>
Gigaproject capital is moving ahead of offtake: where hydrogen economics close first, and who funds the gap.
Volge Quarterly is Volge's signature series, published every quarter for enterprise customers. Each edition takes one subject, sets out where it currently stands on the published record, and makes a set of dated, falsifiable judgements on where it is going, written to give a buyer something firmer than a forecast: every figure carries its source and its date, every judgement carries the date on which it resolves, and the following year's edition grades each one hit, missed or partial.
2027 is the year the green premium meets its first real test. The Gulf ships first production, but its near-term volume is blue, not green. India has awarded a pipeline and commissioned almost none of it. Two bets on the same fuel, read in the physical world rather than the press release.
Figures carry their basis and their date. Where published estimates disagree, the range is given and the basis named rather than averaged away, and where a widely quoted number does not survive checking it is corrected here rather than repeated. Every chart plots a figure stated in the text, and nothing is projected beyond what a named source projects.
A market whose defining number turned the wrong way. The announced-for-2030 pipeline of low-emission hydrogen fell for the first time: from 49 to 37 Mtpa in a single year, and again to 27 Mt in the 2026 review, with barely a tenth at investment decision [IEA, Global Hydrogen Review 2026, October 2026]. 2027 tests the projects that survived the contraction.
2027 is the year the green premium stops being an argument and becomes a measurement. The question is no longer whether green hydrogen can be made, but whether it can be sold: at a price a buyer will pay, from a plant that has actually been built. On present evidence, the answer for most of the pipeline is not yet.
For half a decade the sector was priced on ambition, and in the year to December 2025 the ambition itself contracted. The International Energy Agency records that the world’s announced-for-2030 low-emission hydrogen pipeline fell from 49 to 37 million tonnes a year, and to 27 million tonnes in the 2026 review [IEA, October 2026], the first such contraction on record, with only about 9% of it having reached a final investment decision [IEA, Dec 2025]. New offtake signed in 2024 fell to 1.7 Mtpa from 2.4 the year before and stayed near 1.7 Mt in 2025, with only about a fifth backed by firm contractual commitments [IEA, October 2026]. The market is failing to convert demand, not building it, and 2027 is the year that reality is measured in tonnes rather than targets.
A third fact has since overtaken both, and it is not a cost fact. Iran declared the Strait of Hormuz closed within days of the United States and Israeli strikes of late February 2026, and flows through it remain near zero [IMF PortWatch, via IFPRI, 18 September 2026]. The Gulf's hydrogen case is an export case: it rests on shipping molecules to Europe and to Asia through that water. In the first half of 2026 nitrogen exports fell 63% for Saudi Arabia, 37% for Qatar and 29% for the United Arab Emirates. Ammonia is the carrier the whole Gulf pipeline assumes, and the route it assumes is now the binding constraint, ahead of electrolyser cost and ahead of offtake. An export project that cannot ship is not a project with a premium problem; it is a project without a market. The reader should hold that against every Gulf volume in the pages that follow, including the blue volumes, which travel by the same sea.
Two facts frame the whole edition. The first is that the green premium is not closing: it is, on the cost side,
widening. Green hydrogen in Europe is priced at roughly $4.50 to $6.00 a kilogram against $1.50 to $2.50 for grey and $2.00 to $3.50 for blue, a premium of two to three times over blue and up to four times over grey [Green Fuel Journal, 2026]; CareEdge puts renewable hydrogen at 1.5 to 5 times unabated fossil, a wider band [CareEdge, 2026]. Electrolyser capital cost, the input the entire thesis assumed would fall, rose by roughly 57% at the median between 2022 and 2024. The second fact is that the two geographies attacking this gap look nothing alike, and near-term the Gulf’s producing volume is not even green.
The Gulf is export-led but blue-heavy in the near term. Its one bankable green flagship (NEOM’s Helios plant in Saudi Arabia) is scheduled to reach first production, but the region’s 2026–27 volume leaders are blue-ammonia plants: Qatar’s Ammonia-7 and the UAE’s Ta’ziz. India is demand-led and domestic, its market manufactured by auction; but it had awarded roughly 862,000 tonnes a year of green-hydrogen production while commissioning only about 8,000 (some 0.16% of its 2030 target) by early 2026 [Business Today, Jun 2026]. Both are bets on the same molecule; both are, in the physical world, earlier than the pipeline suggests.
This is not a bearish read so much as a literal one. The direction is not in doubt: renewable power in the Gulf and in India is among the cheapest on earth (Abu Dhabi has contracted solar at $0.0135 a kilowatt-hour, the cheapest electricity recorded anywhere [PV Tech, 2020]), and Europe’s rulebook is creating a class of buyers obliged to pay for the certified renewable molecule. What is in doubt is the pace, and 2027 is the year the pace becomes visible: in first production at NEOM, in the first blue cargoes from Qatar and Ruwais, in the first green ammonia flowing into Indian fertiliser plants, and in the first government-backstopped cargo landing at Rotterdam.
Read as a whole, the year resolves into five movements, each developed in Part III. The cost gap widens before it narrows. The two models (Gulf export and Indian demand) diverge rather than converge, and the Gulf’s near-term output is blue. Offtake hardens as the
true binding constraint, the reference bankable structure being a single creditworthy buyer taking the whole volume for the tenor of the debt. Europe’s renewable-fuel rules, backed from January 2026 by the carbon border levy, become the de-facto passport for any molecule seeking a premium buyer. And somewhere on the map, at least one announced project is delayed, descoped or shelved, the fable risk a contracting pipeline all but guarantees.
The direction is settled; the pace has slipped. A pipeline that shrank for the first time, a premium that widened, and demand contracted rather than built: 2027 is the year the sector is read in what ships, not what is announced.
Every figure is attributed inline and year-stamped, with numbered sources in the annex. Company announcements are treated as claims, not facts, and the status of each agreement (memorandum or binding take-or-pay contract) is stated where known, because most headline export volumes across the Gulf and India are memoranda, not contracts. A small number of developer figures could not be independently reconciled (among them NEOM’s exact daily output and several Indian corporate sizings) and are flagged where they appear.
What follows is the reconciled basis, the two models, the forces, and the Calls.
The reconciled basis for this edition, on one page. The headline is that the two numbers the thesis most needed to move in its favour (the announced pipeline and the electrolyser cost) moved against it in the year to December 2025.
Reconciled from the IEA Global Hydrogen Review 2025, MNRE and SIGHT/SECI results, Hydrom and developer disclosures, and cost work in Green Fuel Journal (2025–26). Ranges reflect genuine source disagreement on scope and grade. Current to July 2026.
A pipeline that shrank, a premium that widened, and a demand market that contracted. The projects that survive 2027 are the exceptions, not the pipeline.
The green premium is not a market failure to be argued away; it is an accounting fact, and in 2026 it is widening on the cost side. Green hydrogen runs two to three times the cost of blue and up to four times grey, and the electrolyser (the input assumed to cheapen) became more expensive.
The premium has three components, and the sector’s hopes rest on only one of them moving. The first is renewable power, the largest input and the Gulf’s and India’s strongest hand: Gulf solar is the cheapest electricity on earth and Indian solar runs near $25–30 a megawatt-hour, with projected cuts of around 40% in the levelised cost of hydrogen into the early 2030s [GFJ, 2026]. The second is electrolyser capital cost, which rose roughly 57% at the median between 2022 and 2024, driven by input-cost inflation and demand for scarce Western manufacturing capacity. The third is the certification and logistics of moving a compliant molecule to a distant buyer.
| Pathway | Cost, 2026 | Notes |
|---|---|---|
| Grey (natural gas) | $1.50–2.50/kg | The unabated benchmark green must displace |
| Blue (gas + CCS) | $2.00–3.50/kg | The Gulf’s near-term producing pathway |
| Green, Europe | $4.50–6.00/kg | RFNBO-grade; 2–3× blue, up to 4× grey |
| Green, India | ₹397–560/kg (~$4.70–6.65) | Refining tenders IOC ₹397, BPCL/HPCL ₹387/kg |
| Green, China | <$2/kg | Cheap alkaline stacks; not a Gulf/India cost |
Cost bands, Green Fuel Journal 2026. Chinese electrolysers ~$600/kW against ~$2,500/kW in the West drive a projected 2030 bifurcation: China and the Gulf at $2–3/kg, the West at $5–7/kg.
Against grey and blue, the green case therefore rests not on cost but on obligation: the existence of buyers
whose rules require the renewable molecule. That is the RFNBO passport, developed in Part III, and it is the reason a widening cost premium and a real market can coexist.
The binding constraint of 2027 is not production capacity but contracted demand. New offtake signed across the whole global market fell to 1.7 Mtpa in 2024 and was flat in 2025, and the reference bankable structure (a single creditworthy buyer taking the entire volume for the life of the debt) exists in only a handful of projects.
Electrolyser supply is being solved; India alone has awarded 3.0 GW of manufacturing capacity across two tranches [SIGHT/SECI, 2024]. The machine is not the scarce input. The scarce input is the buyer willing to sign a take-or-pay contract at a price that pays for the premium. Most of the headline export volumes quoted across the Gulf and India are memoranda or framework agreements, not take-or-pay, and roughly 80% of committed hydrogen investment worldwide still sits in existing industrial uses rather than new clean demand [IEA, 2025].
NEOM – Air Products. A 30-year exclusive offtake of 100% of Helios output: the only structure the Gulf has proven bankable, and the anchor that took the project to financial close in May 2023 [5][6]. Fertiglobe – H2Global. The ADNOC/OCI venture won Europe’s first import auction at €1,000 a tonne into Rotterdam: 19,500 tonnes in 2027 rising to 397,000 cumulative by 2033, government-backstopped [5][29]. ACME – Japan. ACME’s Odisha ammonia, with IHI, was certified for Japan’s contract-for-difference scheme in mid-2026, under a contract for difference covering 228,000 tpa over 25 years from September 2030, with commissioning expected July 2030 [30]. India fertiliser. SECI aggregated 724,000 t/yr of green-ammonia demand; six agreements for 670,000 t/yr were signed on 31 March 2026 across 13 fertiliser units on ten-year terms [27].
Where a large announced capacity is paired with an unnamed buyer or a non-binding memorandum, this edition treats it as intent, not offtake. Readers should apply the same test: a tonne with no take-or-pay contract behind it does not finance a plant.
The Gulf exports; India consumes. But the Gulf’s near-term volume is blue ammonia, not green (one bankable green flagship at NEOM against blue plants at Qatar and Ruwais), while India has auctioned a pipeline it has scarcely begun to build.
NEOM Green Hydrogen Company (the Helios plant at Oxagon) is the Gulf’s only bankable green flagship, and the exception that proves the rule about offtake. At roughly $8.4 billion it reached financial close in May 2023 on the strength of a single 30-year contract, and construction was reported complete to analysts on 6 August 2026, with commercial operations set for 2027 and Air Products guiding to first production in mid-2027 after what it calls a long commissioning process.
The project is an equal joint venture of ACWA Power, Air Products and NEOM, and its structure is the template the rest of the Gulf has not been able to replicate: one very large plant, one deep consortium, and a single long-dated offtake that de-risked the whole. Air Products is the exclusive offtaker of 100% of the output for 30 years [Ammonia Energy, 2023], and it was that contract, not the engineering, that made financial close possible on 22 May 2023. The design pairs about 4 GW of solar and wind with more than 2 GW of thyssenkrupp electrolysis, converting hydrogen to ammonia because ammonia, not hydrogen, is what moves economically across an ocean.
The 600 t/day hydrogen figure is carried from developer disclosure and could not be independently reconciled; it is flagged as directional. First production may slip from end-2026 into 2027.
The Call on NEOM is high-confidence but narrow: first production is a ramp question, not a demand one, because 100% of the volume is already sold. What NEOM does not prove is repeatability: no second Gulf green megaproject has reached FID with binding offtake, and that absence, not NEOM’s progress, is the region’s real signal.
| Attribute | Detail | Source / status |
|---|---|---|
| Project | NEOM Green Hydrogen Co. (Helios), Oxagon | JV ACWA Power / Air Products / NEOM, equal |
| Investment | ~$8.4 billion | Financial close 22 May 2023 [6][17] |
| Renewables | ~4 GW solar and wind | [2][6] |
| Electrolysis | >2 GW, thyssenkrupp | [18] |
| Hydrogen output | ~600 t/day green H₂ | Developer disclosure: marked to verify |
| Ammonia output | >1.2 Mtpa green ammonia | [2][6] |
| Offtake | Air Products, exclusive, 100%, 30 years | Binding [5] |
| Status | Construction complete; first production mid-2027 | [7][8] |
Oman has built the Gulf’s most orderly green-hydrogen programme, and exposed the sector’s defining flaw in doing so: supply is being auctioned ahead of demand. The blocks are allocated and the developers named; the binding buyers are not.
Oman’s distinction is Hydrom, a state orchestrator that auctions defined land blocks at Duqm and Salalah under standard terms rather than negotiating each project bilaterally. Its second round in April 2024 allocated two Dhofar projects at around $11 billion of prospective investment, and a third round for Duqm was launched in May 2025 with awards due in 2026 [Hydrom, 2024–25]. The national ambition is roughly 1.38 Mtpa of hydrogen by 2030, rising toward 7.5–8.5 Mtpa by 2050. The pipeline is unusually legible; its weakness is that the offtake to fill it has not been contracted.
| Project | Detail | Status |
|---|---|---|
| HyPort Duqm (OQ / Uniper / DEME) | 500 MW, ~60,000 t/yr | Phase 1 ~2026 |
| SalalaH2 (OQ / Linde) | 400 MW, ~$1bn | FID targeted 2026 |
| ACME Duqm | 3.5 GW RE → 900,000 t/yr ammonia; ~$3.5bn | Phased; demand-side constrained |
| Dhofar R2 (EDF / J-POWER / Yamna; Actis / Fortescue) | Two projects, ~$11bn | Round 2, Apr 2024 |
| GEO | 25 GW → up to 1.8 Mt H₂ / 10 Mt ammonia | Long-dated |
| Duqm Round 3 | Blocks launched May 2025 | Awards 2026 |
Hydrom and developer disclosures, 2024–25. ACME Duqm’s ammonia capacity is auctioned; a binding full-volume offtake had not been disclosed as of mid-2026, the demand-side constraint in miniature.
The Call on Oman is therefore hedged toward the constraint rather than the pipeline. Oman’s Round 3 Duqm awards were scheduled for the second quarter
of 2026 and none has been announced, and no SalalaH2 FID is on record: both are now overdue rather than pending. The region’s green FIDs beyond NEOM remain thin, and Oman’s legibility does not substitute for the take-or-pay contracts that finance plants.
The correction that most changes the Gulf story is this: the region’s near-term producing volume is blue ammonia, not green. The UAE’s Ta’ziz and Qatar’s Ammonia-7, both blue, are the plants scheduled to start up in 2026–27, while the UAE’s flagship green target has been pushed past 2030.
Ta’ziz closed $1.39bn of financing on 4 November 2026 and reaffirmed 2027 operations. A venture of ADNOC’s Fertiglobe with GS Energy and Mitsui, it is building 1.0 Mtpa of blue ammonia at Ruwais (natural gas reformed with carbon capture), with construction from the third quarter of 2024, first production targeted in 2027, full carbon capture by 2030, and Mitsui offtaking into Japan and Asia [2][21]. Qatar’s Ammonia-7, at Mesaieed, is 1.2 Mtpa of blue ammonia with 1.5 Mtpa of CO₂ capture, about $1.0–1.2 billion, ground broken in November 2024 and start-up in 2026 [MEES, 2024]. These are the Gulf’s real 2026–27 volumes, and they are blue.
| Project | Type | Detail / status |
|---|---|---|
| Ta’ziz (ADNOC/Fertiglobe + GS Energy + Mitsui) | Blue ammonia | 1.0 Mtpa, Ruwais; construction Q3 2024; first production 2027; Mitsui offtake, Japan/Asia |
| Qatar Ammonia-7 (QatarEnergy/QAFCO) | Blue ammonia | 1.2 Mtpa, Mesaieed; SMR + 1.5 Mtpa CCS; ~$1.0–1.2bn; start-up 2026 |
| Masdar green (1 Mtpa) | Green H₂ | Target pushed past 2030 (Oct 2024); migrated to European JVs (OMV, EnBW, TotalEnergies) |
| Helios Industry, KEZAD | Green ammonia | 200,000 t/yr; pre-FID / stalled |
Green Fuel Journal, MEES and Bloomberg, 2024–26. Masdar’s 1 Mtpa is described as “within the next decade” and was explicitly delayed past 2030 in October 2024. It should not be read as an on-track 2030 commitment.
Export-led, blue-heavy near-term, offtake-anchored. One bankable green flagship (NEOM) against blue volume leaders (Qatar, Ta’ziz) and a delayed green target (Masdar). The thesis’s green content is thinner than the pipeline implies.
India’s bet is the mirror image of the Gulf’s: a broad, publicly-funded programme that manufactures demand through auctions rather than backing a few export plants. Its defining number in 2026 is the gap between what has been awarded (roughly 862,000 tonnes a year of green hydrogen) and what has been commissioned: about 8,000 tonnes, some 0.16% of the 2030 target.
The National Green Hydrogen Mission, approved on 4 January 2023, carries an outlay of ₹19,744 crore, a target of at least 5 million tonnes a year of green hydrogen by 2030, and about 125 GW of linked renewable capacity [MNRE, 2023]. Its instrument is SIGHT: ₹17,490 crore split between electrolyser manufacturing (₹4,440 crore) and production (~₹13,050 crore), with a production incentive tapering ₹50 to ₹40 to ₹30 a kilogram over three years. The design is sound; the delivery is early. By 6 August 2026 roughly ₹292 crore, under 1.5% of the ₹19,744 crore outlay, had been spent, and MNRE told Parliament that no electrolyser or production facility supported under SIGHT had been fully commissioned [MNRE, reported 3 November 2026].
| Auction | Volume | Detail |
|---|---|---|
| Production, Mode-1 T-I (Jan 2024) | ~412,000 t/yr | Reliance 90,000 @ ₹18.90/kg (lowest); ACME & Greenko ZeroC 90,000 each @ ₹30/kg |
| Production, Mode-1 T-II (Mar 2025) | 450,000 t/yr | Oversubscribed 1.4×; AM Green/Waaree/Sembcorp/L&T 90,000 each; Reliance +49,000 |
| Cumulative | ~862,000 t/yr | 19 firms; Reliance largest at ~139,000 t/yr |
| Electrolyser mfg | 3.0 GW | T-I 1.5 GW (Jan 2024) + T-II 1.5 GW (Sep 2024); no T-III as of Jul 2026 |
SECI auction results, 2024–25. Reliance’s ₹18.90/kg is the lowest production bid; the production incentive underwrites it. Awarded is not commissioned: the two differ by more than a hundredfold.
The Call follows directly: India commissions less than 5% of its 5-MTPA target by end-2027. The programme is not failing so much as it is early, and the distance between an auction award and a running plant is the whole story of the Indian year.
SECI/SIGHT auction results and developer disclosures, 2024–26. Jamnagar sizing, Adani’s ~1 Mtpa Mundra target and AM Green’s wider programme are
developer figures partly predating July 2026 and are flagged as directional. Awards are production allocations, not commissioned capacity.
The Indian programme is unusually broad (nineteen firms hold production awards), and that breadth is at once its resilience and its fragmentation risk. The table sets out the developers that matter, their awards and their standing.
| Developer | Award / project | Scale | Status |
|---|---|---|---|
| Reliance | Production + electrolyser mfg; Jamnagar | ~139,000 t/yr (largest) + 300 MW; ₹18.90/kg bid | Under build; “$1/kg by 2030” aspiration |
| Adani (ANIL) | Mundra; electrolyser mfg | ~1 Mtpa target; 198.5 + 71.5 + 30 MW | Phased |
| ACME | Production + SECI green ammonia; Bikaner/TN/Oman | 90,000 t/yr + ~370,000 t/yr NH₃ (record ₹49.75/kg, IFFCO Paradeep) | Awarded; Japan corridor via Odisha-IHI |
| L&T | Production + Hazira | 90,000 t/yr + 63 MW | First indigenous electrolyser Mar 2024 |
| NTPC / NGEL | Green ammonia; Pudimadaka hub | 70,000 t/yr @ ₹51.80/kg | Pact signed 31 Mar 2026 |
| Greenko / AM Green | Kakinada | 90,000 t/yr production award | Awarded (T-II) |
| Waaree | Production + Valsad gigafactory | 90,000 t/yr + 300 MW | Awarded |
| Sembcorp | Production + Tuticorin | 90,000 t/yr; Kyushu Electric offtake | First output targeted 2029 |
| Ohmium | Electrolyser mfg; Bengaluru PEM | 274 MW; $250M Series C (TPG) | Scaling |
| Jakson Green | Green ammonia | Record-low SECI ammonia bid | Awarded |
| Others | Hygenco, Welspun, Torrent, CESC, UPL, JSW, John Cockerill Greenko, Jindal, Advait, Matrix, Avaada | Various | Awarded / phased |
India’s answer to the offtake problem is to aggregate demand it can direct: its own fertiliser, refining and steel industries. The largest anchor is fertiliser, where 670,000 tonnes a year of green ammonia was contracted on 31 March 2026, India’s first at-scale green-hydrogen demand.
Through SECI, India aggregated 724,000 tonnes a year of green-ammonia demand and signed six agreements for 670,000 tonnes across 13 fertiliser units on ten-year terms, at prices of ₹49.75 to ₹64.74 a kilogram against grey ammonia at roughly ₹110, an
implied foreign-exchange saving of about $2.5 billion over the decade [Mercom, Mar 2026]. This is the demand that makes the Indian model real: not a merchant export hope but a domestic substitution of imported grey ammonia, directed by the state into plants obliged to buy.
| Anchor | Volume | Terms / price |
|---|---|---|
| Fertiliser (SECI aggregated) | 724,000 t/yr | 670,000 t/yr signed with 11 units of the 13 allocated, 10 yrs |
| – price | – | ₹49.75–64.74/kg vs grey ~₹110/kg; ~$2.5bn forex saving/decade |
| Refining (SIGHT) | 20,000 t/yr + 10,000 | IOC/BPCL/HPCL ₹387–397/kg; Numaligarh 10,000 |
| Steel | First cert Feb 2026 | Green Steel Taxonomy (<2.2 tCO₂e/t) notified 12 Dec 2024; ArcelorMittal Nippon Steel |
Mercom and MNRE, 2026. The fertiliser contracts are binding ten-year offtake, the substance behind the Indian model. There is, however, no economy-wide consumption obligation as of mid-2026: demand is aggregated, not mandated.
Demand manufactured by policy and anchored in fertiliser, refining and steel. The strength is that offtake is contractual and domestic; the exposure is that it is subsidy-dependent, un-mandated, and, so far, barely built.
The Gulf concentrates the bet and ships blue first; India spreads it and has scarcely begun. The corridors and constraints that follow bind both.
There are two export corridors, and they behave differently. Europe pulls the renewable molecule through a rulebook (RED III and, from January 2026, a carbon border levy), but pays only through government-backstopped auctions. Asia pulls ammonia as a co-firing fuel through contracts-for-difference. India’s binding export traction, notably, is Asia-facing, not European.
Europe is the gatekeeper. RED III, in force since November 2023, requires 42% of industrial hydrogen to be RFNBO by 2030 and 60% by 2035, subject to Delegated Act tests on additionality, temporal correlation tightening to hourly from 1 January 2030, and a lifecycle greenhouse-gas saving of at least 70% [EU, 2023]. REPowerEU targets 20 Mt of renewable hydrogen by 2030, split evenly between domestic production and imports. But willingness to pay is thin: Europe’s first import auction, run through H2Global, cleared green ammonia at €1,000 a tonne, and it was won by Fertiglobe, a UAE-parented supplier, delivering 19,500 tonnes into Rotterdam in 2027 [5][29].
IEA, EU, JOGMEC and Ammonia Energy, 2024–26. Ammonia is the carrier of choice (liquid at −33°C against −253°C for hydrogen), and Japan’s JERA is running a 20% co-firing demonstration at its 1,000 MW Hekinan station.
| Corridor | Mechanism | 2027 signal |
|---|---|---|
| MENA/Gulf → Rotterdam | H2Global auction, €1,000/t | First Fertiglobe cargo, 19,500 t, 2027 |
| India → Japan | JOGMEC CfD (¥3tn/15 yrs) | ACME/IHI certified 2026; 228,000 tpa from Sep 2030 |
| Gulf/Aus → Korea | CHPS co-firing auction | Only KOSPO cleared (Dec 2024); ~11% of volume filled |
| EU rulebook | RED III + CBAM (from 1 Jan 2026) | 42% RFNBO by 2030; carbon levy on H₂/ammonia |
The corridor conclusion is uncomfortable for the green thesis: the binding contracts that exist are either government-backstopped (H2Global, the Japanese and Korean schemes) or single-buyer (NEOM). Merchant export without a take-or-pay contract or a public backstop is, in 2027, still a corridor of intent.
Seven constraints bind the sector, and they are ordered. Offtake is the master constraint; everything else is downstream of it. A plant with a binding buyer solves its financing; a plant without one cannot be built however cheap its power.
New offtake signed fell to 1.7 Mtpa in 2024 and was flat in 2025, and roughly 80% of committed hydrogen investment worldwide remains in existing industrial uses rather than new clean demand [IEA, 2025]. Until take-or-pay contracts exist, the pipeline is a wish list.
Green runs two to three times blue, and electrolyser capex rose about 57% between 2022 and 2024. Western stacks cost roughly four times Chinese ones, which decides who can build economically and exposes Gulf and Indian projects to import policy.
Gulf and Indian solar is the cheapest on earth, but round-the-clock firming to keep electrolysers running adds cost that record daytime bids conceal.
Electrolysis needs ultrapure water; in the Gulf that means desalination at scale, a cost and an energy load, as NEOM’s integrated build shows.
Ammonia terminals, cracking capacity and dedicated shipping are nascent. The molecule can be made before it can be moved.
RFNBO rules raise the bar for compliance and CBAM helps the green case at the margin, but merchant European demand above the ~€1,000/t auction clearing price is thin.
About 1.3% of the Mission budget is spent, there is no binding economy-wide consumption obligation, and the grey-to-green switch depends on subsidy holding.
Solve offtake and the rest becomes engineering. Leave it unsolved and cheap power, falling-someday capex and a supportive rulebook do not build a single plant.
Five through-lines cut across the two geographies. Each is a force that will move the sector in 2027, and together they explain the Calls that follow: the cost gap, the two models, the offtake bottleneck, the European passport, and the risk that a flagship does not ship.
The green premium is widening on the cost side, and that is the fact from which the whole year follows. The input the thesis assumed would fall (the electrolyser) became more expensive, and the gap to grey and blue grew rather than shrank.
For a decade the green case rested on a falling cost curve: electrolysers would cheapen, renewable power was already cheap, and the two together would close the gap to fossil hydrogen. Half of that held. Renewable power is cheaper than ever: Gulf solar at little over a cent a kilowatt-hour, Indian solar near $25–30 a megawatt-hour. But electrolyser capital cost rose about 57% at the median between 2022 and 2024 [GFJ, 2026], reversing the decline on which the arithmetic depended, as input-cost inflation and a scramble for scarce Western manufacturing capacity overwhelmed the learning curve.
The result is a premium that, in 2026, runs two to three times blue and up to four times grey. And it fractures along a geopolitical line. Chinese electrolysers cost roughly $600 a kilowatt against about $2,500 in the West, a four-fold gap that increasingly decides who can build. The likely 2030 outcome is not convergence
but bifurcation: China and the Gulf producing at $2–3 a kilogram, the West at $5–7.
This is why the green case cannot rest on cost. Unsubsidised, green does not beat grey or blue in 2027, and the honest reading is that it will not for years. What green has instead is a set of buyers whose rules require the renewable molecule; and that regulatory demand, not a cost crossover, is what a green project must sell into. The cost gap is the reason the next three forces matter as much as they do.
The state of the sector · Volge Quarterly 2027
The Gulf and India are making the same bet on the same molecule in opposite ways, and 2027 tests both structures rather than settling which is right. The Gulf concentrates risk in a few export plants and ships blue first; India spreads risk across many developers and anchors demand at home.
The Gulf model is export-led and capital-concentrated. It backs a small number of very large plants against long-dated foreign offtake, and its bankable instance, NEOM, is de-risked by a single 30-year contract for 100% of output. The strength is bankability; the exposure is concentration, and the awkward truth that the region’s near-term producing volume is blue ammonia from Qatar and Ruwais, not green. The Gulf is decarbonising its existing ammonia business faster than it is building a green export one.
Neither model has been proven, and 2027 does not choose between them. What it does is expose each to its own failure mode: the Gulf to the thinness of green
FIDs beyond NEOM, India to the gulf between award and plant. The edition backs both to survive the year and neither to vindicate itself in it.
The Indian model is demand-led and capital-dispersed. It manufactures demand through auctions, spreads production awards across nineteen firms, and anchors offtake in its own fertiliser, refining and steel industries. The strength is that demand is contractual and domestic (670,000 tonnes a year of fertiliser offtake signed in March 2026) and immune to European willingness-to-pay. The exposure is fragmentation, fiscal dependence, and a commissioning record of 0.16% of target.
Offtake, not electrolysers, is the binding constraint of 2027. The scarce input is not the machine that splits water but the signature of a creditworthy buyer willing to take the whole volume for the life of the debt.
The evidence is in the flows. New offtake signed across the global market fell to 1.7 Mtpa in 2024 from 2.4 the year before and was flat in 2025: the market is failing to convert demand, not building it, and about 80% of committed hydrogen investment still sits in existing industrial uses [IEA, 2025]. Meanwhile electrolyser manufacturing races ahead: India alone has awarded 3.0 GW. The asymmetry is the whole point. Supply is being built faster than demand is being contracted, and a stack with no buyer is a stranded asset in waiting.
The reference bankable structure is narrow and worth stating precisely: a single creditworthy buyer taking 100% of output for the tenor of the debt. NEOM has it: Air Products, 30 years. Almost nothing else does. The government-backstopped auctions (H2Global, Japan’s and Korea’s schemes) substitute the state for the buyer, and India’s fertiliser contracts substitute policy for the market. Every financed project in the sector traces to one of these three: a single buyer, a public backstop, or a mandated one.
This is why the edition reads the year in contracts rather than capacity. A memorandum is a hope; a take-or-pay contract is a plant. The projects that reach first production in 2027 will, without exception, have solved offtake first.
The forces · Volge Quarterly 2027
Europe’s renewable-fuel rulebook is becoming the de-facto passport for any molecule seeking a premium buyer. Compliance with RED III’s RFNBO definition, reinforced from January 2026 by the carbon border levy, increasingly decides which projects can access European demand at all.
RED III, in force since November 2023, requires 42% of industrial hydrogen to be RFNBO by 2030 and 60% by 2035. The Delegated Acts set the tests that make certification hard: additionality (new renewable capacity, with electrolysers commissioned before 2028 grandfathered until 2038); temporal correlation tightening from monthly to hourly matching from January 2030; and a lifecycle greenhouse-gas saving of at least 70% [EU, 2023]. From 1 January 2026 the carbon border adjustment mechanism began its definitive regime, reaching hydrogen and ammonia through fertilisers, tilting the landed economics toward the compliant molecule.
The passport cuts two ways. It is the reason a green project can command a premium at all: the rules manufacture buyers obliged to pay. But it is also a gate: a non-compliant molecule, however cheap, is locked out of the European market, and certification is neither quick nor certain. AM Green’s pre-certification and ACME’s Japanese certification are competitive assets precisely because they are scarce.
The Call is that RFNBO compliance forces at least one announced Gulf or Indian EU-export memorandum to lapse or restructure by end-2027, the gate closing on a project that assumed access it had not earned.
The edition names fable risk as a force rather than a prediction because its precise location is unknowable and its occurrence is near-certain. Some announced project (in the Gulf or in India) that today carries a confident timeline will, by end-2027, carry a slipped one
or none at all. The discipline is to hold every announced figure to the question the whole edition asks: is there a binding buyer behind it, and has ground actually been broken?
A pipeline that shrank for the first time all but guarantees that at least one announced flagship is delayed, descoped or shelved in 2027. The question is not whether fable risk is realised but where.
The IEA’s contraction, 49 to 37 Mtpa of announced 2030 capacity in a single year and 27 Mt by the 2026 review, is fable risk measured in aggregate: projects that existed on paper and then did not. At the individual level the same force operates. Masdar’s 1 Mtpa green target was pushed past 2030 in October 2024 and its green ambition migrated into European joint ventures. The KEZAD green-ammonia project sits pre-FID and stalled. Adani’s TotalEnergies partnership was paused after 2023. None of these is a scandal; each is the ordinary attrition of a young, expensive sector meeting the cost of capital.
The five forces converge on a single instruction for the reader: count tonnes and contracts, discount announcements, and expect the map to lose a project before it gains one. The Calls apply that discipline to the year.
Begin with what is producing, because the list is short and entirely Indian. It comprises demonstration assets: ACME's Bikaner plant, Hygenco's Hisar facility serving Jindal Stainless, a 1 MW unit commissioned at Kandla Port in July 2025 (Indian Infrastructure, 7 May 2026), and an NTPC station at Leh with five fuel-cell buses. Commissioned Indian capacity is given as about 8,000 tpa of hydrogen as at February 2026 (Business Standard, 3 November 2026), while a tracker gives 14,720 tpa of hydrogen and 1,750 tpa of ammonia as at May 2026 (India Infrastructure Research, July 2026). The two are unreconciled and both are set out for that reason. In the GCC no producing green hydrogen asset was found at all.
The commissioning category contains the two assets on which Gulf credibility now rests. NEOM Green Hydrogen Company at Oxagon, held in thirds by ACWA Power, Air Products and NEOM, comprises 2.2 GW of thyssenkrupp Nucera alkaline electrolysis, 600 t/day of hydrogen, about 1.2 Mtpa of green ammonia and 4 GW of wind and solar, at USD 8.4bn as at financial close in May 2023. Construction was reported
complete and the plant in commissioning in August 2026, with commercial operation targeted in 2027 (TradeArabia, 25 August 2026, citing ACWA). That is a slip, and it should be recorded as one: Air Products had filed on-stream in 2026 in its Form 10-K for FY2023 (SEC). No cost overrun has been reported.
The pipeline sorted by what is actually true, the short list of binding offtake, the only prices anyone has published, and who is in fact obliged to buy.
Almost nothing in the GCC or India is producing. The status word attached to a project carries more commercial information than its capacity figure, and nearly all the headline capacity sits in the weakest status of all.
What the pipeline shows is that almost nothing in the GCC or India is producing hydrogen or ammonia at scale, and that the status word attached to a project tells a buyer far more than its capacity figure does. The table below sorts the two regions into the only five categories that carry commercial meaning: producing, commissioning, under construction, FID taken, and announced only. Nearly all the capacity routinely quoted sits in the last of them. It is to be noted that a figure in the announced column is a statement of a developer's intention, carrying no date on which an offtaker or an equity subscriber could rely.
| Project | Location | Scale as stated | Status as at November 2026 | Source and date |
|---|---|---|---|---|
| ACME Bikaner demonstration plant | India | Demonstration scale, capacity not stated | Producing | PIB, 2021 |
| Hygenco Hisar | India | Supply to Jindal Stainless | Producing | Developer material |
| Kandla Port unit | India | 1 MW | Producing, commissioned July 2025 | Indian Infrastructure, 7 May 2026 |
| NTPC Leh station | India | Station and five fuel-cell buses | Producing | Scheme record |
| NEOM Green Hydrogen Company, Oxagon | Saudi Arabia | 2.2 GW alkaline, 600 t/day hydrogen, about 1.2 Mtpa green ammonia | Commissioning complete August 2026, commercial operation targeted 2027 | TradeArabia, 25 Aug 2026, citing ACWA |
| QAFCO 7, Mesaieed | Qatar | 1.2 Mtpa blue ammonia, about 1.5 Mtpa carbon capture | Construction from Nov 2024, Q2 2026 start reaffirmed, start-up not public | QatarEnergy and QAFCO, 31 Aug 2022; QCIntel |
Four projects are genuinely under construction. TA'ZIZ Ruwais, 1.0 Mtpa of blue ammonia held by TA'ZIZ, Fertiglobe, Mitsui and GS Energy, reached financial close on USD 1.39bn for two Ruwais projects on 4 November 2026, over USD 460m of it for the ammonia plant (The National, 4 November 2026). The ammonia EPC, with Tecnimont, is over 90 per cent complete and first production is put at 2027 by Fertiglobe's chief executive (Zawya, around 9 November 2026). No binding third-party offtake has been published. ACME Duqm phase 1 is past 50 per cent complete with electrolysers ordered and first molecule in Q1 2027, at a capital value that is not public (ACME via Oman Observer, 15 December 2025).
| Project | Location | Scale as stated | Status as at November 2026 | Source and date |
|---|---|---|---|---|
| TA'ZIZ Ruwais low-carbon ammonia | UAE | 1.0 Mtpa blue ammonia | Under construction, ammonia EPC over 90 per cent complete, first production 2027 | Zawya, c. 9 Nov 2026 |
| ACME Duqm phase 1 | Oman | 100,000 tpa green ammonia | Under construction, past 50 per cent, first molecule Q1 2027 | ACME via Oman Observer, 15 Dec 2025 |
| AM Green Kakinada | India | 1 Mtpa ammonia in two trains, 1.3 GW alkaline | Under construction, FID end-August 2024, first major erection 17 Jan 2026 | AM Green via Ammonia Energy, Feb 2026 |
| Hygenco Gopalpur | India | Initial phase, capacity not stated | Under construction, initial phase targeted December 2026 | Developer material |
| Indian Oil Panipat | India | 10,000 tpa hydrogen, 25-year build-own-operate supply | FID taken, early works; commissioning conflicted between Dec 2026 and Dec 2027 | L&T exchange filing, 21 Jul 2025 |
| ACWA Power Yanbu | Saudi Arabia | About 2.5 Mtpa green ammonia on ACWA's own material | Announced only, FEED awarded July 2025, no FID, operation targeted 2030 | ACWA; AGBI, Feb 2026 |
| SABIC Agri-Nutrients Jubail | Saudi Arabia | 1.2 Mtpa blue ammonia after a gas allocation reset | Announced only, FID and EPC still to come | Exchange filing, 26 Mar 2026 |
| Project | Location | Scale as stated | Status as at November 2026 | Source and date |
|---|---|---|---|---|
| Hydrom awarded blocks: Amnah, Green Energy Oman, Hyport Duqm, SalalaH2, EDF and J-POWER and Yamna, Actis and Fortescue | Oman | About 200 ktpa, 150 ktpa, over 50 ktpa by 2029, over 1 Mtpa ammonia, about 178 ktpa; one not stated | Announced only, all pre-FID; first FID expected 2026 or 2027 | Hydrom, accessed 10 Nov 2026; MEES, 7 Mar 2025 |
| NTPC Pudimadaka | India | 7 GW renewables, about 1,400 tpd hydrogen | Announced only, jetty detailed-project-report tendering, no FID | Renewable Watch, 25 May 2026 |
| Adani Mundra | India | 1 Mmtpa phase 1 | Announced only; sole commissioned asset is a 5 MW off-grid pilot at Kutch | AEL intimation, 23 Jun 2025; T&D India, 2026 |
| Reliance Jamnagar electrolyser gigafactory | India | Capacity not stated | Announced only, operational by end-2026 claimed, no 2026 update | Reliance AGM, Aug 2025 |
Capacities, capital values and dates in the announced rows trace to developer material or press releases, not to filings or awards. NEOM's USD 8.4bn, NTPC Pudimadaka's Rs 1.85 lakh crore and Reliance's 3 Mtpa by 2032 are all of that character.
Qatar's QAFCO 7 at Mesaieed is the blue equivalent: 1.2 Mtpa with about 1.5 Mtpa of carbon capture, an EPC of about USD 1bn, FID in 2022, construction from November 2024 and a Q2 2026 start reaffirmed in May 2025 (QatarEnergy and QAFCO, 31 August 2022; QCIntel). Actual start-up is not public, and in November 2026 that silence is itself the finding. QatarEnergy is the sole offtaker, an intra-group arrangement rather than third-party offtake, so the project proves engineering rather than demand.
AM Green Kakinada carries the only published FID date in the entire Indian pipeline, taken at the end of August 2024, for 1 Mtpa of ammonia on 1.3 GW of John Cockerill alkaline capacity, with CertifHy RFNBO pre-certification and first major equipment erection on 17 January 2026. First supply has nonetheless slipped from the second half of 2026 to 2028 on AM Green's own material. Indian Oil Panipat sits one rung lower, at
FID with early works: 10,000 tpa of hydrogen on a 25-year supply awarded to L&T Energy GreenTech on 21 July 2025, with commissioning dates that conflict between December 2026 and December 2027.
The announced column is where the large numbers live. ACWA Power Yanbu, at about 2.5 Mtpa of green ammonia on ACWA's own material, has FEED awarded in July 2025 and a 2030 operation target, but neither capital value nor electrolyser capacity is public. SABIC Agri-Nutrients Jubail, reset to 1.2 Mtpa of blue ammonia on a gas allocation change with FID and EPC still to come (exchange filing, 26 March 2026), is the best-sourced Saudi blue datapoint available. Oman's awarded blocks are all pre-FID, with equity splits and capital values undisclosed in every case.
India's announced tier requires the same scepticism. Adani Mundra claims 1 Mmtpa in phase 1, yet its only commissioned asset is a 5 MW pilot at Kutch and MNRE has stated there is no visibility of projects in the Mundra region.
One scheme-level fact governs the whole Indian column. MNRE told Parliament that as at July 2026 no electrolyser manufacturing facility and no green hydrogen production facility supported under SIGHT had been fully commissioned, so no incentive had been claimed or paid, and that Rs 292 crore of the Rs 19,744 crore mission outlay had been spent by August 2026, under 1.5 per cent (Business Standard, 12 August and 3 November 2026). The same reply gives awarded totals of 3,000 MW per year of electrolyser manufacturing to 15 companies and 756,100 tpa of hydrogen production to 16 companies, the latter conflicting with the 862,000 tpa circulated in 2025 reporting.
The 756,100 tpa and 862,000 tpa Indian awarded totals could not be reconciled from any primary source. Both are reported here without preference.
The takeaway for anyone signing an offtake agreement or an equity cheque is a procedural one: the capacity number in a developer's deck is the least load-bearing item in it. The status word, the published FID date, the EPC completion percentage and the identity of the offtaker are the load-bearing items, and in both regions those four fields are empty more often than filled. A pipeline in which one Indian project has a published FID date, and no Gulf green asset is yet producing, is a set of options rather than a supply curve.
Take the binding Gulf-origin agreements first, because there are only two. Air Products takes all NEOM ammonia for 30 years on an exclusive basis, described in its SEC filings as fixed-price with the figure withheld. A downstream marketing agreement with Yara, finalised in 2026 and paid on commission, covers volumes Air Products does not itself reconvert (Air Products 8-K, FY2026). Second, ACME Duqm has a binding 15-year
agreement with Yara Clean Ammonia for 100,000 tpa, the entire phase 1 output, RFNBO-compliant and supplied from 2027, at a price not public. That is the complete list, and no binding GCC-origin offtake with any Japanese or Korean buyer exists on the public record.
Offtake, not capacity and not cost, is what the sector is short of. The list of binding agreements is short enough to be printed in full, and the public record of retreats names the constraint explicitly.
The constraint on this sector is offtake, and the position can be proved by name rather than asserted. The complete list of binding GCC-origin green offtake found runs to two agreements. The list of binding India-origin agreements is longer but still countable on two hands. Against that sits a much longer list of memoranda, letters of intent and single cargoes that are routinely reported as offtake and are not. Most significant of all, the retreats are explicit: Aramco cut its 2030 blue ammonia target from 11 Mtpa to about 2.5 Mtpa expressly for want of long-term offtake (Energy Intelligence, March 2025). Exactly one Gulf green ammonia project is building against a signed binding offtake, and it is Indian-owned.
| Buyer | Seller or origin | Volume | Tenor | Price | Legal character |
|---|---|---|---|---|---|
| Air Products | NEOM Green Hydrogen Company, Saudi Arabia | All ammonia produced | 30 years, exclusive | Fixed price, figure withheld | Binding |
| Yara Clean Ammonia | ACME Duqm, Oman | 100,000 tpa, the entire phase 1 output, RFNBO-compliant | 15 years, from 2027 | Not public | Binding |
| Yara, as marketer | Air Products, for NEOM volumes not reconverted | Not disclosed | Not disclosed | Commission basis, no rate disclosed | Binding, finalised 2026 |
| Thirteen Indian fertiliser plants, via SECI | Indian producers | 724,000 tpa awarded, 670,000 tpa across 11 plants signed March 2026 | 10 years, fixed price | Rs 49.75 to 64.74 per kg | Binding where signed; two plants unsigned |
| Uniper | AM Green, India | Up to 500,000 tpa RFNBO ammonia from 2028 | Long-term, tenor not public | Not public | Binding, announced January 2026 |
| JERA | AM Green, India | 70,000 tpa | Five years | Not public | Binding head of agreement |
| Buyer | Seller or origin | Volume | Tenor | Price | Legal character |
|---|---|---|---|---|---|
| Hintco | AM Green phase 2, India | Minimum EUR 585m of ammonia over the term | 10 years from 2029 | Per-tonne price withheld | Binding, awarded 30 September 2026 |
| IHI | ACME, India | 405,000 tpa for Japan | Long-term, tenor not public | Not public | Binding, reported July 2026 |
| Mitsubishi Gas Chemical | ACME Paradip, India | 100,000 tpa green methanol | 10 years | Not public | Binding |
| EnBW, VNG and Rostock Port | ACWA Power | No volume stated | None | None | Memorandum, 1 February 2026 |
| OMV | Masdar | Not stated | None | None | Letter of intent, 30 April 2025 |
| Yara | Greenko ZeroC | Not stated | None | None | Non-binding |
| RWE, BASF and Keppel | AM Green | Not stated | None | None | Non-binding |
| Mitsui | ADNOC and Fertiglobe | One certified bulk commercial cargo | None | Not public | Cargo, not term offtake |
| Japanese buyer, 2020 | Aramco | 40 t demonstration cargo | None | Not public | Demonstration |
| South Korean buyer, 2022 | Aramco | About 25,000 t | None | Not public | Spot, no term offtake |
The India-origin list is longer and more commercially convincing, in part because its anchor is a state procurement. SECI's SIGHT Mode 2A Tranche I awarded 724,000 tpa across 13 fertiliser plants on 10-year fixed-price agreements with SECI as intermediary procurer, and agreements for 670,000 tpa across 11 plants were signed in March 2026, which leaves two unsigned (SECI; Business Standard, 3 November 2026). AM Green has a binding agreement with Uniper for up to 500,000 tpa of RFNBO ammonia from 2028, announced in January 2026, superseding a non-binding 250,000 tpa arrangement of 2023; a binding head of agreement with JERA for 70,000 tpa over five years; and the sole winning bid in H2Global's round two Asian lot, under which Hintco will buy a minimum of EUR 585m of ammonia over ten years from 2029.
ACME holds a long-term binding agreement with IHI for 405,000 tpa of India-origin ammonia destined for Japan, supported by METI's contract for difference and reported in July 2026, and a ten-year agreement with Mitsubishi Gas Chemical for 100,000 tpa of green methanol from Paradip. MNRE reports aggregate firm offtake of about 555,610 tpa as at October 2026 (Business Standard, 29 October 2026), though figures vary between reports. It is to be noted that the tenors of the ACME and Yara Duqm, ACME and IHI, and AM Green and Uniper agreements are stated only as long-term, and that no contract price among them is public.
Against that short list sits the much longer list of instruments that carry the word offtake in reporting and none of its obligations. ACWA Power's arrangement with EnBW, VNG and Rostock Port, dated 1 February 2026, is a memorandum with no volume and no price (EnBW). Hydrom's arrangements with H2Global and with VNG are memoranda. The Oman to Netherlands to Germany liquid hydrogen corridor of April 2025 is a joint development agreement for infrastructure between eleven parties, not an offtake at all. Masdar and OMV signed a letter of intent on 30 April 2025. Greenko ZeroC with Yara, and AM Green with RWE, BASF and Keppel, are described as non-binding by S&P Global. Aramco has no term offtake of any kind: only a 40 t demonstration cargo to Japan in 2020 and about 25,000 t to South Korea in 2022.
One frequently cited item deserves separating out. ADNOC and Fertiglobe's May 2024 arrangement with Mitsui is a certified bulk commercial cargo, not a term agreement, and Mitsui is simultaneously an equity partner in TA'ZIZ. Volume, tenor and price are not public. A cargo sold to a party that also owns part of the plant says very little about the price at which independent demand can be found.
Contracted tonnage for Yara under the NEOM marketing agreement and for Mitsui under TA'ZIZ is not public. TA'ZIZ equity percentages and any binding third-party ammonia offtake for that plant could not be sourced.
The retreats are where the constraint is named rather than inferred. Aramco's reduction of its 2030 blue ammonia target from 11 Mtpa to about 2.5 Mtpa was attributed explicitly to the absence of long-term offtake (Energy Intelligence, March 2025). ADNOC and Fertiglobe rephased Project Rabdan and abandoned an early-2026 FID on weak buyer interest (Energy Intelligence and QCIntel, 2026). Two of Oman's nine awarded blocks were terminated by mutual agreement in late 2025 (Hydrom, 8 December 2025): BP exited Duqm, a 1.5 GW position, and HyDuqm, reported at
USD 6.7bn, was concluded on the same date. None of these is a technical failure: each is a demand failure, stated as such by the parties.
The demand side has also been losing sinks faster than it has been adding them. Masdar scrapped the 100 MW Emsteel electrolyser after bids had been received, on 1 August 2026, and redirected 6 GW of solar earmarked for a 350,000 tpa ammonia project to data centres (MEED and Enterprise, August 2026). ArcelorMittal cancelled hydrogen direct-reduced iron at Bremen and Eisenhuettenstadt in June 2025, revoking a claim to about EUR 1.3bn of German subsidy and removing a demand sink in the process. Yara shelved green conversion at Porsgrunn and Sluiskil, together 530,000 tpa, in October 2024. Korea's second CHPS auction was cancelled on the bid deadline day itself, 17 October 2025, and SECI scrapped its two hub proposals in July 2025.
The takeaway is that the supply question in both regions has been answered and the demand question has not. A buyer evaluating a project should therefore treat the identity and legal character of the existing offtake as the first diligence item, ahead of capacity, cost and technology. Where the answer is a memorandum, a letter of intent, a single cargo or an intra-group sale, the project has not yet demonstrated that anyone will pay for its output, and on the published record that description fits the overwhelming majority of announced capacity in both regions.
Three price series exist in or adjacent to these two regions, and all three are Indian or European. Every GCC-origin contract price is undisclosed, without exception.
The position on price is that India's SECI auction is the only discovered green ammonia price series in either region, that H2Global's pilot auction is the only published green ammonia clearing price outside India, that Indian refinery hydrogen has two conflicting published prints, and that every GCC-origin contract price is undisclosed without exception. That last sentence is not an omission in the research. It is the finding. A buyer cannot benchmark a Gulf cargo against any published Gulf contract, because none has been published, including the Air Products and NEOM agreement, which filings confirm is fixed-price while withholding the figure.
Two features of that result matter more to a buyer than the headline average. The first is the spread. A band from Rs 49.75 to Rs 64.74 per kg, on the same product under the same scheme in the same tranche,
| Award or auction | Date | Volume | Price as published | Tenor |
|---|---|---|---|---|
| SECI SIGHT Mode 2A Tranche I, weighted average | 2025 | 724,000 tpa awarded across 13 plants | Rs 53.27 per kg, about USD 604 per t, delivered | 10 years, fixed |
| SECI Mode 2A, lowest lot: ACME to IFFCO Paradeep | 2025 | 100,000 tpa | Rs 49.75 per kg | 10 years, fixed |
| SECI Mode 2A, highest lot: ACME to Indorama Haldia | 2025 | 20,000 tpa | Rs 64.74 per kg | 10 years, fixed |
| H2Global pilot auction: Hintco to Fertiglobe, Egypt origin | 2024 | Minimum 40,000 tpa | EUR 811 per t ex-factory, EUR 1,000 per t delivered Europe | Deliveries 2028 to 2033 |
| H2Global round two, Asian lot: Hintco to AM Green phase 2 | 30 September 2026 | Minimum EUR 585m of ammonia over the term | Per-tonne price withheld pending publication of all lots | 10 years from 2029 |
| Indian Oil refinery hydrogen auction | Reported via IEEFA, June 2025 | Volume not stated | Rs 397 per kg, about USD 4.67 per kg | 25 years |
| Lowest reported Indian refinery hydrogen print | Business Standard, 3 November 2026 | Volume not stated | Rs 279 per kg | Not stated |
| Indian refinery awards as tabulated by the IEA | Global Hydrogen Review 2025 | Volume not stated | Near Rs 328 per kg, about USD 3.8 per kg | Not stated |
| Every GCC-origin contract, including Air Products and NEOM | To November 2026 | Includes all NEOM ammonia | Undisclosed; filings confirm fixed price without giving it | 30 years for NEOM |
is about 30 per cent, and it tracks lot size: the cheapest lot is five times the volume of the dearest. Scale and the identity of the receiving plant are therefore doing significant work in the price. The second is concentration. ACME holds six of the 13 lots and 370,000 tpa of the award, slightly more than half. A series in which one supplier holds half the volume is a real price, but a price discovered in a thin market.
The second series is European and sits one step removed. H2Global's pilot auction, run in 2024, saw Hintco contract Egyptian-origin ammonia from Fertiglobe at EUR 811 per t ex-factory and EUR 1,000 per t delivered into Europe, for a minimum of 40,000 tpa with deliveries from 2028 to 2033, against a subsidy of about EUR 397m (Hintco; Veyt). This remains the only published green ammonia auction clearing price outside India. Its mechanism matters: Hintco buys high and resells at whatever the market will bear, with the
public purse absorbing the difference, so EUR 1,000 per t is the price at which supply could be secured, not the price at which demand was found.
The SECI auction is the centrepiece because it is the only place where competitive bidding has produced a disclosed, contracted, long-dated price for green ammonia in either region. Mode 2A Tranche I awarded 724,000 tpa across 13 fertiliser plants on ten-year fixed-price agreements, with SECI acting as intermediary procurer so that individual fertiliser plants did not have to carry counterparty risk on a first-of-its-kind supplier. The band was Rs 49.75 to 64.74 per kg delivered, with a weighted average of Rs 53.27 per kg, or about USD 604 per t. The lowest lot was ACME supplying IFFCO Paradeep at 100,000 tpa and Rs 49.75 per kg; the highest was ACME supplying Indorama Haldia at 20,000 tpa and Rs 64.74 per kg (SECI; PIB release 2153006, 6 August 2025).
Round two of the same scheme is the single most consequential price disclosure still pending for either region. AM Green's phase 2 won the Asian lot as sole winner on 30 September 2026, with Hintco committing to buy a minimum of EUR 585m of ammonia over ten years from 2029. The per-tonne price has been withheld until Hintco publishes all lots, which it has said it will do before the end of 2026. Until then it exists as a contract value without a unit price, and no inference about the per-tonne figure should be drawn from the aggregate.
The third series is Indian refinery hydrogen, and it is the least clean of the three. Indian Oil's refinery hydrogen auction cleared at Rs 397 per kg, about USD 4.67 per kg, on a 25-year offtake (reported via IEEFA, June 2025). A lowest refinery print of Rs 279 per kg is also reported (Business Standard, 3 November 2026), and the IEA tabulates the Indian refinery awards near Rs 328 per kg, about USD 3.8 per kg, in its Global Hydrogen Review 2025. Three figures, three sources, no reconciliation. What can safely be said is that Indian refinery hydrogen has cleared between Rs 279 and Rs 397 per kg on very long tenors.
The silence from the Gulf is therefore total, and it is worth stating precisely what is silent. No price is public
for the Air Products and NEOM agreement, although SEC filings confirm the arrangement is fixed-price. No price is public for ACME Duqm's supply to Yara. No price is public for the ADNOC and Fertiglobe cargo to Mitsui, nor for any TA'ZIZ volume, nor for any awarded Omani block. Aramco has no term agreement to price. Gas transfer prices to Gulf producers, which decide blue and grey cost entirely, are not public either, so even a reconstructed cost benchmark is unavailable.
A fourth number is often quoted alongside these and is not a price at all. Reliance bid Rs 24.99 per kg for 49,000 tpa in SIGHT Mode 1 Tranche II (JMK Research, 21 March 2025). That is a bid for a production incentive, expressed as the subsidy per kilogram the bidder is willing to accept, and it is not comparable with a delivered contract price. Mixing the two produces a false impression of Indian cost levels.
No Argus or Platts assessed green or blue ammonia level, and no attributable current Middle East free-on-board ammonia print, could be obtained: both sit behind subscription. The Korean CHPS round one clearing price is undisclosed.
The takeaway is uncomfortable but actionable. A buyer negotiating Gulf-origin ammonia today has no regional comparator and must price against Indian and European discovery: a ten-year Indian fixed price averaging about USD 604 per t, a European pilot clearing at EUR 1,000 per t delivered, and an Indian refinery band of Rs 279 to 397 per kg. Each of those references carries a different tenor, a different delivery point and a different subsidy structure, so none is a clean substitute. The practical consequence is that the first party to publish a Gulf contract price will reset the negotiating position for every subsequent cargo, and that until one does, every Gulf quotation is being tested against foreign evidence.
India's discovered green ammonia price sits between about 10 and 31 per cent above grey, depending entirely on which grey reference is chosen. For part of 2026 it sat below spot grey altogether.
The honest statement of the green premium is that it is as much a choice of reference as a measurement. India's discovered green ammonia average of about USD 601 to 604 per t sits against grey ammonia at USD 515 per t as at March 2025, a published gap of about USD 86 to 126 per t, or between roughly 10 and 31 per cent depending on which green print and which grey reference is used (IEEFA, November 2025; JMK Research via pv magazine India, 28 August 2025). IEEFA states about 31 per cent against a grey range of USD 400 to 515 per t. Tampa cost-and-freight
settlements during 2026 ran USD 555 to 825 per t (Fertilizer Daily and QCIntel, January to November 2026), so for part of 2026 the Indian green print was below spot grey and the premium, measured that way, was negative.
| Reference used | Figure | Date and source | Premium that results |
|---|---|---|---|
| Indian discovered green ammonia, weighted average | About USD 601 to 604 per t | SECI Mode 2A award, 2025 | The green side of every comparison below |
| Grey ammonia, single published print | USD 515 per t | As at March 2025 | About USD 86 to 126 per t, or roughly 10 to 31 per cent |
| Grey ammonia, range | USD 400 to 515 per t | IEEFA, November 2025 | About 31 per cent, on IEEFA's own statement |
| Tampa cost-and-freight settlements | USD 555 to 825 per t | Fertilizer Daily and QCIntel, January to November 2026 | Negative for part of 2026 |
| Pre-auction expectation | USD 707 to 715 per t | IEEFA | Nearly twice grey; not realised |
| Green steel, modelled | Rs 55,000 to 65,000 per t | IEEFA, source of the estimate unnamed | About 20 to 30 per cent above conventional |
The green steel figure is a modelled estimate and IEEFA does not name the source of the estimate. It is reported here on that basis and should not be treated as a transacted price.
Why does the number move so violently? Because the green side is a fixed ten-year contract price and the grey side is a volatile spot assessment driven by gas. The Indian green print is a single number, set once, for a decade, under a procurement in which the state intermediated counterparty risk. Grey ammonia at Tampa moved across a range of USD 270 per t within a single year. Comparing a fixed decade-long price with a spot series will produce a different premium every month, and in a soft gas year no premium at all. This is not a presentational trick: it is the arithmetic consequence of comparing two instruments of entirely different character.
The second reason the number moves is that the expected premium was wrong, and by a wide margin. Pre-auction expectation had been USD 707 to 715 per t, described as nearly twice grey (IEEFA). The auction cleared at about USD 604 per t on a weighted average basis. The discovered price came in roughly 15 per cent below the expectation built into most published analysis of Indian green ammonia, so a buyer modelling the premium on pre-auction consensus would have overstated it substantially.
Beneath the contracted prices sits a cost stack that is published only as ranges. IEEFA, in November 2025, gives grey hydrogen at USD 1.5 to 3 per kg against green at USD 3 to 6 per kg, with Indian solar tariffs at Rs 2.5 to 3 per kWh and electrolyser capital cost at Rs 26,500 to 106,000 per kW. That electrolyser range is a factor of four, which is to say that the published cost of green hydrogen in India is dominated by an input nobody can price within a factor of four. GPCA, in March 2025, gives Gulf blue hydrogen at USD 2.8 to 3.5 per kg depending on gas prices, with the gas assumption unstated.
That unstated gas assumption is the single largest hole in the Gulf cost case. Gas transfer prices to Gulf producers decide blue and grey cost entirely, and they are not public. No Gulf grey hydrogen cost is public. No blue ammonia levelised cost is public for either region. A Gulf blue ammonia project therefore presents a cost range whose dominant variable is undisclosed, so the range cannot be independently tested and its lower bound cannot be relied upon.
The IEA's forward projections should be read with the same care. The Global Hydrogen Review 2026 states that by 2030 the Middle East, India and Chile could reach just above USD 3 per kg, and that gas-rich regions could reach potentially just above USD 1 per kg for blue hydrogen, a figure the IEA itself qualifies as largely theoretical, with no live project meeting it at 95 per cent capture. These are modelled projections, not quotations. The assumption annex carrying weighted average cost of capital, capacity factor, electrolyser capital cost and gas price could not be retrieved, so the figures are quoted here as ranges without their assumptions, which is how they should be used.
The frequently quoted framing of USD 4 to 5 per kg falling to USD 2 per kg for India could not be traced to any MNRE or NITI Aayog document. Its only observable anchors are the Rs 397 per kg award and government projections of Rs 260 to 310 per kg reported secondhand. It should not be cited as official.
The IEA's 2025 edition adds a point that cuts directly against the Gulf green case. It tabulates the Indian refinery awards near Rs 328 per kg, about USD 3.8 per kg, and notes that in cheap-gas regions including the Middle East the green-to-grey gap will remain larger, and that carbon capture is likelier to be competitive in the near term. The implication is uncomfortable for green Gulf exporters and comfortable for blue ones: the cheaper the domestic gas, the wider the premium a green molecule must bridge.
The demand-side arithmetic is harsher still. The IEA's 2026 edition states that absent policy support the
maximum acceptable hydrogen cost is below USD 2 per kg for most sector and region combinations, and that for steel the hydrogen cost must be negative to reach parity. Set that against a published green range of USD 3 to 6 per kg and the gap is not a premium to be absorbed in a procurement decision: it is a gap that only a subsidy, a mandate or a carbon price can close.
The takeaway for a buyer is to fix the reference before negotiating the premium. A ten-year fixed price should be compared with a ten-year grey forward view, not with last week's Tampa settlement, and the comparison should be stated with its reference attached every time. Where a counterparty quotes a premium without naming its grey reference and its date, the quotation is not a measurement. On the record for 2025 and 2026, the same Indian green price supports a premium of about 31 per cent, one of about 10 per cent, or none at all.
Across India, the European Union, Japan, Korea and the IMO, binding purchase obligation is far scarcer than the headline targets suggest. India has notified none at all.
The question that decides whether this sector has a market is who is legally obliged to buy, and the answer is that very few parties are. India has notified no green hydrogen or green ammonia consumption obligation for refining, fertiliser or steel. Japan's scheme binds the government to pay rather than any party to buy. Korea's second clean hydrogen auction was cancelled on its bid deadline day. The IMO's Net-Zero Framework is not adopted. The European Union is the exception that proves the point: RED III and FuelEU Maritime are binding, and they are the only instruments in this survey that compel a purchase. Target, obligation and government payment are three different things and are separated below.
| Jurisdiction | Instrument | Legal character | What it actually binds | Status as at November 2026 |
|---|---|---|---|---|
| India | Green hydrogen or ammonia consumption obligation | None notified | Nothing | Discussed for years and yet to be notified, per counsel quoted in Business Standard, 3 November 2026 |
| India | SECI SIGHT Mode 2A procurement | Contract | The 13 fertiliser plants that signed | 670,000 tpa of the 724,000 tpa awarded signed March 2026 |
| Jurisdiction | Instrument | Legal character | What it actually binds | Status as at November 2026 |
|---|---|---|---|---|
| India | Carbon Credit Trading Scheme | Binding intensity targets | 208 units notified in tranche 2; about 490 entities in total | Notified 8 October 2025 and mid-January 2026; iron and steel still draft at a revised notification of 26 June 2026 |
| India | Green Steel Taxonomy, S.O. 763(E) | Labelling | Nobody to buy anything | Notified December 2024 |
| India | Green Hydrogen Certification Scheme | Voluntary labelling at 2 kg CO2e per kg | Nothing | MNRE, April 2025; portal live 17 June 2026 |
| European Union | RED III Article 22a | Binding on member states | 42 per cent of industrial hydrogen to be RFNBO by 2030, 60 per cent by 2035, 1 per cent in transport | Transposition due 21 May 2025; only eight of 27 member states had both industry and transport mandates in national law as at mid-2026 |
| European Union | FuelEU Maritime | Binding on ship operators | Minus 2 per cent well-to-wake intensity from 2025, minus 6 per cent from 2030, minus 14.5 per cent from 2035 | In application since 1 January 2025 |
| Japan | Hydrogen Society Promotion Act 2024 | Binds the government to pay | 15 years of contract-for-difference price-gap support against the incumbent fuel | Certifications announced 30 September 2025, 19 December 2025 and 27 March 2026 |
| Japan | 20 per cent ammonia co-firing | Policy target from the 2021 Strategic Energy Plan | Nothing | Four coal units plan commercial 20 per cent operation between FY2027 and FY2030; Hekinan Unit 4 is the only unit to have demonstrated it at scale |
| Korea | CHPS clean hydrogen auction | Contract where awarded | 750 GWh of the 6,500 GWh per year offered in round one | Round two, 3,000 GWh per year, cancelled 17 October 2025; 2026 volume cut to 500 GWh per year |
| IMO | Net-Zero Framework | Not adopted | Nothing yet | Adoption sits at the resumed extraordinary session of 4 December 2026 |
India first, because the gap between perception and position is widest there. There is no notified green hydrogen or green ammonia consumption obligation for refining, fertiliser or steel. Business Standard, on 3 November 2026, quotes counsel directly to the effect that the obligation has been discussed for years but is yet to be notified. The 5 per cent by 2029-30 and 10 per cent by 2035-36 refinery figures that circulate widely could not be traced to any notification or official proposal: all circulating numbers are drafts. One blog claim of a notified purchase obligation is contradicted by the primary record and should be treated as wrong.
What India has instead is procurement, which is stronger for the suppliers who win it and weaker for the market as a whole. The 724,000 tpa SECI ammonia award is a contract, binding on the 11 plants that signed in March 2026. A further 30,000 tpa of refinery hydrogen has been procured for IOCL, BPCL, HPCL and Numaligarh, and SECI has stated an intent to procure 1 Mtpa more for fertiliser (IGHA, August 2026). An intent to procure is not an obligation to consume, and procurement can be paused, retendered or scrapped, as SECI's own cancellation of its two hub proposals in July 2025 demonstrates.
India does, however, have binding adjacent instruments, and these are where pressure will first be felt. The Carbon Credit Trading Scheme notified intensity targets on 8 October 2025 for aluminium, cement, chlor-alkali and pulp and paper, and in mid-January 2026 for refining, petrochemicals and textiles, covering 208 units, with compliance years 2025-26 and 2026-27 measured against an FY2023-24 baseline and about 490 entities in total; iron and steel remained draft at a revised notification of 26 June 2026 (ICAP). The scheme publishes no hydrogen volume, so it creates a price on intensity without specifying the abatement route. The Green Steel Taxonomy is notified but binds nobody to buy, and the Green Hydrogen Certification Scheme is voluntary labelling at a 2 kg CO2e per kg threshold.
Japan's Hydrogen Society Promotion Act of 2024 is the most generous instrument in this survey and the most
frequently misdescribed. It provides fifteen years of contract-for-difference support covering the price gap against the incumbent fuel, with certifications announced on 30 September 2025, 19 December 2025 and 27 March 2026. It binds the government to pay. It does not bind any party to buy. The much-quoted 20 per cent ammonia co-firing figure is a policy target from the 2021 Strategic Energy Plan and not a binding obligation: only four coal units plan commercial 20 per cent operation between FY2027 and FY2030, and Hekinan Unit 4 is the only unit to have demonstrated 20 per cent co-firing at scale (IEEFA).
Korea has moved in the opposite direction. The CHPS mechanism is a contracting scheme, and its first round offered 6,500 GWh per year and awarded 750 GWh, about 11 per cent, to KOSPO Samcheok on imported blue ammonia supplied by Samsung C and T, at an undisclosed clearing price. Round two, at 3,000 GWh per year, was cancelled by KPX on 17 October 2025, on the bid deadline day. The volume was then cut to 500 GWh per year of clean hydrogen for 2026, and coal-ammonia co-firing will be excluded from CHPS altogether (Seoul Economic Daily, 9 June 2026). A buyer counting on Korean ammonia co-firing demand is counting on a scheme that has shrunk by an order of magnitude and has removed the use case.
The IMO is the largest single potential demand creator in this survey and it has not voted. The Net-Zero Framework was approved in principle at MEPC 83 in April 2025, with 63 in favour. Adoption failed at the October 2025 extraordinary session when a Saudi motion to adjourn for twelve months passed 57 to 49. MEPC 84, held from 27 April to 1 May 2026, did not adopt the framework but carried it forward as the sole basis for work. Adoption now sits at the resumed extraordinary session of 4 December 2026 (Carbon Brief Q and A, 5 May 2026, updated 15 July 2026; Seatrade and Lloyd's Register). If adopted in December 2026, the earliest entry into force is 1 March 2028 under MARPOL tacit acceptance.
IMO primary documents could not be consulted: imo.org served empty responses. The IMO positions above rest on Carbon Brief, Seatrade and Lloyd's Register reporting.
The approved trajectory, should it be adopted, would require a base intensity reduction of 4 per cent in 2028 rising to 30 per cent in 2035, with a direct target of 17 to 43 per cent against 2008, and remedial units priced at USD 100 and USD 380. That would be a meaningful carbon price on marine fuel. It is also conditional on a vote that has already been adjourned once at the instance of a Gulf state. FuelEU Maritime, by contrast, is binding and has been in application since 1 January 2025.
The takeaway is that one jurisdiction compels purchase and the rest either pay, aspire or have not decided. For an exporter from the Gulf or India, that concentrates commercial risk severely: the only reliable obligated demand is European, and it is being transposed into national law slowly. A project whose business case assumes Japanese, Korean or Indian obligation is assuming something that does not currently exist in law.
The only binding demand in this survey is European, which makes European rules the gate every Gulf and Indian exporter must pass. RED III, the RFNBO delegated act and CBAM decide what a molecule is worth.
Because Europe is the only jurisdiction in this survey that compels a purchase, European rules, and not Gulf or Indian ambition, determine what an exported molecule is worth. Three instruments do the work: RED III Article 22a, which obliges member states to reach 42 per cent RFNBO in industrial hydrogen by 2030; the RFNBO delegated act, which decides whether a molecule qualifies at all, with hard dates in 2028 and 2030; and CBAM, whose definitive regime began on 1 January 2026 with hydrogen in scope and whose free allocation ladder descends to zero by 2034. A Gulf exporter is selling into that, and compliance with it is the product.
| Year | CBAM free allocation retained | Other binding change in the same year |
|---|---|---|
| 2026 | 97.5 per cent | Definitive regime begins 1 January, hydrogen in scope; the 50 t de minimis does not apply to hydrogen or electricity |
| 2027 | Not stated in the sources used here | CBAM certificate pricing moves to the weekly average ETS price from 1 January |
| 2028 | Not stated in the sources used here | RFNBO additionality grandfathering ends 1 January; scheduled review of the RFNBO methodology in July |
| 2030 | About 51.5 per cent | Hourly temporal correlation applies from 1 January; RED III 42 per cent industrial and 1 per cent transport targets bite; FuelEU zero-emission at berth in AFIR-covered ports from 1 January |
| 2034 | Zero | FuelEU Article 5(3) 2 per cent RFNBO sub-quota applies from 1 January if the 2031 RFNBO share is below 1 per cent |
CBAM factors for 2027, 2028 and intervening years were not retrieved and are not interpolated here. A reported Commission proposal to reinstate 15 per cent free allocation from 2028 and extend the phase-out to 2038 is uncorroborated and should not be planned around.
RED III Article 22a is binding on member states and sets the demand that a Gulf or Indian exporter is ultimately selling into: renewable fuels of non-biological origin to be at least 42 per cent of industrial hydrogen
by 2030 and 60 per cent by 2035, with a separate 1 per cent transport target (Directive (EU) 2023/2413). Two qualifications are commercially significant. Transposition was due on 21 May 2025 and only eight of 27 member states had both industry and transport mandates in national law as at mid-2026 (European Hydrogen Observatory, 2026). And refinery hydrogen falls to the 1 per cent transport obligation rather than the 42 per cent, which removes Europe's largest existing hydrogen consumer from the stronger mandate.
The delegated act, 2023/1184, is the gate itself, and it is already binding. Additionality does not apply to electrolysers commissioned before 1 January 2028, which creates a hard commissioning deadline with real economic value attached: an electrolyser energised on 31 December 2027 is grandfathered and one energised a day later is not. Temporal correlation is monthly to the end of 2029 and hourly from 1 January 2030, which is a significant change in how a plant must be dispatched and contracted against renewable generation. A Commission revision extending the transition to 2035 is a proposal only and carries no legal effect at present.
For an exporter, the practical effect of those two dates is that the delegated act, rather than any Gulf or Indian scheme, sets the construction schedule. A project aiming to sell RFNBO-compliant ammonia into the European compliance market has a commissioning target of 1 January 2028 for the grandfathering benefit and a dispatch redesign due by 1 January 2030 for hourly matching. It is to be noted that ACME Duqm is pre-certified as RFNBO-compliant and AM Green Kakinada holds CertifHy RFNBO pre-certification: competitive advantage currently lies in certification rather than capacity. A scheduled review of the RFNBO methodology falls in July 2028, which introduces a known regulatory event inside the life of every contract now being signed.
CBAM's definitive regime began on 1 January 2026 with hydrogen in scope, and its design reaches further into Gulf and Indian industry than hydrogen alone. The 50 t de minimis threshold does not apply to hydrogen or electricity, so there is no small-consignment
escape. The CBAM factor, which is the proportion of free allocation retained by European producers and therefore the proportion of the carbon cost not yet charged at the border, starts at 97.5 per cent in 2026, falls to about 51.5 per cent in 2030 and reaches zero from 2034. From 1 January 2027, certificate pricing moves to the weekly average ETS price, which removes a smoothing mechanism and transmits European carbon price volatility directly into an exporter's landed cost.
The consequence for the Gulf and India is that CBAM binds European importers and therefore binds exporters of ammonia, fertiliser and steel commercially, even though it imposes no obligation on them in their own law. It works by changing what a European buyer can afford to pay for a high-carbon tonne rather than by prohibiting its production. As the factor descends, the advantage of a certified low-carbon tonne rises mechanically, without any new policy being made. That is the most predictable source of demand growth available to either region.
It should be recorded that the rulebook is already deciding projects, and not only at the margin. Equinor and Linde cancelled H2M Eemshaven, a 1 GW project at about 210,000 tpa, as reported on 20 February 2026, citing among other things RED III's treatment of blue hydrogen (Hydrogen Insight). That is a European project cancelled by European rules, which is the clearest available evidence that the delegated act and the directive are operative constraints rather than aspirations. An exporter choosing between blue and green for a European-facing project should treat that cancellation as the governing precedent.
Shipping adds a second European layer. FuelEU Maritime binds operators of ships above 5,000 gross tonnes calling at European ports under any flag, with its first annual report, covering 2025, due on 31 January 2026 and recurring thereafter. The intensity trajectory runs minus 2 per cent from 2025, minus 6 per cent from 2030 and minus 14.5 per cent from 2035. RFNBO energy counts double to 31 December 2033, which front-loads the value of early RFNBO bunkering, and from 1 January 2030 zero-emission at berth applies in AFIR-covered ports. Under Article 5(3), a 2 per cent RFNBO sub-quota applies from 1 January 2034 if the RFNBO share in 2031 is below 1 per cent, which converts a shortfall into a mandate automatically.
The takeaway is that the European rulebook, not the Gulf's resource endowment, is the asset that a Gulf or Indian project most needs to secure. Certification status, commissioning date relative to 1 January 2028, dispatch design relative to 1 January 2030, and the CBAM exposure of the buyer's own product decide whether a tonne sells into obligated demand at a premium or into unobligated demand at the grey price. Very few announced projects in either region have addressed all four, and the two furthest advanced on certification are both Indian-owned.
The IEA's own 2030 projection has fallen 45 per cent in two editions. Every major forecaster has cut, and the write-offs carry figures filed with regulators.
The most useful single piece of evidence available to a buyer is not any forecast but the direction in which the forecasts have moved. The IEA's estimate of potential 2030 low-emissions hydrogen production from announced projects rose from over 17 Mt in 2021 to 49 Mt in 2024, then fell to 37 Mt in 2025, the first ever decline, and to 27 Mt in 2026: a cut of 45 per cent in two editions. BloombergNEF cut its 2030 global clean hydrogen supply estimate from 16.4 Mt per year to about 5.5 Mt per year. DNV cut its mid-century clean hydrogen forecast by 45 per cent against its 2022 forecast. The sector is not being revised downward by its critics. It is being revised downward by its own institutions.
| IEA edition | Potential 2030 low-emissions production from announced projects | Share or volume at final investment decision |
|---|---|---|
| 2021 | Over 17 Mt | Not stated |
| 2022 | 16 to 24 Mt | 4 per cent |
| 2023 | 38 Mt | 4 per cent |
| 2024 | 49 Mt | 3.4 Mtpa |
| 2025 | 37 Mt, the first ever decline | 9 per cent |
| 2026 | 27 Mt | Only about 300 ktpa of new FID taken in the year |
Read as a series rather than as a set of annual headlines, the IEA numbers describe a pipeline that inflated and then deflated while the committed portion of it barely moved. The share at FID was 4 per cent in 2022 and 4 per cent in 2023; the 2024 edition gives 3.4 Mtpa at FID; the 2025 edition gives 9 per cent. The 2026 edition is the sharpest line available to a buyer anywhere in the public record: committed plus strong-potential 2030 volume fell from 10 Mt to just above 6 Mt, only about 300 ktpa of new FID was taken, and 22 Mt of announced production may lose any chance to
begin operation by 2030 if investment decisions are not taken by early 2027.
Two further figures from the same edition deserve to be set beside one another. Signed agreements in 2025 totalled about 1.7 Mt, of which only about 20 per cent is firm. Installed electrolysis doubled in 2025 to over 4 GW, with China accounting for about 75 per cent of additions. The first says that the paper flow has continued while its binding content has not. The second says that the electrolysis actually being installed is being installed somewhere other than the Gulf or India.
The private forecasters tell the same story in their own terms. BloombergNEF cut global clean hydrogen supply in 2030 from 16.4 Mt per year, as at 14 May 2024, to about 5.5 Mt per year, as at 5 November 2025, which in BNEF's own words is half of what was expected just a few years ago. The Hydrogen Council and McKinsey, in the Global Hydrogen Compass of September 2025, record the most instructive pairing in the literature: committed post-FID investment to 2030 rose from USD 10bn in December 2020 to USD 110bn in July 2025, while feasible 2030 supply fell to 9 to 14 Mtpa, or 20 to 30 per cent of announced. Binding offtake stood at 3.6 Mtpa, and over 50 projects had been publicly cancelled in the eighteen months to July 2025.
That pairing is the central fact of the sector: committed capital rose elevenfold while the volume it is expected to produce fell. The committed money has gone into fewer, larger, better-contracted projects, and the long tail has been written down. Westwood quantifies the tail in Europe: 23 projects and 29.2 GW cancelled or stalled by the end of 2024, 20.3 per cent of a 111.5 GW pipeline (17 December 2024), and then an expectation that only 17 per cent of the EU pipeline, about 12 GW, would materialise by 2030 (1 April 2025).
The write-offs cannot be presented as sentiment, because they are filed with regulators and carry numbers. Air Products cancelled Massena, World Energy Paramount and a Texas carbon monoxide project on 24 February 2025 at a pre-tax charge not exceeding USD 3.1bn (8-K). It then exited the Louisiana Clean Energy Complex and Casa Grande, decided on 26 June 2026, at a pre-tax charge of up to USD 2.9bn and a cash cost not exceeding USD 925m, and the stated reason is the single most important sentence in this
survey: it could not secure firm hydrogen and nitrogen offtake and could not divest the ammonia loop (8-K Item 2.06, 30 June 2026).
A circulating figure of USD 3.2bn for the Air Products charge exceeds the company's own stated ceiling and is unverified. NEOM was explicitly retained in both filings and is not part of either write-off.
BP cancelled HyGreen Teesside, 500 MW, in March 2025 and scrapped H2Teesside, up to 1.2 GW, in 2026, citing a competing data centre consent and deteriorating demand across the hydrogen sector. The data centre point recurs: Masdar redirected 6 GW of solar earmarked for ammonia to artificial intelligence data centres in August 2026. Equinor and Linde cancelled H2M Eemshaven, 1 GW and about 210,000 tpa, as reported on 20 February 2026. Shell scrapped its Rotterdam biofuels and blue hydrogen complex in 2025 after USD 1.4bn had been spent. It should be recorded that Holland Hydrogen II is not confirmed cancelled and that Holland Hydrogen I, 200 MW, connected to the Dutch network in March 2026, so the European record is not uniformly negative.
What should a buyer take from a sector that has revised itself down by this much? Not that it will fail, but that the base rate of delivery against announcement is now measurable and low. On the Hydrogen Council's own arithmetic, feasible 2030 supply is 20 to 30 per cent of announced. On Westwood's, 17 per cent of the European pipeline will materialise. On the IEA's, 9 per cent of the 2025 pipeline had reached FID. Three independent estimates converge on between one fifth and one tenth of announced capacity.
The takeaway is a discount rate applied to announcements rather than a verdict on the technology. A project list should be read with the delivery base rate attached, and the base rate should be applied to the announced tier specifically, because the producing, commissioning and under-construction tiers have already passed the test the base rate measures. Applied that way, the Gulf and Indian pipelines look smaller but firmer than the headline capacity suggests: a handful of assets that will exist, and a long list that mostly will not.
Three jurisdictions have published quantified support and two of the most-discussed have published none. What matters is not the headline envelope but the fraction actually disbursed, which in India is under 1.5 per cent.
Subsidy decides whether these molecules clear, and the position is that the published envelopes are large while the disbursements are small or absent. India has committed Rs 19,744 crore and had spent Rs
292 crore by August 2026, under 1.5 per cent, with no SIGHT incentive claimed or paid as at July 2026 because nothing supported under the scheme had been fully commissioned. The EU Hydrogen Bank has cleared three auctions and is publishing its own attrition. Japan has an envelope of about JPY 3 trillion over fifteen years. Saudi Arabia and the United Arab Emirates have no publicly quantified production support of any kind, and that absence is itself a finding a buyer should price.
| Jurisdiction | Instrument | Envelope as published | Per-unit support | What has actually been paid or signed |
|---|---|---|---|---|
| India | National Green Hydrogen Mission and SIGHT | Rs 19,744 crore total; SIGHT Rs 17,490 crore, split Rs 4,440 crore electrolyser manufacturing and Rs 13,050 crore production | Production incentive caps of Rs 50, 40 and 30 per kg in years one to three; Mode 2A ammonia incentives of Rs 8.82, 7.06 and 5.30 per kg | Rs 292 crore spent by August 2026, under 1.5 per cent; no SIGHT incentive claimed or paid as at July 2026 |
| European Union | Hydrogen Bank, three auctions | Second auction EUR 992m to 15 projects; third auction EUR 1.09bn to nine projects, about 1.1 GW; fourth round of EUR 500m opens December 2026 | Pilot cleared at EUR 0.48 per kg; second auction EUR 0.20 to 0.60 per kg general and EUR 0.45 to 1.88 per kg maritime; third auction EUR 0.44 to 3.49 per kg | Seven second-auction projects totalling 1.9 GW withdrew; only six signed grant agreements in January 2026, about 17 per cent of awarded capacity |
| Japan | Hydrogen Society Promotion Act, via JOGMEC | About JPY 3 trillion over 15 years | Per-unit support rates not published | JERA, 492,144 tpa, and Mitsui, 280,000 tpa, certified 19 December 2025; about USD 6.8bn, or about 35 per cent of the envelope, on Wood Mackenzie's estimate |
| Oman | Hydrom Round 3 terms, announced 13 to 14 August 2025 | No envelope in the official release | 90 per cent reduction in land lease fees during development, reduced early royalties, up to 10 years of corporate tax exemption; no per-kilogram value | No Round 3 award, shortlist or named bidder published as at 10 November 2026 |
| Jurisdiction | Instrument | Envelope as published | Per-unit support | What has actually been paid or signed |
|---|---|---|---|---|
| Saudi Arabia | None found | None published | None published | Exclusive green hydrogen export right granted to ACWA Power (Zawya, July 2026): an exclusivity, not a payment |
| United Arab Emirates | None found | None published | None published | None found |
India's mission is the largest quantified programme in either region, and its structure should be understood before its size is admired. It carries Rs 19,744 crore in total, of which SIGHT accounts for Rs 17,490 crore, split Rs 4,440 crore for electrolyser manufacturing and Rs 13,050 crore for production. Production incentive caps run at Rs 50, 40 and 30 per kg in years one to three, and the Mode 2A ammonia incentives are Rs 8.82, 7.06 and 5.30 per kg. Reliance bid Rs 24.99 per kg for 49,000 tpa in Mode 1 Tranche II (JMK Research, 21 March 2025), which is to say that competitive bidding has already pushed the realised incentive to half the first-year cap.
The disbursement position is the one that bears on a buyer. MNRE told Parliament that as at July 2026 no facility supported under SIGHT had been fully commissioned, so no incentive had been claimed or paid, and that Rs 292 crore of the Rs 19,744 crore outlay had been spent by August 2026 (Business Standard, 12 August and 3 November 2026). Under 1.5 per cent of a committed envelope has moved. The design is not in question: delivery is untested, and a project relying on the per-kilogram incentive relies on a payment no party has yet received.
SIGHT per-kilogram rates could not be obtained from a primary notice. The incentive caps and Mode 2A rates above are reported as given in secondary sources.
India's non-cash support may prove more valuable than its cash. The inter-state transmission charge waiver runs for 25 years for projects commissioned on or before 31 December 2030 (MNRE and PIB PRID 2329321), which is a long-dated, mechanical benefit requiring no claim process and no commissioning certificate from a scheme that has yet to certify one. By contrast, hub funding of Rs 4 billion for at least two hubs has not been deployed: the August 2024 tender was cancelled in July 2025. The instruments which pay automatically have performed; those requiring administration have not.
The EU Hydrogen Bank is the only mechanism in this survey that publishes a per-kilogram clearing premium, which makes it the only place where the true cost of bridging the gap can be observed. The pilot cleared at EUR 0.48 per kg. The second auction awarded fixed premiums of EUR 0.20 to 0.60 per kg for general projects and EUR 0.45 to 1.88 per kg for maritime, totalling EUR 992m across 15 projects. The third auction, with results on 7 May 2026, awarded EUR 1.09bn to nine projects at about 1.1 GW, with premiums of EUR 0.44 to 3.49 per kg, against EUR 8.4bn bid for a EUR 1.3bn budget. The clearing premium has risen by a factor of seven from pilot to third auction, and the oversubscription is six to one.
The Hydrogen Bank also publishes what nobody else does, which is its own attrition, and the figures are severe. Seven second-auction projects totalling 1.9 GW withdrew (S and P Global, 18 September 2025) and only six projects signed grant agreements in January 2026, representing about 17 per cent of awarded capacity (H2UB). A fourth round of EUR 500m opens in December 2026. On the second auction's evidence, an award under this scheme has roughly a one-in-six chance of becoming a signed grant agreement, which corrects any reading of auction results as capacity.
Japan's envelope is the largest nominal figure in the survey, at about JPY 3 trillion over fifteen years through JOGMEC, and its certifications are the most relevant to Indian exporters. JERA, at 492,144 tpa, and Mitsui, at 280,000 tpa, were certified on 19 December 2025, with Wood Mackenzie estimating about USD 6.8bn of support, or about 35 per cent of the envelope. Per-unit support rates are not published, so the value to a specific cargo cannot be calculated from the outside. ACME's 405,000 tpa agreement with IHI is supported by this mechanism, which is the
clearest instance in either region of a foreign subsidy underwriting an export contract.
The Gulf is where the record is thinnest and the claims are loudest. Oman's Hydrom Round 3, announced on 13 to 14 August 2025, offers a 90 per cent reduction in land lease fees during development, reduced early royalties and up to ten years of corporate tax exemption. No per-kilogram value and no envelope appears in the official release. A widely repeated USD 3.6bn incentive package is a secondary characterisation and should not be presented as official. For Saudi Arabia and the United Arab Emirates, no publicly quantified production subsidy, contract for difference or per-kilogram support could be found in any form. The Saudi grant of exclusive export rights to ACWA Power is an exclusivity, not a payment.
The takeaway is that a buyer should value subsidy by its disbursement record and its per-unit transparency, not by its headline envelope. On that test, the EU Hydrogen Bank is the most informative scheme because it publishes both its clearing premium and its attrition; Japan's is the most valuable to an Indian exporter but is opaque per unit; India's is large and entirely untested in payment; and the Gulf's cannot be valued at all, because nothing has been published to value. On the current record, no Gulf per-kilogram support figure exists in public.
The next four years contain a small number of dates that genuinely bind and a larger number that merely aspire. The most consequential is not a legal deadline at all: it is the IEA's early-2027 commercial cliff.
The calendar below separates what binds from what is aspired to, and the most consequential entry in it carries no legal force whatsoever. It is the IEA's finding that 22 Mt of announced 2030 capacity may lose any chance of beginning operation by 2030 if investment decisions are not taken by early 2027. Everything else on the list, including CBAM's definitive regime, the end of RFNBO additionality grandfathering on 1 January 2028 and the IMO adoption vote of 4 December 2026, operates on projects that take FID. The early-2027 date decides how many projects there are for the rest of the calendar to govern.
Entries marked target carry no obligation on any party. Entries marked binding, contractual are scheme or contract milestones whose breach has commercial rather than regulatory consequences. The Hydrom Round 3 award, shortlist and named bidders had not been published as at 10 November 2026 and no Round 4 has been announced.
| Date | Event | Character | Whom it reaches |
|---|---|---|---|
| 1 January 2026 | CBAM definitive regime begins, hydrogen in scope | Binding | EU importers, and therefore GCC and Indian ammonia, fertiliser and steel exporters commercially |
| 31 January 2026 | FuelEU Maritime first annual report, covering 2025, recurring | Binding | Operators of ships over 5,000 gross tonnes calling at EU ports, any flag |
| 31 January 2026 | Hydrom Round 3 final bids, one Duqm block up to 300 sq km | Binding on bidders | Omani licensing applicants |
| Mid-January 2026 | India Carbon Credit Trading Scheme tranche 2 notified, 208 units | Binding | Refining, petrochemicals and textiles |
| 7 May 2026 | EU Hydrogen Bank third auction results, nine projects, EUR 1.09bn | Award | Awardees |
| Q2 2026 | Hydrom Round 3 award, indicative | Target | No award, shortlist or named bidder published as at 10 November 2026, and no Round 4 |
| 9 to 28 August 2026 | India SIGHT electrolyser Tranche I scheduled commissioning, 1,500 MW across eight awardees | Binding, contractual | Passed with none confirmed commissioned |
| 30 November to 4 December 2026 | MEPC 85 and the resumed extraordinary session: the Net-Zero Framework adoption vote | Decision | Shipping, and ammonia and methanol suppliers to it |
| December 2026 | EU Hydrogen Bank fourth round opens, EUR 500m | Target | Bidders |
| Before end-2026 | Hintco to publish H2Global round two prices, including the AM Green per-tonne price | Disclosure | Every buyer benchmarking Indian RFNBO ammonia |
| 2026-27 | India CCTS second compliance year, about 60 per cent of the required reduction | Binding | Notified units |
| 1 January 2027 | CBAM certificate pricing moves to the weekly average ETS price | Binding | EU importers |
| 2 February 2027 | India SIGHT Mode 1 Tranche I, 412,000 tpa across ten awardees | Binding, contractual | Includes Reliance, ACME and Greenko at 90,000 tpa each |
| 26 March 2027 | India SIGHT electrolyser Tranche II, 1,500 MW across thirteen awards | Binding, contractual | Awardees |
| 2027 | First production at NEOM, TA'ZIZ Ruwais and ACME Duqm; first ACME Duqm cargo to Europe; RWE offtake from AM Green begins | Target | Buyers awaiting first molecules |
| Early 2027 | The IEA's FID deadline beyond which 22 Mt of announced 2030 capacity loses any chance of 2030 operation | Commercial, not legal | Every pre-FID developer in both regions |
| Date | Event | Character | Whom it reaches |
|---|---|---|---|
| 1 January 2028 | EU RFNBO additionality grandfathering ends | Binding | GCC and Indian exporters selling into EU compliance markets |
| 1 March 2028 | IMO Net-Zero Framework earliest entry into force if adopted in December 2026; 4 per cent base and 17 per cent direct intensity reductions in 2028 | Conditional | Shipping |
| 28 February and 1 March 2028 | India SIGHT Mode 1 Tranche II, 450,000 tpa | Binding, contractual | Awardees |
| 2028 | AM Green supply to Uniper begins, up to 500,000 tpa | Contractual | Uniper |
| July 2028 | Scheduled review of the EU RFNBO methodology | Binding review | Exporters holding RFNBO certification |
| 2029 | Hintco deliveries from AM Green phase 2 begin; Hyport Duqm phase 1 target | Contractual and target | Hintco; Omani licensees |
| 1 January 2030 | FuelEU zero-emission at berth in AFIR-covered EU ports | Binding | Ship operators and port authorities |
| 2030 | RED III 42 per cent industrial RFNBO and 1 per cent transport; hourly temporal correlation from 1 January 2030; CBAM factor about 51.5 per cent | Binding on member states | Exporters into the EU |
| 2030 | India NGHM 5 MMTPA; Oman 1 Mtpa; ACWA Yanbu commercial operation; ACME Odisha commissioning by July 2030 under the IHI contract for difference; India transmission waiver eligibility ends 31 December 2030 | Target | Both regions |
What the calendar shows is that 2026 and 2027 are dense with dates that have already been missed. The India SIGHT electrolyser Tranche I commissioning window of 9 to 28 August 2026, covering 1,500 MW across eight awardees, passed with none confirmed commissioned. The Hydrom Round 3 award, indicatively due in the second quarter of 2026, has produced no award, no shortlist and no named bidder as at 10 November 2026. QAFCO 7's reaffirmed Q2 2026 start-up is not publicly confirmed. Reliance's SIGHT commissioning date of 9 August 2026 passed without confirmation. A calendar whose recent entries have mostly been missed should be read as indicative for its future entries too.
The single most important date on the list is the IEA's early-2027 FID deadline, and it is important precisely because it is commercial rather than legal. The arithmetic behind it is straightforward: a project that has not taken FID by early 2027 cannot be engineered, financed, constructed and commissioned in time to operate in 2030, which means 22 Mt of announced production loses any chance of 2030 operation. Nothing happens on that date. No notification lapses and no penalty attaches. What happens is that, after it, the announced tier of the pipeline stops being a 2030 supply story and becomes at best a 2032 or 2033 one.
The consequences for both regions are direct. On the Gulf side, every awarded Omani block is pre-FID, with the energy minister expecting a first FID only in 2026 or 2027 (MEES, 7 March 2025). ACWA Power Yanbu, at about 2.5 Mtpa and targeting commercial operation in 2030, has FEED awarded and no FID. SABIC Agri-Nutrients Jubail has FID and EPC still to come. On the Indian side, NTPC Pudimadaka is at detailed-project-report tendering, Adani Mundra has no FID and MNRE has stated there is no visibility of projects in that region, and Reliance's gigafactory has published no 2026 update. Each of those sits on the wrong side of the early-2027 line as matters stand.
The cliff also interacts with the European rulebook in a way that compounds it. A project reaching FID in early 2027 will not be commissioned by 1 January 2028, and will therefore lose the additionality grandfathering available to electrolysers commissioned before that date. A project missing early 2027 for FID purposes is thus likely to miss the grandfathering date as well, and to face hourly temporal correlation from 1 January 2030 with no transitional benefit. Two deadlines that look independent are, for most of the announced pipeline, the same deadline.
Three nearer dates are worth watching for what they reveal rather than what they require. Hintco's publication of H2Global round two prices before the end of 2026 will give the market its second European clearing price for Indian-origin RFNBO ammonia, and the only independent test of whether the Indian cost base holds at European delivery. The IMO vote at the resumed extraordinary session on 4 December 2026 will determine whether the largest potential demand sink for ammonia as a fuel exists at all, with entry into force no earlier than 1 March 2028 if it passes. And the India SIGHT Mode 1 Tranche I date of 2 February 2027, covering 412,000 tpa across ten awardees including
Reliance, ACME and Greenko at 90,000 tpa each, will be the first real test of whether the Indian scheme can convert awards into operating plants.
The takeaway, and the closing observation of this survey, is that the calendar should be used as a filter rather than a forecast. A project that cannot name its FID date, its commissioning date relative to 1 January 2028, its certification status and its binding offtake counterparty is not a 2030 asset, whatever its announced capacity. By the middle of 2027 the published record will have answered those questions for most of the pipeline in both regions, and the answers will be visible without any forecasting at all. Until then, the status word, the signed offtake and the published price remain the only three things in this sector that can be relied upon, and in the Gulf the third of them does not yet exist. It is to be noted that the first two are available to any buyer who asks for them in writing, and that the answers are usually shorter than the decks.
Ten dated, falsifiable predictions for 2027: the heart of Volge Quarterly. Confidence is stated so a contrarian Medium Call and a High base case are read, and scored, differently. Each is written to be graded hit, miss or partial in the 2028 Scorecard.
Ten dated, falsifiable predictions for the year. The pattern is deliberate: the high-confidence Calls are the ones the physical world has already largely decided (NEOM’s completion, the Gulf’s blue-before-green sequence, India’s commissioning gap), while the genuine uncertainty sits in the corridors and the rulebook.
The reasoning behind each Call, with confidence and resolution date. The first three concern the Gulf, where the near-term outcome is largely set: NEOM ships, and blue precedes green.
NEOM’s Helios plant produces its first green ammonia during 2027. The target has already slipped from end-2026, and Air Products guides to mid-2027.
Confidence: High · Resolves H1 2027
Construction is complete and 100% of output is sold to Air Products for 30 years, so the risk is ramp and commissioning rather than demand or completion. ACWA Power told analysts on 6 August 2026 that commercial operations are set for 2027 [7][8][5].
Qatar’s Ammonia-7 (blue) starts up before any GCC green ammonia reaches scale : the Gulf’s near-term producing volume is blue, not green.
Confidence: Medium · Resolves 2026–27
Ammonia-7 is under construction, but its target moved to the second quarter of 2026 and no start-up has been announced, against a 37% fall in Qatari nitrogen exports in the first half of 2026. Green FIDs beyond NEOM remain thin; the region decarbonises existing ammonia faster than it builds green export [11].
The UAE’s Ta’ziz blue-ammonia plant at Ruwais reaches first production in 2027, on the Mitsui offtake into Japan and Asia.
Confidence: High · Resolves 2027
Construction began in the third quarter of 2024 with a named offtaker; the schedule is tight but the structure is complete, and $1.39bn of financing closed on 4 November 2026 with 2027 operations reaffirmed [2][21].
Calls four to six turn from the Gulf’s completions to its FID drought and to India’s two defining facts: the commissioning gap and the fertiliser demand anchor.
No new GCC green-hydrogen megaproject beyond NEOM reaches final investment decision with binding offtake by end-2027.
Confidence: Medium–High · Resolves end-2027
Oman’s blocks are auctioned ahead of demand, Masdar’s green target has slipped past 2030, and in August 2026 it cancelled a 100 MW electrolyser for Emsteel’s Musaffah complex and redirected 6 GW of solar from green ammonia to data centres, and the binding-offtake bar that took NEOM to FID has not been cleared again [1][3][9].
India commissions less than 5% of its 5-MTPA-by-2030 green-hydrogen target by end-2027.
Confidence: High · Resolves end-2027
About 0.16% was commissioned by February 2026 on 1.3% of the budget; even rapid acceleration leaves
the commissioned base far below 5% within the window [4].
India’s green-ammonia-to-fertiliser programme becomes the country’s first at-scale green-hydrogen demand, ahead of any export use.
Confidence: High · Resolves 2027
670,000 tonnes a year was signed with 11 of the 13 units to which 724,000 tonnes a year was allocated, on ten-year terms: contractual, domestic demand that begins delivering in 2027 [27].
Calls seven and eight concern the economics and the first European import: the premium that will not close, and the auction-backed cargo that will land.
The green premium does not close to below 1.5 times blue in any Gulf or India merchant project by end-2027.
Confidence: Medium–High · Resolves end-2027
Green runs 2 to 3 times blue, and CareEdge puts renewable hydrogen at 1.5 to 5 times unabated fossil; electrolyser capex rose about 57% since 2022; no unsubsidised merchant project has a path to 1.5× within the year [13].
The first H2Global / Fertiglobe green-ammonia cargo lands in Rotterdam in 2027 : Europe’s first auction-cleared import.
Confidence: Medium · Resolves 2027
19,500 tonnes is scheduled for 2027 at €1,000 a tonne under a government-backstopped contract, but the Haber-Bosch plant is only scheduled to be operating during 2027 and at full capacity in early 2028, and the Port of Rotterdam anticipates commercial activity in 2028 [29].
The final two Calls carry the year’s genuine uncertainty: which corridor delivers first, and whether Europe’s rulebook claims a casualty.
India secures a second certified green-ammonia export corridor, on binding terms, before any Gulf-to-Northeast-Asia green offtake delivers. The original Call, that ACME reaches Japan ahead of the Gulf, has been retired: its premise rested on about 405,000 tonnes a year from 2027, where the contract for
difference in fact covers 228,000 tonnes a year from September 2030.
Confidence: Medium · Resolves 2027–28
ACME’s Odisha certification stands but delivers at the end of the decade, so it cannot grade a 2027 race. Competing Gulf green routes still lack binding delivery dates, and the Gulf's export routing is now itself in question [30].
EU RFNBO compliance, rather than general viability, forces at least one announced Gulf or India EU-export memorandum to lapse or restructure by end-2027. Retreats on techno-commercial grounds do not count: INOX Air Products abandoned a $3bn Indian green ammonia project in October 2026 and Masdar cancelled an Emsteel electrolyser in August 2026, neither on certification grounds.
Confidence: Medium · Resolves end-2027
Certification is hard and merchant European demand is thin above the €1,000/t auction price; a non-compliant announced route is the likeliest casualty [32][5].
Four Calls are High, four Medium–High, two Medium. The High Calls are near-certainties the year merely confirms; the Medium Calls are where the edition takes real risk, and where the 2028 Scorecard will most test its judgement.
The Calls are reasoned judgements, not forecasts dressed as certainties. Each is written to a single standard: dated, falsifiable, sourced, and gradable by an outsider without access to the author.
A Call earns its place only if it can fail. “Green hydrogen will grow” is not a Call; “India commissions less than 5% of its 5-MTPA target by end-2027” is, because a single number in December 2027 settles it. Every Call names the year it resolves, the threshold that decides it, and the evidence it rests on. Where the evidence is a memorandum rather than a binding contract, the Call is written to account for the difference.
Confidence is stated explicitly (High, Medium–High, Medium, Low) so that a contrarian low-confidence Call and a high-confidence base case are read, and scored, differently. A High Call that hits earns little credit; a Medium Call that hits earns more, and a High Call that misses is the most damaging outcome of all.
The bands are a claim about the author’s certainty, not a hedge.
The 2028 edition opens with the Scorecard: each of these ten Calls marked hit, miss or partial, with a running hit-rate by confidence band. A forward-look that refuses to mark its own homework is a brochure; Volge Quarterly marks it. No Scorecard appears in this volume; it opens the accountability loop the next edition closes.
Dated, falsifiable, sourced, gradable. A Call that cannot be proven wrong by a specific date is not published.
The year by domain, three scenarios for how it resolves, the operator’s toolkit, and the reference annex (project data, numbered sources and a glossary) against which every figure in this edition can be checked.
The same year reads differently through each lens. Technology says the machine got more expensive; markets say demand contracted; regulation says the rulebook hardened; delivery says almost nothing has been built. All four are true at once.
The electrolyser is the story, and it moved the wrong way: median capex rose about 57% between 2022 and 2024, and a four-fold cost gap opened between Chinese alkaline stacks (~$600/kW) and Western PEM (~$2,500/kW) [GFJ, 2026]. Ammonia remains the carrier of choice, liquid at −33°C against −253°C for hydrogen. The technology works; its economics regressed.
The defining market fact is contraction: the announced 2030 pipeline fell 49 to 37 Mtpa and then to 27 Mt [IEA, October 2026], and new offtake has been flat near 1.7 Mt with only about a fifth under firm contract. The Gulf’s near-term volume is blue; India’s demand is domestic and policy-made. Merchant markets barely exist.
Europe’s RED III and, from January 2026, the carbon border levy make the RFNBO standard the export passport. Japan and Korea run contracts-for-difference; India has no economy-wide consumption mandate. Policy, not price, is the demand.
The hardest lens. NEOM is built and commissioning; Qatar and Ta’ziz will ship blue; India has commissioned 0.16% of its target. Delivery is where announcements meet the ground, and in 2027 the ground is mostly still empty.
Technology regressed, markets contracted, regulation hardened, delivery lagged. The year rewards the few projects that solved offtake before they began.
One year, three ways it can resolve. The base case is that 2027 confirms the physical facts already in train; the bull and bear cases turn on offtake and on whether a flagship slips.
Bull The molecule ships and demand hardens NEOM produces on schedule and ramps faster than feared; Qatar and Ta’ziz start up on time; the first H2Global cargo lands at Rotterdam and ACME’s ammonia reaches Japan; India’s fertiliser deliveries begin and a second Gulf green project reaches FID with binding offtake. The premium narrows at the margin. Probability: lower. It requires offtake to solve faster than the 1.7 Mtpa trend suggests. Base Completions land, the pipeline stays thin NEOM ships first green ammonia in H1 2027; blue precedes green across the Gulf; India commissions well under 5% of target but begins fertiliser deliveries; the first Rotterdam cargo lands; no new Gulf green megaproject reaches FID; the premium does not close. The year confirms the facts already set. Probability: highest. Bear A flagship slips and demand stalls NEOM’s first production slips beyond 2027; an announced Gulf or Indian flagship is descoped or shelved; H2Global or a corridor cargo is delayed; India’s commissioning gap widens as subsidy disbursement lags. The contraction in the pipeline accelerates. Probability: material, given a pipeline that already shrank once.
Scenarios are the author’s framing of the same sourced evidence, not separate forecasts. The base case carries the weight of the ten Calls; the bull and bear mark the range around them.
The signals to watch through 2027, each of which will shape or settle a Call and seed the 2028 edition. These are the events that, more than any announcement, will tell an operator where the sector is actually going.
NEOM Helios first-production date, and the shape of its ramp against nameplate through 2027. Qatar Ammonia-7 and Ta’ziz start-up confirmations, the blue-before-green sequence in fact. Any new Gulf green-hydrogen FID with binding, full-volume offtake beyond NEOM. Oman Round 3 Duqm awards and the SalalaH2 final investment decision, both due in 2026. India’s commissioned capacity and Mission disbursement rate, the gap between award and plant. First deliveries under the 670,000 t/yr fertiliser offtake signed in March 2026. The first H2Global / Fertiglobe cargo into Rotterdam, and its landed price. ACME/IHI first delivery to Japan under the JOGMEC contract-for-difference. Any announced Gulf or Indian EU-export memorandum that lapses or restructures on RFNBO grounds. Electrolyser capex direction: whether the 2022–24 rise reverses.
The 2028 edition opens with the Scorecard: each of this year’s ten Calls graded hit, miss or partial, with a running hit-rate by confidence band. A forward-look that refuses to mark its own homework is just a brochure; Volge Quarterly marks it.
Developer, IEA, Hydrom, SECI/MNRE, Mercom and Ammonia Energy disclosures (2023–26). Figures marked to verify in the body carry the same caveat here. Nameplate capacities at full build unless stated.
The projects referenced in this edition, with capacity, status and offtake on one page. Green and blue are distinguished; binding offtake is marked, and memoranda are not counted as anchors.
| Project | Type | Capacity | Offtake | Status / date |
|---|---|---|---|---|
| NEOM Helios (Saudi) | Green | >1.2 Mtpa NH₃; ~600 t/d H₂ | Air Products, 100%, 30 yrs (binding) | Built; first production mid-2027 |
| Qatar Ammonia-7 | Blue | 1.2 Mtpa NH₃ | QAFCO/QatarEnergy | Start-up 2026 |
| Ta’ziz (UAE) | Blue | 1.0 Mtpa NH₃ | Mitsui (Japan/Asia) | First prod. 2027 |
| Masdar green (UAE) | Green | 1 Mtpa target | European JVs | Delayed past 2030 |
| HyPort Duqm (Oman) | Green | 500 MW; ~60,000 t/yr | OQ/Uniper/DEME | Phase 1 ~2026 |
| SalalaH2 (Oman) | Green | 400 MW; ~$1bn | OQ/Linde | FID 2026 |
| ACME Duqm (Oman) | Green | 900,000 t/yr NH₃ | Not disclosed binding | Phased; demand-constrained |
| NTPC Pudimadaka (India) | Green | Hub; 70,000 t/yr NH₃ @ ₹51.80/kg | SECI/fertiliser | Pact 31 Mar 2026 |
| AM Green Kakinada | Green | 90,000 t/yr (SECI T-II award) | SECI production | Awarded |
| Reliance Jamnagar | Green | ~139,000 t/yr + 300 MW | SECI production | Under build |
| ACME (India) | Green | 90,000 t/yr + ~370,000 t/yr NH₃ | IFFCO Paradeep (record ₹49.75/kg); Japan via IHI | Awarded |
| Fertiglobe H2Global | Green | 19,500 t (2027) → 397,000 t cum. | H2Global auction, €1,000/t | First cargo 2027 |
| India fertiliser (SECI) | Green | 670,000 t/yr NH₃ with 11 units; 724,000 t/yr allocated across 13 | 10-yr, ₹49.75–64.74/kg (binding) | Signed 31 Mar 2026 |
[1] IEA, Global Hydrogen Review 2025 (9 Dec 2025) [3] Bloomberg, “Masdar Delays Green H2 Target Beyond 2030” (28 Oct 2024) [4] Business Today, “India’s green hydrogen reality check” (11 Jun 2026); MNRE [5] Ammonia Energy Assoc., NEOM financial close and 30-year offtake (2023) [6] NEOM / PRNewswire, $8.4bn financial close (2023) [7] Oil & Gas Middle East, NEOM ~90% built (2026) [8] Smart Water Magazine, NEOM first production 2026 [9] GH2.org, Oman national programme (2025) [10] gasworld / PRNewswire, Hydrom Round 3 (2025) [11] MEES, Qatar Ammonia-7 (29 Nov 2024) [12] MNRE, National Green Hydrogen Mission PDF (Jan 2023) [13] Green Fuel Journal, Cost economics (2026) [14] BloombergNEF, Electrolyser cost bands (9 Aug 2023) [15] PV Tech, Al Dhafra $0.0135/kWh (2020) [16] RenewEconomy, Saudi solar 1.04c/kWh (2021) [17] ACWA Power, NEOM equity (2023) [18] thyssenkrupp, >2 GW electrolysis (2023) [19] gasworld, ACWA export rights (2023) [20] Ammonia Energy, SalalaH2 FID 2026 [21] ADNOC / Mitsui / MEES, Ta’ziz blue ammonia (2024) [22] Energetica India, SECI production auction winners (3 Jan 2024) [23] Renewable Watch, SECI production Tranche II (20 Mar 2025) [24] JMK Research, SECI Tranche II analysis (2025) [25] Renewable Watch, Electrolyser-manufacturing auction, T-I (Jan 2024) [26] Renewable Watch, Electrolyser-manufacturing auction, T-II (Sep 2024) [27] Mercom, SECI 724,000 tpa green ammonia (31 Mar 2026) [28] Down To Earth, Green Steel Taxonomy (30 Dec 2024) [29] Fertiglobe / Port of Rotterdam / H2Global, Pilot auction (Jul 2024) [30] Ammonia Energy, ACME–IHI Japan certification (3 Jul 2026) [31] EU Hydrogen Observatory, RED III (in force Nov 2023) [32] European Commission, Renewable-hydrogen delegated acts (20 Jun 2023) [33] REPowerEU / European Commission, 20 Mt renewable H2 by 2030 (2022) [34] European Commission, CBAM definitive regime (from 1 Jan 2026) [35] CSIRO HyResource, Japan JOGMEC CfD (2025) [36] Ammonia Energy, Korea CHPS auction (16 Dec 2024)
RFNBO. Renewable Fuel of Non-Biological Origin: the EU RED III class of hydrogen/derivatives that counts toward renewable targets; the de-facto export passport.
All figures were refreshed to November 2026 and are attributed inline by source and year. Source numbering follows Volge’s research dossier. Volge Quarterly is governed by Volge’s Morpheus standard; this edition is not investment advice.
RED III. EU Renewable Energy Directive (in force Nov 2023): 42% of industrial hydrogen RFNBO by 2030, 60% by 2035.
CBAM. Carbon Border Adjustment Mechanism: the EU carbon levy on imports; definitive regime from 1 Jan 2026, reaching hydrogen and ammonia.
Grey / blue / green. Hydrogen from unabated gas (grey), gas with carbon capture (blue), or electrolysis on renewable power (green).
FID. Final Investment Decision: the point at which a project is committed and financed.
Take-or-pay / MoU. A binding contract obliging a buyer to pay for volume, versus a non-binding memorandum of understanding.
SIGHT. Strategic Interventions for Green Hydrogen Transition: India’s incentive scheme under the Mission (via SECI).
NGHM. India’s National Green Hydrogen Mission (approved Jan 2023): ₹19,744 cr, 5 MTPA by 2030.
CfD. Contract-for-Difference: a government-backstopped price support, used by Japan (JOGMEC) and Korea (CHPS).
H2Global. EU-backed import auction mechanism; cleared green ammonia at €1,000/t into Rotterdam.
Mtpa / tpd. Million tonnes per annum; tonnes per day.
Alkaline / PEM. The two main electrolyser technologies: cheaper alkaline (often Chinese) and Western PEM.
Volge Quarterly is Volge's signature series, published every quarter for enterprise customers. Each edition takes one subject, sets out where it currently stands on the published record, and makes a set of dated, falsifiable judgements on where it is going, written to give a buyer something firmer than a forecast: every figure carries its source and its date, every judgement carries the date on which it resolves, and the following year's edition grades each one hit, missed or partial.
Where a figure is not published, the edition says so rather than estimating it. Where published sources disagree, both are given and the disagreement is stated rather than averaged away. A number that does not survive checking is corrected here rather than repeated.
The edition carries no recommendation of any supplier and no offer of service. Where a vendor, platform or deployment is named it is because it is part of the published record of the subject.
This edition may be shared freely. It may not be altered or republished without permission, and figures carry their stated basis and date. Names and marks appearing in it are the property of their respective owners.