Duranium Is Reshoring Critical Metals, and the Math Actually Works

The Macro: America Forgot How to Make Metal

I want you to try something. Go look up where the United States gets its magnesium. Not the magnesium in your multivitamin. The magnesium that goes into aircraft frames, missile casings, automotive parts, and the lightweight alloys that the defense industry cannot function without. The answer is going to make you uncomfortable.

Three foreign producers control 97% of global magnesium supply. The situation is nearly identical for titanium sponge, where a handful of countries dominate 96% of production. The United States used to be a major producer of these metals. We are not anymore. The last domestic magnesium electrolyzer shut down over twenty years ago.

This is not a hypothetical supply chain risk. It is an active one. When China temporarily restricted magnesium exports in 2021, European automakers scrambled. Prices spiked. Production lines slowed. The entire downstream manufacturing ecosystem discovered, in real time, how dependent it had become on a single source.

The cost gap is the reason nobody fixed this. U.S. magnesium production costs roughly $7,700 per ton. The global price hovers around $2,900. That is not a margin you can subsidize away with tariffs or tax credits. The fundamental production economics are broken because American metallurgy has not been meaningfully updated in decades.

There have been attempts to address this. Companies like Western Magnesium and Alliance Magnesium have pursued various approaches, from electrolytic processes to thermal reduction. None have achieved commercial-scale production at competitive pricing. The gap between “technically possible” and “economically viable” has swallowed a lot of investor capital in this space.

The Department of Defense knows this is a problem. The Department of Energy knows this is a problem. Congress has held hearings about it. Reports have been written. And yet the number of operational domestic magnesium production facilities remains the same: zero.

The Micro: A McKinsey Consultant and a Metallurgist Walk Into a Reactor

Brenden Prins-McKinney spent his time at McKinsey advising the Department of Defense on exactly this problem: how to reshore critical chemical production. He is a Stanford Law graduate and Knight-Hennessy Scholar. His co-founder, Berkley Noble, built what Duranium describes as “the only U.S. pilot magnesium electrolyzer in the last 20+ years.” Noble holds a patent for an electrochemical metal cell. The team is four people, based in Alameda, California. They came through Y Combinator’s Summer 2025 batch.

The technical approach is what makes Duranium interesting rather than just another reshoring pitch. They are modernizing carbochlorination metallurgy, which is the chemical process of converting metal oxides into chlorides using chlorine and carbon. That sounds incremental. It is not. Their reactor creates a closed-loop system that recycles CO emissions back into the production cycle as feedstock. The byproduct is hypochlorite, a bleaching agent with its own commercial market. So instead of paying to dispose of waste, they sell it.

This is the part that matters: Duranium claims they can achieve Chinese-level pricing without subsidies. That claim lives or dies on the closed-loop economics. If the CO recycling works at scale and the hypochlorite revenue offsets enough of the operating costs, the math could actually pencil out. That is a big “if,” but it is an engineering “if,” not a physics “if.” The chemistry is sound. The question is whether it works at production volumes.

Their pilot facility is targeting operational status by June 2026. The projections from there are ambitious: supplying 12% of U.S. magnesium demand by 2029 at 50%+ EBITDA margins. Those are the kinds of numbers that sound too good until you remember that the domestic competition is literally nobody. When the baseline is zero percent market share for all American producers combined, capturing twelve percent is less about beating competitors and more about proving the process works.

The market opportunity extends beyond magnesium. Duranium’s reactor technology applies to titanium, aluminum, zirconium, and hafnium. Each of those metals has its own supply chain vulnerabilities and its own set of defense and industrial applications. If the core process scales for magnesium, the expansion path is clear.

The Verdict

I think Duranium is one of the most important companies to come out of YC in recent batches, and I do not say that because reshoring is trendy or because defense tech is having a moment. I say it because the unit economics story is specific, testable, and grounded in real chemistry rather than handwaving about national security premiums.

The risk is execution. Scaling a novel metallurgical process from pilot to commercial production has killed plenty of companies with good science. The jump from “it works in our facility” to “it works at thousands of tons per year” is where capital requirements explode and unexpected engineering problems emerge.

Thirty days, I want to see progress on the pilot facility timeline. Sixty days, whether they have secured offtake agreements with defense contractors or industrial buyers. Ninety days, the question is whether the closed-loop economics hold at pilot scale or whether the CO recycling efficiency degrades in ways that change the cost picture. If the June 2026 pilot delivers the numbers they are projecting, Duranium will have no trouble raising whatever capital it needs to go big.