Sold as a way to ‘decarbonize’ rail, questions remain around how useful hydrogen actually is

If Google search results are to be believed, hydrogen fuel is big news in rail. More than 13,000 news stories were published in the past year alone, reporting on projects in India, South Korea, Russia, Japan, Switzerland, Turkey, Romania, the UK and the US.

These trains are billed as “a climate-friendly alternative to conventional diesel trains, particularly on non-electrified lines,” because they emit water vapor, rather than a cocktail of greenhouse gases. They work by combining hydrogen gas – usually stored in heavy-duty tanks on the train’s roof – with oxygen from the air, inside a stack of fuel cells. These devices split the hydrogen into protons and electrons. The electrons are then directed through a circuit, generating an electric current that powers the train. At the other side of the fuel cell, the protons, oxygen and electrons are combined, forming water and generating a little bit of heat – these are the only emissions from a hydrogen train.

So, when, back in September 2018, a 100 km-long train route in Lower Saxony, Germany was announced as the world’s first all-hydrogen passenger rail line, excitement was high. Designed to refuel just once a day, with a range of 1000 km, these trains – the Coradia iLint, manufactured by Alstom – were described as “a revolution…for the future of mobility”, heralding “a new era in emission-free rail transport”.

It started well, with the service operating successfully for several years. But fast-forward to 2024, and issues with the supply of hydrogen gas had begun impacting services . In August 2025, delays in the delivery of replacement fuel cells meant that only 4 of the 14 hydrogen trainsets were available for service. At around the same time, Alstom reportedly told staff that the company would be pausing all further development of its hydrogen-powered trains – while pledging to meet existing orders and maintenance contracts – due to a reduction in government funding. In 2025, Lower Saxony’s regional public transport authority announced a tender for 70 battery trains , with no mention of further investment in hydrogen.

Does all this mean hydrogen’s moment in rail is over? Before we can reach a verdict either way, I think it’s worth asking a more basic question: where does the hydrogen actually come from?

Rather than being an energy source like petrol, hydrogen (H 2 ) is an energy carrier , which means it can store and move energy, a bit like a battery. Hydrogen is also the most abundant element in the universe, but it doesn’t tend to exist by itself; instead forming bonds with other elements. So, getting access to it means producing – or separating – it from a variety of sources; typically, by splitting water (into H 2 +O), or extracting it from fossil fuels or biomass . In addition, each of these processes require energy to run, and the form of that energy defines how ‘environmentally-friendly’ the hydrogen is.

There are many classes of hydrogen , but the main two (for our purposes) are labelled green and grey Hydrogen can only be considered ‘green’ if it is produced by the electrolysis of water “with 100% or near 100% renewable energy”, producing close-to-zero net emissions across its lifecycle. ‘Grey’ hydrogen is made by taking natural gas (sometimes mixed with coal), and putting it through an industrial process called steam reforming. As you might imagine, this comes with a substantial carbon footprint – in the range of 830 million metric tons of CO 2 per year; equivalent to 2.5% of all global CO 2 emissions.

According to the International Energy Agency (IEA), of the almost 100 million metric tons of hydrogen produced – for all purposes – in 2024, less than 1% of it qualified as low- or no emissions. The rest was made using fossil fuels. Green hydrogen is currently more expensive to produce than grey – in an earlier report , the IEA said it was “one-and-a-half to six times more costly than unabated fossil-based production.”

What this points to is that, unless clearly stated otherwise, today’s hydrogen trains are most likely to be running on grey hydrogen, rather than green.

That was certainly the case for the iLint trains of Lower Saxony. In 2022, they were powered using hydrogen produced “as a by-product” of the chemical industry, “using energy from gas or coal-fired power plants.” The goal was to switch the production over to renewables – namely, photovoltaics and wind turbines installed at the hydrogen filling station – by 2026. From what I can find, that hasn’t happened. The services that are still running are still using grey hydrogen. Elsewhere in Germany, another fleet of H 2 trains have been experiencing similar issues, with delays in the roll-out and an ongoing reliance on grey hydrogen.

For the past year, California has had its own hydrogen train too, providing commuter services on a nine-mile route between San Bernardino Downtown and Redlands University. Manufactured by Stadler, the FLIRT H 2 ZEMU (Zero-Emission Multiple Unit) was designed for “longer, non-electrified routes where diesel trains are still in use”. The source of its hydrogen has proven tricky for me to track down, as the supplier , Air Liquide, operates both grey and green hydrogen production in California.

Brahim Soua, a vice-president at train manufacturer Alstom argues that grey hydrogen shouldn’t be written off entirely, “Starting a project using grey hydrogen is already a good solution; you’re already reducing your emissions by 45 percent compared to diesel. However, on the trajectory to net zero, the more green hydrogen can be produced, the better.”

Here in the capital city of Aotearoa New Zealand, at the bottom of the North Island, rail passengers are awaiting the arrival of a new feet of trains. Called Tūhono , these units will replace the ageing diesel-powered ones that currently run on two long-distance commuter lines from Wellington (Te Whanganui-a-Tara); one to Palmerston North ( Te Papa-i-Oea ) and another to Masterton ( Whakaoriori ). In addition to providing faster, more frequent services, the Tūhono fleet is exciting for another reason – its trains can operate on both electrified and non-electrified tracks, seamlessly switching power source when needed.

Rather than using hydrogen fuel to do this, the fleet is powered by batteries, making them a Southern Hemisphere first. Tūhono’s battery-electric multiple units (BEMUs) will be charged from the existing overhead catenary wires on the electrified parts of the network, and then run on stored power the rest of the time. The ability to do both is important because of the ~4,130 km of rail track in New Zealand, less than 600 km (about 14%) is electrified , and all of that is on the North Island.

The appeal of batteries over hydrogen is partly to do with the cost of installing infrastructure to enable it, and partly to do with avoiding unnecessary energy conversion. A train running on green hydrogen relies on electricity to turn water into hydrogen, then that hydrogen needs to be compressed and stored, before being turned back into electricity on board via a fuel cell. Energy is lost at every step. By one estimate , the well-to-wheel efficiency of an EMU running on green hydrogen is 25-35%. For a battery EMU that skips straight from grid electricity to storage to motion, the efficiency is more like 70-80%.

In 2020, VDE (German Association for Electrical, Electronic & Information Technologies) apparently reported that battery trains could be built and run for roughly a third less than hydrogen equivalents on comparable routes. I say ‘apparently’ because I can’t find the actual technical report anywhere – just a press release .

So why would anywhere choose hydrogen? Well, it really comes back to the existing rail infrastructure. On routes with some overhead electrification, the typical BEMU range of 40 – 120 km works brilliantly. Wellington’s Tūhono trains, for instance, are designed for non-electrified stretches of 58.5 km and 81 km. But where there is no overhead or track electrification available at all, battery EMUs have limited utility. Increasing their range means increasing the battery mass, which then takes more energy to haul around. Beyond a certain distance, batteries offer diminishing returns.

And while a H 2 train’s hydrogen ‘power plant’ is also heavy, comparing its energy-output-per-kilogram to that of batteries tells another story. Battery-electric trains typically carry traction packs delivering ~ 120 to 160 watt-hours of energy per kilogram , once battery management electronics, thermal management and housing are accounted for. Hydrogen storage systems used on trains offer considerably higher outputs. A 2021 review of rail energy storage puts usable system-level hydrogen energy density at roughly 200 to 600 Wh/kg. For long distance travel on non-electrified routes, hydrogen is a good option…. assuming they can access hydrogen fuel when and where needed.

But ultimately, if the goal is both energy-efficiency and low lifecycle-emissions, full electrification – especially via a renewables-dominated grid – is the way to go for rail.

Overhead catenary, with a tensioned wire strung above the track, typically runs at higher voltages, is good for high-speed and long-distance mainlines, and is the default choice for new electrification almost everywhere. It’s what’s used here in New Zealand.

The other option is a conductive rail that runs alongside or between the running rails. This is used on lots of London’s suburban and regional network, as well as on New York’s commuter networks. It’s preferred for dense urban and metro environments, because it operates at lower voltages and where few trains ever reach their top speeds.

Neither of these options for electrifying a route are cheap. Catenary networks can cost anywhere from $600,000 to $3 million per kilometer , and that’s before accounting for tunnels, bridges or the complex clearance work that’s often required.

In the absence of large-scale infrastructure investment, hydrogen can be a useful bridging technology; a stepping stone towards lower-carbon mass transit and freight. On the right corridor, a hydrogen train – running on green or grey fuel – is likely still an improvement over a diesel one.

And remember, even diesel trains can be a more environmentally-friendly option than road vehicles. An expert advice paper published by Royal Society Te Apārangi states that, “at the present level of rail electrification, the transport of freight by rail on average per tonne kilometre typically produces only around a third of the emissions compared with transporting freight by road.” In 2024, KiwiRail – which runs NZ’s country’s national rail network – estimated that rail freight saved 230,000 tonnes of emissions in just one year.

So, is the hydrogen hype-train running out of steam? As a universal fix for rail decarbonization, I think the answer is yes. What seems to be replacing it isn’t a single technology, but a more practical sorting exercise, with wires and third rails going where there’s enough traffic and infrastructure investment to justify them, batteries filling in non-electrified gaps, and hydrogen where distances are larger or the freight is heavier. It’s not as tidy a picture as you might like, but it’s the one that’s actually showing up on the tracks.