A stainless steel built to survive inside a hydrogen electrolyzer costs about 40 times less than the coated titanium it replaces, while roughly a fifth of the alloy is cobalt

Inside a working electrolyzer the environment is hot, acidic and saturated with oxygen.

Metal sitting in that atmosphere is held at a voltage that would strip an ordinary protective film off almost anything.

Titanium survives it. That is why titanium is in there, usually with a thin coating of gold or platinum on top.

The result works because the manganese oxide is doing a job the chromium oxide cannot do at that voltage, not because the old objection was wrong at lower voltages.

Both of those are expensive, and both are a reason green hydrogen equipment costs what it does.

A team in Hong Kong made a steel that holds.

The price gap runs near fortyfold.

The trick is a second protective layer that takes over from the first

Stainless steel resists corrosion because chromium in the alloy grows a thin oxide film that seals the surface.

That film is tough within a window and useless outside it. Push the voltage high enough and the chromium oxide itself starts to dissolve, which metallurgists call transpassive behavior.

Conventional grades give up somewhere around 1,000 millivolts. Splitting water demands roughly 1,600, so ordinary stainless has never been a candidate for the job.

The alloy in question does something different. As the chromium film begins to fail, a second film built on manganese forms over it, starting near 720 millivolts.

The handover carries the surface to about 1,700 millivolts in chloride bearing water, past the threshold that stopped every previous attempt.

One film fails while another takes hold.

What manganese was supposed to ruin

The odd part is that manganese has a bad name in this field.

For decades it was treated as something to minimize, because manganese sulfide inclusions are classic starting points for pitting corrosion in steel.

Loading an alloy with it and expecting better corrosion behavior runs against a century of practice, which is roughly why nobody had tried.

The result works because the manganese oxide is doing a job the chromium oxide cannot do at that voltage, not because the old objection was wrong at lower voltages.

That is a narrow and specific claim, and it is the reason the work got published rather than filed away.

The objection held everywhere else, just not up there.

The line in the composition that nobody quotes

Here is what the coverage leaves out, and it undercuts the cheapest part of the pitch.

As the composition has been reported, the alloy runs to about 20 percent chromium, roughly 18 percent manganese, a little silicon, and about 20 percent cobalt.

Cobalt is not a commodity the way iron is. It is a critical raw material, mined overwhelmingly in one central African country and refined overwhelmingly in one other country.

Claims that no exotic supply chain is involved, or that the raw material cost starts low because iron ore is abundant, do not survive a fifth of the alloy being cobalt.

The savings against gold coated titanium may still be real. The story about a humble everyday metal is not.

Titanium has one problem. This has a different one.

What the forty times figure is actually measuring

The number is a material cost comparison made by the researchers, not an independently audited bill of materials for a finished stack.

It matters anyway, because structural parts are not a rounding error in this equipment.

The team puts structural components at as much as 53 percent of the cost of a 10 megawatt system, which they price at roughly 2.3 million dollars.

Halve that share and the effect on installed cost per kilowatt is larger than most catalyst improvements deliver, since catalyst loading is a small slice of the total.

Catalyst work runs on a separate track, as with the iron based cathode that set a record while avoiding platinum entirely.

The passivation mechanism, the voltages and the cost estimate are described by a technology outlet.

Structure is the expensive half, not the catalyst.

How far this is from a component you could buy

The underlying paper is not new. It appeared in a materials journal in 2023, and the wave of attention in August was a reissue rather than a fresh result.

Production so far means wire. The team reports tons of it made with a mainland factory, which is a real manufacturing step and not the same as a porous transport layer or a bipolar plate.

Electrolyzers want meshes, foams and pressed plates held flat under load in hot water, and turning an alloy into those shapes is where promising metals usually stall.

No long duration data at stack scale has been published either. Brief electrochemical tests have flattered marginal materials before.

Commercial hardware moves on a different clock, which is what an Ohio maker demonstrated by shipping its first unit to Antwerp.

The composition, the journal and the scale up claim are set out by a science service.

The stainless steel works in a beaker, and the question is whether it survives a pressed plate and a purchase order.