Green hydrogen’s big promises, yet to match reality
By Professor Dato Dr Ahmad Ibrahim
The technology works in the lab. Whether it works as a business is still an open question.
Many have probably heard the pitch: green hydrogen will clean up steel, shipping, and fertilizer. Made by splitting water with renewable electricity, it burns clean, with water as its only “exhaust.” Governments are pouring billions into it. Companies are racing to build factories.
A new scientific review in the journal Energies takes stock of where this technology actually stands. The authors, led by Rafika Louli at the University of Technology of Belfort-Montbéliard in France, did a careful job. But reading between the lines, a less comfortable story emerges: much of what we “know” about cheap green hydrogen comes from small lab tests and computer models that have never been proven in the real world.
Four technologies, four sets of problems.
There are four main ways to split water into hydrogen and oxygen.
The oldest, called alkaline electrolysis, has been used in factories for about a century. It’s reliable. But it’s slow to start up and shut down—a real problem when your electricity comes from sun and wind, which come and go.
The newer favorite, PEM, responds quickly and works at high power. But it needs iridium and platinum, two of the rarest metals on Earth. You can’t build millions of these things if the ingredients barely exist.
Solid oxide systems are the most efficient on paper—up to 95%. But that number only holds if you feed them a constant supply of high-quality waste heat. In a real factory, that heat usually has other jobs. The efficiency claim quietly assumes a free lunch.
The fourth, AEM, sits somewhere in the middle. It’s promising but not ready. Its membranes wear out, and scientists don’t yet agree on how to predict how long they’ll last. That’s a big deal for anyone signing a 20-year contract.
The cost numbers are all over the map.
Here’s where things get messy. The standard way to compare energy costs is something called LCOH—the levelized cost of hydrogen. It’s supposed to tell you the true cost per kilogram over a project’s lifetime.
The problem? There’s no single agreed way to calculate it. Different methods give different answers. A review of 96 studies found solar-powered hydrogen costs ranging from $3.70 to $12.50 per kilogram. That’s more than a threefold spread. Most of that gap comes from assumptions—where you build, how you finance it, how often the equipment runs—not from the technology itself.
So when someone says “green hydrogen is getting cheap fast,” ask: cheap compared to what, calculated how, and proven where?
The experiment nobody talks about
Here’s the finding that should worry the industry.
A real-world study of a 30-megawatt solar-and-electrolysis system at China’s Yumen Oilfield found that simply running the equipment smarter—matching it to the sun’s rhythms and electricity prices—cut costs by 21% to 51% compared to standard planning.
That’s a huge range. And it points to something important: most published cost estimates assume the system runs steadily, the way a coal plant does. Real renewable power doesn’t work that way. When you account for the actual stop-start reality, the numbers can shift dramatically.
In other words, many of the cost figures floating around in reports and press releases may be wrong—not by a little, but by a lot.
A lab result that’s too good to be true for now.
One recent study made headlines: researchers produced hydrogen for under $1 per kilogram by swapping in a different chemical reaction. Sounds like a breakthrough.
But look closer. The test cell was 25 square centimeters—smaller than a credit card. Industrial systems are several square meters. Scaling up that gap is the graveyard of countless clean-tech dreams. Lots of clever chemistry works beautifully in a beaker and fails completely in a factory.
Without a pilot plant proving it works at size, the “$1 per kilogram” claim is a promise, not a result.
What this means for you.
Green hydrogen may well become a major clean energy tool. The chemistry is sound. The need is real.
But the confident cost projections you see in headlines rest on shaky foundations: lab data, optimistic models, and assumptions that haven’t survived contact with actual projects. The technology works. The business case is unproven.
So the next time you hear that green hydrogen is about to be cheap, ask three questions. Has it been built at scale? Has it run for years, not hours? And has anyone published the real numbers?
Until the answers are yes, treat the hype with caution. The promise is real. The proof isn’t here yet.
The author is affiliated with the Tan Sri Omar Centre for STI Policy Studies at UCSI University and is an Adjunct Professor at the Ungku Aziz Centre for Development Studies, Universiti Malaya. He can be reached at ahmadibrahim@ucsiuniversity.edu.my.