Hydrogen's Reality Check: Targets Shrink, Plumbing Gets Built, and Niches Decide Who Wins
Across 89 developments in two weeks, the hydrogen industry traded headline ambition for permits, pipes and proof points, and the demand side is where the story gets decided.
The loudest hydrogen news of the past fortnight was not a record or a breakthrough. It was a downgrade. The Netherlands cut its 2030 electrolyser outlook, an Italian gigafactory lost part of its funding, and a steel giant handed back German support. Meanwhile the unglamorous plumbing of the industry kept moving: pipeline consultations, salt-cavern drying systems, tax-certificate rules, infringement notices. Our reading: hydrogen is leaving the announcement phase and entering the permit-and-offtake phase, and that phase is far less forgiving.
1. Europe's ambition reset: fewer gigawatts, harder questions
The sharpest data point came from the Netherlands. The Planbureau voor de Leefomgeving (PBL) now puts 2030 green hydrogen electrolyser capacity at roughly 1.2–1.5 GW, down from an earlier national ambition of 3–4 GW. It cites higher electrolyser capital costs, higher grid connection and network charges, slower industrial offtake agreements and limited delivery time.↗ The government's plan still cites 4 GW, but stresses that direct electrification should be preferred where possible, with green hydrogen reserved for uses without practical alternatives. The report also makes a useful technical point: electrolyser capacity measures maximum electrical input, not annual hydrogen output, which depends on utilisation and operating hours.
Galicia told a similar story at its H2Gal summit. The regional agenda targets 500–1,000 MW of electrolysers and up to 100,000 tonnes a year by 2030. Yet of about 650 MW of proposed capacity mapped by the regional industrial alliance, only around 20 percent remains in active planning, and current aid calls are expected to deliver roughly 150 tonnes annually.↗ Delegates named grid connections, offtakes and pipeline infrastructure as the barriers, citing the downgrade of the Guitiriz-Zamora corridor.
Manufacturing is not immune. Italy revised public support for the De Nora electrolyser gigafactory at Cernusco sul Naviglio to €50,348,569, replacing an earlier plan of about €63 million. The revised package is split between €32,250,000 from the PNRR (Italy's recovery plan) and €18,098,569 from the IPCEI fund. The target is up to 2 GW of annual output by 2030, and the company reassessed its plans in response to shifting market dynamics.↗
| Where | Headline ambition | What the fortnight showed |
|---|---|---|
| Netherlands | 3–4 GW electrolysers by 2030 | PBL outlook now 1.2–1.5 GW |
| Galicia, Spain | 500–1,000 MW by 2030 | ~650 MW mapped, ~20% in active planning |
| Italy (De Nora) | Gigafactory, up to 2 GW/yr by 2030 | Support revised to €50.35M from ~€63M |
| Chubut, Argentina | ~13,000 MW wind, ~632,000 t/yr by 2035 | High-level roadmap, not committed projects↗ |
The policy backdrop is both a lifeline and a stress test. A coalition of 170 companies, coordinated by Hydrogen Europe, asked the Commission to keep binding RFNBO targets beyond 2030. RFNBO means renewable fuels of non-biological origin, essentially electrolytic hydrogen made with additional renewable power. Under RED III, member states must reach at least 1% RFNBOs in transport and 42% RFNBO hydrogen in industry by 2030, rising to 60% by 2035.↗ The coalition cites over €15 billion already committed and a potential €50–60 billion more tied to existing targets. It also notes that the European Court of Auditors has warned current 2030 production and import targets may be missed. Separately, the Commission sent formal notices to 26 member states, all but Italy, for failing to communicate full transposition of the Hydrogen and Decarbonised Gas Market directive by the 5 August deadline. This is only the first stage of infringement and carries no fines.↗
The global lens is sobering too. The IEA's Global Hydrogen Review 2026, as summarised this period, put 2025 hydrogen demand above 100 million tonnes but low-emissions production at almost 1 million tonnes. It also noted that clean hydrogen still costs more than conventional hydrogen.↗ The Hydrogen Council, backed by around 140 companies, responded with a call to embed hydrogen in crisis-response measures and to back contracts for difference and offtake guarantees.
The pattern: the industry's own asks have converged on demand-side instruments such as offtake guarantees, mandates and contracts for difference. Supply-side capital alone is no longer treated as sufficient, and developers now say so openly.
2. Steel: the sector where hydrogen has to prove itself
If one industry carries the weight of the green hydrogen thesis, it is steel. Germany shows both the promise and the strain. The Power4Steel programme in Saarland involves about EUR 4.6 billion of investment with EUR 2.6 billion in public support. Salzgitter's SALCOS aims to cut emissions by 60–95 percent through phased hydrogen use. thyssenkrupp's tkH2Steel in Duisburg is backed by roughly EUR 2 billion in public funds.↗ The IEA is quoted saying early hydrogen-DRI (direct reduced iron) plants may cost 50–140 percent more than conventional ones. Reuters, meanwhile, reports that ArcelorMittal withdrew from planned German projects and returned about EUR 1.3 billion in support, citing high energy costs and uncertain hydrogen economics.
Duisburg itself keeps moving. Jacobs extended its programme and construction management role on the 2.5-million-tonne DRI plant. Initial commissioning is planned for late 2026, first hydrogen use in 2028 and full hydrogen operation by 2029.↗ Public funding comprises up to €550 million in EU grants and €2 billion from German federal and state sources. Analysts caution that commercial success depends on affordable renewable hydrogen, stable power and customer commitments to green-steel premiums. The plant is slated to need hydrogen supply of about 143,000 tonnes per year.
Outside Germany, the steel pipeline broadened:
- Hydnum Steel is developing a greenfield DRI-EAF (electric arc furnace) plant in Puertollano, Spain, targeting up to 2.7 million tonnes of flat steel a year. Financing exceeds €1.65 billion, including a €150 million equity commitment from the Cofides FOCO fund, and the plant has secured 500 MW of grid access.↗
- Calix appointed Perdaman under a Heads of Agreement to develop the hydrogen supply solution for its ZESTY demonstration in Kwinana, Western Australia. The plant is designed for up to 30,000 tonnes a year, supported by up to A$44.9 million from ARENA and A$35 million from Rio Tinto. Perdaman has 12 months of exclusivity, and long-term hydrogen purchase terms remain to be negotiated.↗
- The African Development Bank selected four development-stage projects for proposed reimbursable grants totalling US$20 million. These include the Saldanha hydrogen DRI project led by Enertrag with ArcelorMittal South Africa, offered US$5.24 million for pre-feasibility work.↗
Our reading: steel is the clearest case of a buyer with a decarbonisation obligation and a physical need for hydrogen at scale. It is also where the cost gap bites hardest, and the Calix detail shows why. The delivered hydrogen price is still listed as unresolved while the plant itself is already funded. Expect more projects to hinge on a single question: who signs the green-steel premium?
3. Transport: fuel-cell trucks stall, while niches and engines advance
The most uncomfortable number of the period belongs to road freight. BloombergNEF and the Smart Freight Centre found global medium- and heavy-duty fuel-cell truck sales fell 55% year over year to about 530 units in the first half. China accounted for roughly 507, or 95%, while US and European sales combined stayed under two dozen. Battery-electric trucks sold nearly 158,000, up 75%, and took over 99% of low- and zero-emission deliveries.↗ BNEF links a late-2025 hydrogen peak to Chinese incentives that later tapered, which shows how policy-dependent the segment remains. The report's prescription is a split: batteries for most uses, hydrogen where rapid refuelling and distance are critical.
The station operators are living that reality. H2 MOBILITY cut retail prices at 15 German stations from 1 October, after an earlier cut at five. The company says it is now supplying certified green hydrogen from electrolysis and benefiting from greenhouse-gas quota credits, without direct government aid.↗ But it also said it had to consolidate less-used sites and adapt 350 bar dispensers for buses and trucks as passenger uptake grew slowly, and that fleet operators often want long-term contracts before committing. Lower pump prices are an attempt to build volume where the network was originally planned around hydrogen cars that never arrived in numbers.
Hydrogen combustion is the pragmatic hedge
A cluster of internal-combustion stories suggests OEMs and engineers are hedging against fuel-cell cost. SwRI completed a medium-duty spark-ignited hydrogen engine aimed at diesel-like torque with ultra-low NOx.↗ Efficient Hydrogen Motors unveiled its E265 five-stroke engine for retrofits, targeting thermal efficiency above 50% and a pilot with Transdev. The company itself suggests a 30–40% capital cost premium over diesel engines.↗ Toyota filed a US patent for a hydrogen series-hybrid powertrain, though the entry notes such a layout is generally less efficient tank-to-wheel than fuel-cell or battery-electric vehicles.↗ Notably, MissionH24 and Toyota Racing said the H24EVO prototype will use a Toyota 3.5-litre V6 hydrogen-combustion engine with a hybrid system targeting 650 kW, replacing the previously planned fuel-cell architecture.↗ Even the motorsport flagship is changing powertrain to reach existing supply chains.
Fuel cells are not retreating everywhere. Toyota and Isuzu launched the next-generation Sora fuel-cell bus with a range above 300 km, using a 113 kW PEM stack on a shared battery-electric platform, with production scheduled for fiscal 2026.↗ Toyota Motor Europe said a fuel-cell Hilux pickup is gearing up for European production by 2028, targeting over 400 km WLTP range.↗ Quantron and HyperView formed a joint venture with Foxconn as manufacturing partner to build a Gen2 platform for both battery-electric and fuel-cell configurations. The company's targets include 100 zero-emission vehicles in Europe by end-2026, but no binding customer agreements have been disclosed.↗
India is running the large state-backed experiment. Its transport pilots have grown from 37 vehicles and nine stations to 70 vehicles (27 buses, 43 trucks) and 16 stations across 21 highway routes, testing both fuel-cell and hydrogen-combustion vehicles. The target is 400–500 hydrogen buses and trucks by 2030.↗
Our reading: the BNEF numbers and the engine announcements point the same way. Hydrogen vehicles will be sold on duty cycle, such as rapid refuelling, multi-shift operation and retrofit economics. The generic claim that hydrogen is the clean alternative to diesel is losing ground. Fuel-cell hydrogen cars and light trucks look like a long game; heavy-duty is being decided by batteries first.
4. Shipping, rail, ports and aviation: where hydrogen's fit is least contested
Away from road, the fortnight read more constructively, though almost entirely at demonstration scale.
Maritime
The HyShip project selected two Samskip SeaShuttle container vessels for an 18-month liquid hydrogen demonstration on the Rotterdam–Oslofjord route. Each ship will carry 16 Ballard fuel-cell modules (3.2 MW per vessel), with the first vessel expected in service by mid-2027 and a goal of more than 3,000 hours of operating data.↗ H2SITE won a NOK 39.1 million Enova grant for HydraNord Power, an onboard ammonia-to-hydrogen membrane reactor and fuel-cell system targeting 700 kW net, with commissioning in early 2028. Maritime regulators are still finalising safety and classification rules for onboard ammonia.↗ NYK Line launched the 48-metre AMANE fuel-cell dining cruise ship for Tokyo Bay, with service expected around May 2027.↗ DLR launched the hull of its MODULARIS floating laboratory, backed by €36 million of federal funding, to test hydrogen fuel cells, batteries and methanol- or ammonia-capable engines at sea from 2027.↗ A first green hydrogen refuelling of a recreational boat in the Balearics, at Alcudia port, was framed as a blueprint for port permissions and safety protocols.↗
Rail
At InnoTrans, CRRC unveiled the 2,000 kW VELFORCE fuel-cell hybrid locomotive, with refuelling of 15–20 minutes. It is part of a modular platform that also supports battery and diesel-battery variants.↗ Hyundai Rotem showed a 220 km/h hydrogen multiple unit and a freight locomotive prototype. It noted that dedicated hydrogen refuelling infrastructure is required before prototype construction, and that orders are pending.↗
Ports and aviation
A hydrogen fuel-cell ReachStacker was delivered to the Port of Tilbury in a pre-production trial. GeoPura will supply green hydrogen under a 10-year agreement from a 1 MW solar-powered electrolyser.↗ In aviation, Honeywell completed integration of the main subsystems of a fuel-cell power source on a TRL 4 ground rig under the EU's NEWBORN project, backed by roughly €33 million.↗ Ebara Elliott Energy will supply a liquid-hydrogen pump for a refuelling trailer in the Airbus-coordinated GOLIAT project, with high-flow transfer tests in Bristol.↗ On the Rhine, RH2INE won Lloyd's Register Approval in Principle for swappable 380-bar hydrogen containers. Each batch and installation still needs certification under the relevant transport regulations.↗
The signal: these segments share a feature that road freight lacks. Captive operators, fixed routes and single-site refuelling mean the chicken-and-egg problem is smaller. Our reading is that this is where the first genuinely bankable hydrogen demand outside industry may form, but nearly every item above is a trial, a prototype or a certification step, not a fleet order.
5. Production and infrastructure: the plumbing is being laid
The most reassuring developments for long-term investors were not flashy. They were procedural.
- Spain's backbone. The BOE opened public consultation on the 62.095 km Tivissa–Tarragona segment proposed by Enagás: a buried DN900 carbon-steel pipeline at 100 bar with a segment budget of €126,564,435. Commissioning is targeted for 2030 but remains conditional on environmental approval, authorisation and final investment decisions.↗ Enagás's market consultation figures, prospective production of 2.5 million tonnes a year against demand near 1 million tonnes by 2030, are non-binding. Spain's prime minister announced the start of administrative processing for the first backbone sections, within a planned network of approximately 2,750 km.↗
- Storage. EWE GASSPEICHER selected Bilfinger to deliver the first full-scale H2Dry hydrogen dryer at its Huntorf salt-cavern site. A single converted cavern would offer about 16 million Nm³ of working gas, with connection to Germany's planned core network targeted for 2028.↗
- Flagship electrolysis. OMV is progressing alone with its 140 MW plant at Bruck an der Leitha after Masdar withdrew. Financing includes a €450 million EIB loan, covering three-quarters of the estimated €600 million cost, and up to €123 million in European Hydrogen Bank support, with commissioning on track for late 2027.↗ Moeve is reorganising its executive committee to speed the Onuba project in Huelva: 300 MW of alkaline electrolysis from thyssenkrupp nucera, total investment above €1 billion, and around €304 million of Spanish public support.↗
- Smaller electrolyser deals. Sungrow Hydrogen and DOT Engineering reportedly signed for a 21 MW PEM project in Poland, still awaiting permits and RFNBO certification.↗ Plug Power shipped a 1 MW PEM electrolyser to Invercargill, New Zealand, for on-site fuelling of trucks retrofitted with hydrogen-diesel dual-fuel systems.↗
Two things stand out. First, the OMV case shows that a project with refinery offtake, EIB debt and Hydrogen Bank support can survive the loss of a joint-venture partner. A funding stack matters more than the partner list. Second, Arkema inaugurated a €10 million steam methane reforming unit at its Jarrie site, a conventional grey hydrogen facility with no carbon capture announced. The company frames it as supply security, not decarbonisation.↗ That is a reminder that much of the hydrogen actually being built today is captive, fossil-based and operational rather than aspirational.
Policy plumbing elsewhere: Australia's tax office issued guidance on the Hydrogen Production Tax Incentive, a refundable $2 per kilogram offset running from 1 July 2027 to 30 June 2040. It requires Product Guarantee of Origin certificates and a final investment decision by 30 June 2030, and excludes non-renewable production.↗ Hunan Province in China launched a multi-user green-electricity direct-connection policy that prioritises green hydrogen, ammonia and methanol, though it names no specific projects and sets no prices.↗ A Californian dairy pathway received provisional CARB approval with a carbon-intensity score of -1,887.35 gCO2e/MJ, generating credits that can exceed the value of the hydrogen itself. Environmental groups are challenging the crediting assumptions.↗ Credit-driven economics can be generous, but they are exposed to rule changes.
6. Emerging markets and the finance question
The geography of green hydrogen ambition keeps widening, and so does the gap between pipeline and finance.
India delivered the most price-relevant signal. A lowest discovered tender price of ₹279/kg was announced for 10,000 tonnes a year of green hydrogen to the Numaligarh Refinery, with nine bidders and the awardee not disclosed in independent reports. In the same round, green ammonia drew a lowest price of ₹49.75/kg.↗ Observers caution that tender prices may incorporate subsidies, transmission waivers and site-specific power costs, so this is a market signal and not proof of unsubsidised competitiveness. The mission director says the programme could attract ₹8–10 lakh crore of investment by 2030, but government materials support only the lower bound of more than ₹8 lakh crore.↗ India is the clearest test of whether a state-run reverse auction can discover prices that refiners will sign.
Elsewhere, the picture is pipeline-heavy:
- Africa: the four AfDB-selected projects represent an estimated US$23 billion pipeline, but the grants fund studies, not construction. South Africa launched six priority projects, led by a Phelan Green Group electro-SAF venture with a US$100 million equity commitment and offtake agreement, aiming for first exports in early 2029.↗ A Kenyan demonstration with a 220 kW electrolyser is aimed at fertiliser and medical oxygen.↗
- Brazil: the EU committed €3 million via H2Uppp to nine new projects. The Brazilian hydrogen association maps 34 low-emission projects with potential of up to 4.6 million tonnes a year by 2035, and the programme coordinator names offtake as the main challenge.↗
- Central Asia: Allied Biofuels is developing a roughly US$6.1 billion e-SAF complex in Uzbekistan with up to 2.4 GW of electrolysers, targeting a final investment decision in early 2027 while remaining at FEED stage.↗ YPP Corporation is exploring a roughly $2 billion green hydrogen and ammonia complex in East Kazakhstan under an exploratory memorandum.↗
- Cooperation frameworks: Switzerland identified Morocco as a priority partner without creating new public support instruments.↗ The UAE and Mongolia signed an MoU.↗ Gas Malaysia and Air Liquide signed an MoU to explore low-carbon hydrogen, Bio-LNG and CO2 off-take, with no investment commitment.↗
Our reading: derivatives such as ammonia, methanol and e-SAF are the export currency, because they solve transport. But the Uzbek project has no disclosed water-allocation plan and the Kazakh one still lacks financing and offtake. The finance gap is the real constraint, and the AfDB study grants show that even the bank-supported projects are years from financial close. There is also an investor-side signal: Shanghai REFIRE proposed shifting RMB50 million of subscription proceeds from R&D to commercialisation and delivery capabilities.↗ Suppliers are being pushed to prove sales before they prove technology.
7. The frontier: natural hydrogen, cheaper electrolysis and fuel cells for data centres
Geological hydrogen moves from curiosity to regulatory category
A cluster of items treated naturally occurring hydrogen as an asset class. QIMC reported mud-gas readings up to 30.0% hydrogen at its Bennett Hill project in Nova Scotia, while stressing the data are preliminary and no resource estimate exists.↗ H2Au began drilling in Kansas. The USGS warns that prospectivity does not guarantee economic viability.↗ New Zealand's Cabinet agreed in principle to bring white and orange hydrogen under its Natural Environment Act framework, though neither approach has a verified commercial site there.↗ Edith Cowan University reported lab hydrogen release from Pilbara magnetite.↗ And Eclipse Energy published a Wood-modelled cost as low as $0.56/kg for its subsurface microbial route, which Wood frames as a favourable case, not operating data.↗ Per the QIMC entry, only Mali's Bourakébougou field currently produces natural hydrogen at scale. Rules are arriving before resources are proven. Treat the cost numbers with care.
Electrolysis cost-down efforts
H2Pro is advancing membrane-less decoupled water electrolysis from a 0.5 MW Israeli pilot to planned Spanish demonstrations of 5 MW and 25 MW.↗ HKU and a mainland partner are scaling an SS-H2 stainless alloy intended to replace titanium components in PEM stacks, with an estimated 40-fold reduction in structural material costs in a 10 MW system.↗ Other lab work, including a sulfur-linked framework, a palladium membrane for ammonia electrolysis and a metal-free carbon catalyst, remains explicitly early stage, with stability and efficiency data missing.↗↗↗ None are near deployment.
Fuel cells find a buyer in computing
One non-hydrogen-fuel demand story deserves attention. Hilliard, Ohio dropped its appeal against the permit for a 72.9 MW Bloom Energy solid oxide fuel cell installation for AWS, which AWS is funding.↗ Bloom also introduced an 800 VDC architecture for AI data centres, with projected savings that are company-model estimates and not measured results.↗ These fuel cells are not tied to the hydrogen economy's offtake bottleneck, and the permit clearance suggests data-centre power is one of the faster routes to commercial fuel-cell volume. A sourcing note for suppliers: Evolution Energy Minerals reported that graphite from its Chilalo project in Tanzania meets PEM bipolar plate specifications after independent testing.↗
The through-line: from ambition to obligation
Taken together, the fortnight points to three tensions.
Supply is being built faster than demand is being contracted. Almost every project that stalled, shrank or lost a partner did so on offtake, power cost or grid charges. The Netherlands, Galicia, ArcelorMittal in Germany and Masdar at OMV all point the same way. The remedies proposed, from RFNBO mandates to contracts for difference, are demand instruments.
Regulation is moving at the speed of administration. 26 member states missed a transposition deadline. Spanish pipelines are in consultation windows. Australia's incentive does not start until 2027. None of this is bad news, since bankability requires it. But it means 2030 targets now depend on approval timelines as much as on technology.
Hydrogen is sorting itself into winners by application. Steel, refining and chemicals have the volume. Ships, rail and ports have captive routes. Road freight, by BNEF's figures, is conceding the mass market to batteries. Our view: the next phase rewards companies that sell to a specific, contracted user. It punishes those still selling a general promise of the hydrogen economy.
What to watch
- The EU response to the RFNBO coalition. No public response has been issued, and EU law is unchanged until the Commission proposes a post-2030 framework. The tone of that proposal will set the investment climate for European electrolysis.
- Whether India's ₹279/kg becomes a contract. Watch for the awardee, supply-contract finalisation, commissioning timelines and lifecycle-emissions verification.
- The infringement clock. Member states have two months to respond to the Commission's notices before a reasoned opinion is possible.
- Duisburg's commissioning and the Calix FID path. Initial commissioning at Duisburg is planned for late 2026. Calix's 12-month exclusivity window will reveal whether a delivered hydrogen price can be agreed.
- The REFIRE vote. Shareholders vote on 12 October on the reallocation, an early read on whether investors accept commercialisation over R&D.
The period's lesson for anyone allocating capital in hydrogen news: read the offtake line first, the megawatts second.