gyami, and Pere Miró. The Department of Energy and the Murdock Charitable Trust paid for the work. Roseborough is now a metallurgist at the department’s National Energy Technology Laboratory.

We describe in atomic-level detail how the separation works and how precipitation-based separations can compete with solvent extraction.

May Nyman, Terence Bradshaw Chemistry Professor, Oregon State University

The atomic-level account of the precipitation is in that Journal of the American Chemical Society article, received May 24, 2026, accepted August 31, and published September 7, 2026. Computation in the study finds lower-nuclearity hafnium species more favored than the zirconium analogues, which matches the solids they isolated.

What a Separation Factor of 33 Means

A separation factor above 1 means two components can be pulled apart; a higher number means a cleaner cut per step. Oregon State says the industry standard for this pair is 6 to 7, and the water process posted a top score of 33.

SEPARATION FACTOR, LAB VERSUS PLANT STANDARD

That gap is why the university talks about precipitation competing with solvent extraction at all. It is not a plant throughput number, a cost per kilogram, or a certificate of nuclear-grade purity. It is a single-step selectivity score, measured on the solids that fall when choline is added and the cluster size is tuned.

Older solvent patents show how tightly the incumbent method is specified. One granted method keeps an aqueous feed of zirconium and hafnium oxychlorides with a thiocyanate salt and methyl isobutyl ketone, and holds a narrow acid-to-metal window so the extractant behaves. The Oregon State work leaves that organic layer out. It does not leave thiocyanate chemistry behind.

Albany Made Nuclear Zirconium in 1956

The Albany line is older than commercial nuclear power as most people met it. Trevor Giroux, president of ATI’s Specialty Alloys and Components division, has said the site produced the world’s first nuclear-grade zirconium product on Christmas Day 1956, work that began under Admiral Hyman Rickover and later ran through Wah Chang before ATI. Giroux also described the tall extraction buildings as the place where the 50-to-1 ore mix is cut apart, with zirconium wrapping fuel and hafnium acting as the brake.

Western Zirconium in Ogden has made zirconium and hafnium products since 1978, according to the plant’s own materials. A Utah air-permit file for that site walks through feed mix tanks, three hafnium extraction columns, thio-removal columns, and scrubbers, the classic American MIBK flowsheet.

FROM THE FIRST SPONGE TO THE WATER PAPER

1. December 25, 1956: Albany produces what ATI’s Giroux calls the first nuclear-grade zirconium product.
2. 1978: The Western Zirconium plant in Ogden begins metal production.
3. Nuclear-age plant practice: Selective hafnium-isothiocyanate extraction into methyl isobutyl ketone becomes the U.S. industrial method described in the new paper.
4. 2025: Roseborough, Nyman, and coworkers show that Hf-rich oxo clusters selectively precipitate from water when the counterion is pushed from ammonium toward tetramethylammonium, a solvent-free split built on the same industrial clusters.
5. September 7, 2026: The choline paper appears in the Journal of the American Chemical Society, raising the precipitation score to 33.

The 2025 cluster study already argued that mixed Zr/Hf tetramers form instead of separate pure clusters, and that hafnium-rich clusters fall first. The 2026 paper adds choline, sulfate, and acid as the tools that push that effect to a factor of 33.

Nyman Still Has Milestones to Hit

Oregon State’s own release is careful on the next step. Nyman said the group still has questions to answer and milestones to achieve, even as she tied the work to carbon-free, high-density electricity, including nuclear power. That is the right scale of claim. A separation factor of 33 in water is a chemistry result. The plants in Albany and Ogden are steel, solvent inventories, permits, and decades of raffinate recycle.

Thiocyanate stays in the recipe, so the green claim is the missing ketone, not a reagent-free process. Throughput, recycle of choline, residual sulfur, and the extra stages needed to reach cladding-grade zirconium are still open. The university says the process is patented; no patent number was attached to the announcement.

Roseborough’s move to the National Energy Technology Laboratory puts one of the authors inside the energy department that funded the study. The two solvent plants remain the U.S. path from zircon to nuclear metal. The paper gives them a water-based alternative to test, not a schedule for tearing down the columns.