What We Publish

Utilizing curtailed wind and solar power to scale up electrolytic hydrogen production in Europe

Alissa Ganter, Tyler H. Ruggles, Paolo Gabrielli, Giovanni Sansavini, and Ken Caldeira · Environmental Science & Technology 59, 3495-3507 · 2025

Key finding. Across 27 European countries, hydrogen made from surplus wind and solar electricity could substitute 30% of fossil-derived hydrogen — 1.9 million tonnes a year — cutting ammonia and refinery emissions by 18%, or 20 million tonnes of CO2 a year.

Four maps of Europe in two columns. The left column is the fuel-saving scenario and the right the fuel-replacing scenario. The top row shades the share of hydrogen demand that can be met cost-effectively from 0 to 1, with Germany, Denmark and the Netherlands darkest; the shading is markedly paler throughout the fuel-replacing column. The bottom row shades the share of available surplus electricity actually used, which is high across most of continental Europe in the fuel-saving case and much lower in the fuel-replacing case.
The two columns are the same resource used two ways. Substituting electrolytic hydrogen whenever surplus happens to be available (left) serves a much larger share of demand and consumes far more of the surplus than committing to replace a fossil plant outright (right), because the second has to buy storage to deliver hydrogen on schedule. Figure 5 from Ganter et al. (2025), Environmental Science & Technology 59, 3495-3507. Reproduced under CC BY 4.0. Extracted from the published PDF and resized for web display.

What question did this research address?

As European wind and solar capacity grows, so does curtailment — generation thrown away because the grid cannot absorb or move it. Meanwhile hydrogen for ammonia and refining is made almost entirely from fossil fuels, and accounts for roughly 3% of global greenhouse gas emissions.

Those two facts suggest an obvious pairing, but surplus electricity is intermittent and hydrogen demand is not. This paper asked how much fossil hydrogen that surplus could actually displace, and what it costs to bridge the mismatch.

What did we find?

Surplus electricity potential is estimated from historical data for 27 European countries, then run through a cost-optimization that sizes and operates the electrolysis system, including the option of battery and hydrogen storage.

Two uses are separated. In a fuel-saving scenario, electrolytic hydrogen substitutes for fossil hydrogen whenever surplus electricity happens to be available. In a fuel-replacing scenario, it takes over entirely from a subset of fossil hydrogen plants.

Fuel-saving is the cheap win: 30% substitution, 1.9 million tonnes of hydrogen a year, and an 18% cut to ammonia and refinery emissions, amounting to 20 million tonnes of CO2 a year.

Fuel-replacing costs much more, because a plant that has actually shut down needs hydrogen on demand rather than when the wind blows — which means paying for battery or hydrogen storage to bridge the gap.

Even so, about 19% of fossil hydrogen production, 1.2 million tonnes a year, can be replaced outright and still be cost-effective.

Why does it matter?

It treats curtailment as a resource rather than a defect. Surplus generation is usually counted as waste to be engineered away; here it is an input whose zero opportunity cost is what makes the hydrogen competitive.

The gap between the two scenarios is the real finding. Substituting *when you can* is far cheaper than replacing *for good*, because firmness has to be bought with storage — the same intermittency problem that governs electricity, appearing again one step downstream.

It also bounds the ambition honestly. This displaces roughly a third of Europe's fossil hydrogen, not all of it, so surplus electricity is a useful head start on industrial decarbonization rather than a route to finishing it.

Citation

Alissa Ganter, Tyler H. Ruggles, Paolo Gabrielli, Giovanni Sansavini, and Ken Caldeira (2025). Utilizing curtailed wind and solar power to scale up electrolytic hydrogen production in Europe. Environmental Science & Technology 59, 3495-3507.

Related