Oregon State’s New MOF Makes Hydrogen Without Extra Platinum

Oregon State chemists report a zinc metal-organic framework that made 938 micromoles of hydrogen per hour per gram without an added platinum catalyst. The powder, BVR-19-Zn, uses a sulfur-sulfur bond in an amino-acid linker to catch light and move electrons. Kyriakos Stylianou, who directs the university’s Materials Discovery Laboratory, said the organic part does the work that most of these crystals assign to metal atoms.

The same briefing put hydrogen from methane-steam reforming at $1.50 a kilogram and green hydrogen at about $5. Light-driven hydrogen has been chasing that gap since 1972.

The Sulfur Bond That Breaks in Light

BVR-19 is a family of metal-organic frameworks, crystals in which metal ions sit in a lattice of organic linker molecules. Stylianou said chemists have already made almost 100,000 of these structures, predicted properties for about another 500,000, and can imagine millions more. For this study the linkers are L-cystine, an amino acid that already carries a disulfide bridge, and the metals are magnesium, manganese, copper, zinc, and cadmium.

The paper, published online September 14, 2026 in the Journal of the American Chemical Society and collected in volume 148, issue 37, calls the family BVR-19-M. First author Emmanuel Nyela Musa and the Corvallis group report that light can temporarily split the sulfur-sulfur bond, giving a thiyl radical and a thiolate anion that help separate charge. Zinc’s filled d-orbitals favor that path. Open-shell metals either open a mid-gap recombination site or pull the chemistry back onto the metal, and those versions lag.

Instead of relying primarily on the metal atoms, our material uses its sulfur-containing organic building blocks to capture light energy and move electrons where they are needed to produce hydrogen. This represents a different way of thinking about how these materials should be designed.

Kyriakos Stylianou, OSU College of Science

The intraligand charge transfer mechanism is the claim the journal is actually selling. The zinc ions hold the lattice; the cystine units absorb light and shuttle electrons. The crystals form on their own in deionized water at room temperature, which cuts the heat bill that many MOF recipes run up in ovens and solvents.

Intraligand Charge
Transfer in Metal-Organic Frameworks
Facilitates Radical Anion-Mediated Hydrogen Evolution https://t.co/7bdLYELvMO

— J. Am. Chem. Soc. (@J_A_C_S) September 15, 2026

Zinc Led the Rate Table, Copper Barely Moved

The zinc crystal was the clear winner in the hydrogen-evolution tests. Cadmium, another closed-shell ion, came next. Copper, manganese, and magnesium trailed by a wide margin, which is how the group argues that the metal’s electron configuration, not just the presence of a disulfide, sets the rate.

HYDROGEN EVOLUTION RATES IN THE BVR-19 FAMILY

Adding nickel phosphide, a cheaper stand-in for the platinum dots that litter this literature, raised the zinc rate from 938 to 1,223 micromoles per hour per gram, about 30 percent. That small bump is the design working as advertised: the linker already does most of the hydrogen-making, so an extra metal co-catalyst is optional rather than required. Stylianou said skipping an expensive extra metal could simplify future light-driven systems.

The zinc powder was run through six straight two-hour cycles and stayed above 350 micromoles per hour per gram. After long irradiation the long-range crystal order faded even though the local chemistry of the cystine linker held, a durability problem the authors flag rather than bury.

Photocatalytic Hydrogen, 54 Years After Honda

In July 1972 Akira Fujishima and Kenichi Honda reported photoelectrolysis of water in 1972 on a titanium dioxide electrode paired with platinum, using a xenon lamp in an electrochemical cell. The short Nature letter became the Honda-Fujishima effect, and it launched a field that still lives mostly in flasks. Fifty-four years later, BVR-19-Zn is another material paper in that line: a new way to separate charge, tested as hydrogen evolution, written up as a design rule.

LIGHT-DRIVEN HYDROGEN SINCE 1972

1. July 1972: Fujishima and Honda publish water photolysis on a titanium dioxide anode in Nature, volume 238, pages 37 to 38.
2. 1975: The same pair, with Kohayakawa, show hydrogen under sunlight with titanium dioxide and platinum, using a pH difference instead of an applied bias.
3. September 14, 2026: The BVR-19-M study goes online at JACS, with zinc leading a five-metal series and no required co-catalyst.
4. September 28, 2026: Oregon State releases the work as a path to cheaper green hydrogen.

Kazunari Domen, who has spent decades on powder photocatalysts at Shinshu University and the University of Tokyo, has described a 100-square-meter panel array that used aluminum-doped strontium titanate and stayed below 1 percent solar-to-hydrogen efficiency because the material absorbs mainly ultraviolet light. He has also written that a practical system needs at least 5 percent, and ideally about 10 percent. BVR-19-Zn does not report a solar-to-hydrogen efficiency. It reports a lab rate, in micromoles per hour per gram, which is how this literature keeps score long before anyone builds a field.

Factories Still Buy Almost Every Kilogram

The Oregon State release leads with fuel cells for cars, then lists ammonia, metal refining, and plastics. The market still runs in the other order. The International Energy Agency says global hydrogen demand surpassed 100 million tonnes in 2025, almost all of it in industry and refining. Low-emissions supply grew 20 percent that year and reached close to 1 million tonnes, still about 1 percent of the pile.

HYDROGEN DEMAND IN 2025

• Total use: Demand grew almost 3 percent in 2025 and moved past 100 million tonnes, the IEA’s Global Hydrogen Review 2026 finds.
• Clean slice: Low-emissions production reached close to 1 million tonnes after a 20 percent rise.
• Old jobs: In 2024, refining, ammonia, methanol, and fossil-based iron still took almost all demand; new uses stayed under 1 percent.
• Vehicles: Fuel-cell vehicle stock grew 20 percent in 2025 to almost 130,000, with China trucks and a Korea car rebound doing the work.

If BVR-19 ever leaves the beaker, the first buyers will not be drivers. They will be ammonia plants, refineries, and steel shops that already burn hydrogen as a chemical, not a fuel. Those users already have cheap grey hydrogen on site. A powder that needs sunlight, water, and a donor chemical has to beat that incumbent on cost and uptime, not on a press-release sentence about cars.

What the Oregon State Tests Left Out

The JACS paper is a hydrogen-evolution study. That is the half-reaction that makes H2. It is not overall water splitting, which also has to oxidize water to oxygen and keep the two gases apart. The proposed mechanism in the Oregon State figure has electrons from a sacrificial donor driving hydrogen at the disulfide site, which is the standard lab bargain in this field: feed the holes a chemical so the electrons can make hydrogen on a clock the authors can publish.

WHAT THE BEAKER TEST DID NOT DO

• Both gases: The work tracks hydrogen evolution, not a 2:1 hydrogen-to-oxygen split from pure water.
• No donor: The published mechanism still uses a sacrificial donor to supply electrons, so the test is not a closed sunlight-and-water device.
• Long life: Six two-hour cycles kept rates above 350 micromoles per hour per gram, and crystal order still slipped after long light exposure.
• A plant price: Nothing in the paper converts 938 micromoles per gram into a dollar-per-kilogram figure against steam reforming.

Steam reforming still sets the floor Stylianou himself cited, at about $1.50 per kilogram, against about $5 for green hydrogen. On those two figures, the clean product costs more than three times the dirty one. The U.S. Department of Energy has also noted that steam methane reforming emits about 10 tons of carbon dioxide per ton of hydrogen, which is the climate reason to care. Electrocatalysis, the other water route, only stays clean if the electricity is clean and cheap. Photocatalysis tries to skip the electrolyzer and the grid. That is the appeal, and it is also why every new powder gets billed as a cost cut before anyone has run a reactor in the sun for a year.

Posts that circulated the Oregon State release treated BVR-19 as a way to retire steam reforming. The zinc result does not do that. It shows that a cystine linker can carry the photochemistry so the crystal does not need a platinum patch, and that zinc and cadmium beat copper and manganese in this lattice. That is a materials finding. The $3.50 gap between $5 green hydrogen and $1.50 grey hydrogen is still an engineering and electricity problem.

Stylianou’s Design Rule Stops at the Beaker

Stylianou said the work gives a blueprint: change the metal, keep the rest of the crystal, and learn why some versions work. The rule that falls out is specific. Closed-shell ions and a redox-active disulfide favor ligand-centered charge transfer. Open-shell ions waste the light. Make the material in water at room temperature if you can. Do not assume you need a second expensive metal to evolve hydrogen.

By changing the metal while keeping the rest of the material essentially the same, we discovered why some versions of the MOF work much better than others. These findings provide new design rules for creating more effective materials for solar fuel production.

Kyriakos Stylianou, director of OSU’s MaD Lab

The Murdock Charitable Trust, the National Science Foundation, and the OSU College of Science funded the study. The U.S. Department of Energy’s Hydrogen Shot still aims for clean hydrogen at $2 per kilogram by 2026 and $1 per kilogram by 2031, targets written for electrolyzers as much as for any powder. BVR-19-Zn does not land on those numbers. It lands on 938 micromoles an hour, a 30 percent optional nudge from nickel phosphide, and a sulfur bond that breaks in the light and, for now, a crystal that slowly loses its long-range order when the lamp stays on.

Steam reforming still sells hydrogen at about $1.50 a kilogram. The zinc framework is a design rule in a beaker, and the next crystal that matters will have to keep its lattice in sunlight and make oxygen as well as hydrogen.