Originally published December 28, 2010. Updated July 25, 2026.
Hydrogen fuel cells only cut carbon emissions if the hydrogen feeding them is clean, and less than 1% of the world’s hydrogen supply qualifies as low-emissions today. The world used roughly 100 million tonnes of hydrogen in 2024. Fuel-cell cars haven’t caught on either. US drivers bought 3,143 hydrogen cars in 2023, against 380,000 battery-electric cars the same year. Federal hydrogen funding kept flowing through early 2025, before a wave of cancellations hit clean-energy awards generally, and a 2025 tax law moved the deadline for hydrogen’s biggest tax credit five years earlier.
How much hydrogen the world makes
Global hydrogen demand reached roughly 100 million tonnes in 2024. Almost none of it helps the climate. Under 1% of that supply counted as low-emissions, the kind made without adding the usual fossil carbon to the total. The clean share is growing, even if slowly. Low-emissions hydrogen production climbed about 10% from 2023 to 2024, putting 2025 on track for close to 1 million tonnes. That’s still a small fraction of total demand, and the rest still comes from the same natural gas and coal that dominate hydrogen production today. At hydrogen’s standard heating value of 33.3 kilowatt-hours per kilogram, that 100 million tonnes holds roughly 3,330 terawatt-hours of energy in a single year’s global supply. We’d call the low-emissions share the number that decides whether a fuel cell in a car or a power plant counts as a climate technology or a different way to burn gas.
Where most of the world’s hydrogen comes from
Producing hydrogen the standard way used 290 billion cubic meters of natural gas and 90 million tonnes of coal on a coal-equivalent basis in 2024. That’s grey hydrogen, split out of fossil fuels, and it still dominates the market. Green hydrogen works differently, splitting water with electricity in machines called electrolyzers. Global electrolyzer capacity reached about 2 gigawatts in 2024, with more than 1 more gigawatt added by July 2025. Most of that equipment comes from one country. China holds 65% of the world’s installed electrolyzer capacity and roughly 60% of global manufacturing capacity for the machines. Even with that growth, fossil-based production still supplies most of the world’s new hydrogen.
The hydrogen project pipeline shrank in 2025
The pipeline of announced hydrogen projects shrank for the first time instead of growing. Companies had floated projects that could produce up to 37 million tonnes of hydrogen a year by 2030, down from a 49-million-tonne estimate the year before. Most of that total exists only on paper so far.
| Pipeline stage | Hydrogen output by 2030 |
|---|---|
| Announced projects | 37 million tonnes a year |
| Share that reached a final investment decision | 9% |
| Realistic estimate from projects running or funded | 4.2 million tonnes a year |
Source: IEA, Global Hydrogen Review 2025
A final investment decision is the point where a company commits the money to build. Only 9% of the announced pipeline had reached that point as of the 2025 count, which is why the realistic 2030 total comes in at 4.2 million tonnes a year, a little over a tenth of what got announced.
Fuel-cell cars and hydrogen stations never caught on in the US
The US has roughly 44 public hydrogen stations, and 42 of them are in California. Nationally, 38 of those are retail stations and 6 serve fleets or research, and in California the split is 37 retail and 5 non-retail. The rest of the country has two. If you’ve ever filled up at a hydrogen pump, it was almost certainly there. Toyota’s Mirai sold 1,702 units worldwide through November 2024. Ownership hasn’t been trouble-free. Hyundai recalled all 1,600 Nexos it had sold in the US as of 2024, over a fire risk, and filling up cost $36 a kilogram at California stations as of September 2023. For the fuller picture on the vehicle that’s outselling it, see our explainer on electric vehicle adoption. Hydrogen isn’t the only alternative fuel that stalled out in the US. Our explainer on natural gas vehicles covers the other one.
What happened to the US hydrogen hubs program
The Bipartisan Infrastructure Law set aside up to $7 billion of a roughly $8 billion program to fund 6 to 10 regional hydrogen hubs. The last two of those hubs, in the Heartland and Mid-Atlantic regions, were awarded on Jan. 17, 2025. Then the cuts began. The Department of Energy terminated 24 awards worth about $3 billion on May 30, 2025, and 223 projects across 321 awards worth about $7.5 billion more on Oct. 1. Both announcements describe the cuts as targeting carbon capture and other decarbonization awards broadly. Neither one names a specific hydrogen hub as terminated, so it isn’t clear from the public record which hubs, if any, came through intact. Federal money for hydrogen hasn’t disappeared, but it’s now flowing alongside a broader round of clean-energy program cuts.
The 45V tax credit’s deadline moved up five years
The federal credit that subsidizes low-emissions hydrogen production is called 45V. Under the One Big Beautiful Bill Act, signed July 4, 2025, a hydrogen project now has to start construction before Jan. 1, 2028 to qualify, down from a Jan. 1, 2033 deadline in the original law. A separate summary of the same law also lands on a December 2027 cutoff, matching the statute’s language. That’s five fewer years for a project to reach a final investment decision, and only 9% of the announced pipeline had reached that point as of the 2025 count. For developers still lining up financing, that means breaking ground years earlier than the original law required.
What would make hydrogen’s climate case stronger
Whether hydrogen’s climate case gets better or worse from here depends on two numbers. One is the low-emissions share of global supply, under 1% in 2024. The other is how many more hydrogen projects clear a final investment decision before construction has to start by Jan. 1, 2028 to keep the 45V credit. If low-emissions production keeps growing past the roughly 1 million tonnes expected for 2025, the case for hydrogen fuel cells gets stronger. If the pipeline stays stuck at today’s 9% conversion rate, the 4.2-million-tonne estimate for 2030 is probably close to where things land. The hub program was built to help close that gap. Whether the hubs that already have funding keep it will show up in next year’s low-emissions total. Getting there depends on projects that already exist finishing what they started, on the shorter timeline the 45V deadline now sets. Anyone weighing a hydrogen fuel cell purchase, in a car or in an industrial fleet, is betting on how fast those two numbers move.
