Per-fuel well-to-wake intensity: ammonia grades

Background: ammonia as a marine fuel

Ammonia is a colourless, pungent, alkaline gas with the chemical formula NH3 and a molecular weight of 17.03 g/mol. At ambient pressure it liquefies at minus 33.34 degrees Celsius, and at ambient temperature it liquefies at approximately 8.6 bar gauge. The two storage modes (cryogenic at minus 33 degrees and atmospheric pressure, or pressurised at ambient temperature and roughly 9 bar) frame the bunker-tank design choice for an ammonia-fuelled vessel. The lower heating value of approximately 18.6 MJ/kg combined with the liquid density of approximately 682 kg/m3 at minus 33 degrees yields a volumetric energy density of approximately 12.7 GJ/m3, which is roughly 36 percent of VLSFO and roughly 80 percent of methanol on the same basis. The bunker-tank size penalty for an ammonia-fuelled ship is therefore approximately 2.7 to 3.0 times the equivalent VLSFO tank for the same range, slightly worse than methanol but materially better than liquid hydrogen.

The case for ammonia as a marine fuel rests on four structural advantages. First, the molecule contains no carbon, so the TtW CO2 is zero by chemistry, and the only GHG terms at the engine are N2O slip, NOx (which converts indirectly to N2O in the atmosphere) and a small unburned-ammonia term. Second, the global ammonia industry already produces approximately 185 million tonnes per year for fertiliser, refrigeration, plastics and explosives applications, with deep-water trading on more than 200 ammonia gas carriers and a network of import terminals at major chemical and fertiliser ports. Third, ammonia is one of the few practical hydrogen carriers at marine scale, so the renewable-hydrogen value chain that builds for stationary energy applications can supply marine bunkers through the same logistics. Fourth, the synthesis route is well-understood Haber-Bosch chemistry that scales linearly with hydrogen feedstock, which means a low-carbon hydrogen supply translates directly into a low-carbon ammonia supply.

The case against rests on three considerations. The toxicity profile is severe, with acute inhalation effects at concentrations above 25 ppm and an immediately-dangerous-to-life-and-health threshold of 300 ppm; the safety case for ammonia bunkering and onboard handling is therefore substantially heavier than the methanol or LNG envelope. The N2O slip risk is structural to ammonia combustion, and a 0.5 percent slip can dominate the TtW intensity at GWP100 of 273. The WtT intensity question is sharp: a vessel that bunkers grey ammonia from a Saudi or Trinidadian Haber-Bosch plant operates at a higher WtW intensity than the same vessel burning VLSFO, so the fuel choice only delivers a GHG benefit when the ammonia carries a low-carbon production certificate.

The shipboard architecture is unaffected by which grade is bunkered. The pumps, the cryogenic or pressurised piping, the inert-gas blanketing, the leak-detection layout, the engine fuel injectors, the pilot-fuel skid, the SCR or N2O abatement reactor, and the IGF Code Part B safety case are identical across the three grades, because the molecule is identical. The certificate of analysis, the proof of sustainability, the chain-of-custody document, the FuelEU verifier statement, and the MEPC.391(82) WtT value are different. The compliance value of a tonne of ammonia bunker is therefore set by the upstream certificate, not the engine.

Haber-Bosch synthesis: hydrogen sourcing determines the grade

Ammonia is synthesised at industrial scale through the Haber-Bosch process, which combines hydrogen and atmospheric nitrogen at 400 to 500 degrees Celsius and 150 to 300 bar over an iron catalyst (or a more modern ruthenium catalyst in some plants). The reaction is N2 + 3H2 to 2NH3, exothermic by approximately 92 kJ per mol of ammonia formed, and runs as a recycle loop with single-pass conversion of approximately 15 to 20 percent and total conversion approaching 97 percent over the recycle. The nitrogen feedstock is air-separated through a cryogenic ASU, contributes a small electrical energy demand of approximately 0.1 GJ per tonne of ammonia, and is essentially carbon-free in its supply.

The hydrogen feedstock dominates both the cost and the carbon footprint. Each tonne of ammonia consumes approximately 178 kg of hydrogen, which represents roughly 88 percent of the molecular mass and approximately 82 percent of the energy input to synthesis. The grade label (grey, blue, green) is determined entirely by how this hydrogen is produced. Grey ammonia uses hydrogen from steam-methane reforming of natural gas, blue ammonia uses the same SMR or autothermal reforming pathway with carbon capture and storage applied to the synthesis-gas chain, and green ammonia uses hydrogen from electrolysis of water using renewable electricity. The Haber-Bosch reactor itself is identical across the three grades.

The synthesis-loop electricity demand is approximately 0.6 to 0.8 MWh per tonne of ammonia, supplying the recycle compressors, the refrigeration to liquefy the product, and ancillary loads. This electrical demand is a meaningful term in the green-ammonia carbon footprint because a renewable-electrolyser plant that runs the synthesis loop on grid electricity (rather than on the same renewable supply as the electrolyser) carries the grid carbon intensity into the WtT calculation. The RED III RFNBO methodology requires the synthesis-loop electricity to meet the same additionality and correlation criteria as the electrolyser electricity if the producer wants to claim full RFNBO status across the integrated plant.

The conversion efficiency of hydrogen to ammonia in the Haber-Bosch loop is approximately 80 to 85 percent on an energy basis, with the balance lost as heat (recovered in the steam system and the synthesis-gas preheat), as purge gas (vented or burned for process heat), and as compression work. The energy ladder for an integrated electrolyser-plus-Haber-Bosch plant runs at approximately 35 to 42 MWh of renewable electricity per tonne of green ammonia delivered to the bunker manifold, depending on electrolyser efficiency, synthesis-loop integration, and product-storage chain.

Grey ammonia: SMR-based pathway

Grey ammonia is the conventional production pathway and accounts for approximately 72 percent of global ammonia output. The hydrogen feedstock is produced through steam-methane reforming of natural gas at 800 to 1,000 degrees Celsius over a nickel catalyst, with a primary reformer fed by natural gas and steam, a secondary reformer fed by air to introduce the nitrogen for the downstream Haber-Bosch synthesis, water-gas shift reactors to convert CO to CO2 and additional hydrogen, CO2 removal by amine absorption, and methanation to clean residual CO. The cleaned synthesis gas (a 3:1 H2:N2 ratio) feeds the Haber-Bosch loop.

The carbon balance is dominated by the natural gas feedstock and the reformer flue gas. Roughly 0.45 to 0.55 tonnes of natural gas (as methane) are consumed per tonne of ammonia produced for the hydrogen feedstock, plus another 0.20 to 0.30 tonnes for process heat in the primary reformer. The CO2 vented from the amine-absorber unit is approximately 1.6 tonnes per tonne of ammonia (high-purity CO2, captured as part of the synthesis-gas cleanup), and the CO2 vented from the reformer flue gas is approximately 0.5 tonnes per tonne of ammonia (lower-purity CO2 mixed with combustion gases). Total direct CO2 emissions from a typical grey ammonia plant therefore run at approximately 2.0 to 2.4 tonnes per tonne of ammonia, plus an upstream methane-leakage term that depends on the gas supply chain.

The well-to-tank intensity therefore lands in the 110 to 130 gCO2eq/MJ range for the upstream stage. With ammonia’s TtW CO2 at zero (no carbon in the molecule), the WtW intensity equals the WtT plus the engine-side N2O contribution. For a modern ammonia dual-fuel engine with N2O slip below 0.1 percent of fuel input, the engine-side contribution is approximately 1 to 3 gCO2eq/MJ on a GWP100 basis, which leaves the WtW intensity for grey ammonia at 111 to 133 gCO2eq/MJ under MEPC.391(82) Annex 1.

The MEPC.391(82) Annex 1 default for grey ammonia from natural gas in non-CCS configuration sits at approximately 121 gCO2eq/MJ WtW for the global average, with regional defaults varying based on the natural gas supply chain (US Gulf, Trinidad, Middle East, China-coal-based) and the reformer thermal efficiency. The China coal-based ammonia pathway, which uses gasification of coal to produce the hydrogen feedstock, sits substantially higher at approximately 180 to 220 gCO2eq/MJ WtW because of the higher carbon intensity of coal versus natural gas. The FuelEU Annex II default for the natural-gas grey ammonia pathway sits at approximately 120 gCO2eq/MJ WtW, with the two figures aligned within methodological tolerance.

The compliance implication is direct. A vessel that bunkers grey ammonia is not reducing GHG intensity relative to VLSFO. Under FuelEU 2026 with a baseline target of approximately 88 gCO2eq/MJ falling to roughly 78 gCO2eq/MJ in 2030, grey ammonia fails the standard outright and accumulates a substantial compliance deficit on every tonne consumed. The shipowner pays a penalty per the FuelEU schedule, and the ammonia fuel switch produces a worse compliance outcome than continuing on conventional residual fuel. The role of grey ammonia in the marine market is therefore even more transitional than grey methanol, and the offtake agreements signed in the 2024 to 2026 window for ammonia-fuelled newbuilds are uniformly tied to certified blue or green volumes rather than grey.

Blue ammonia: SMR/ATR + CCS pathway

Blue ammonia uses the same Haber-Bosch synthesis loop as grey ammonia, but the upstream hydrogen production is configured to capture and store the CO2 streams. Two reformer architectures are deployed. The first is conventional steam-methane reforming with carbon capture applied to the synthesis-gas amine-absorber stream and optionally to the reformer flue gas. The second is autothermal reforming (ATR), which combines partial oxidation with steam reforming in a single reactor and produces a pressurised, high-purity CO2 stream at the synthesis-gas cleanup that is far more amenable to capture than the SMR flue gas. ATR is the preferred architecture for greenfield blue-ammonia plants because the capture rate can reach 95 percent at lower marginal cost.

The capture and storage chain typically targets the high-purity CO2 stream from the amine-absorber on the synthesis-gas chain, which can be captured at the lowest cost and represents approximately 70 percent of the total plant CO2 emissions. The flue-gas CO2 from the primary reformer combustion is lower purity and higher cost, and is captured only when the project economics justify the additional kit. Capture rates of 60 to 95 percent of the production-stage CO2 are technically feasible, with 85 to 90 percent representing the typical greenfield ATR design point and 70 to 80 percent representing a typical retrofit on an existing SMR plant.

The captured CO2 is compressed to pipeline pressure (approximately 100 to 150 bar), transported by pipeline or ship to a geological storage site, and injected into a depleted gas field, saline aquifer or enhanced-oil-recovery formation. The Northern Lights and Northern Endurance projects in the North Sea, the 45Q-credited storage sites in the US Gulf, and the Aramco Jafurah developments in Saudi Arabia represent the leading 2026 blue-ammonia storage chains.

The WtT intensity reduction depends on the capture rate, the energy penalty of the capture and compression chain, the storage-site monitoring requirements, and the upstream natural gas methane leakage. A blue ammonia plant with 90 percent capture and a low-leakage gas supply (less than 0.5 percent upstream methane loss) reaches a WtT intensity of approximately 15 to 35 gCO2eq/MJ. With ammonia’s TtW CO2 at zero and a small N2O term of 1 to 3 gCO2eq/MJ, the WtW intensity for high-quality blue ammonia lands in the 30 to 60 gCO2eq/MJ band. Lower-capture-rate or higher-leakage configurations push the intensity above 70 gCO2eq/MJ, at which point the GHG benefit relative to grey ammonia erodes meaningfully.

Several blue ammonia projects entered front-end engineering design or final investment decision in the 2024 to 2026 window. CF Industries’ Donaldsonville Blue Point project in Louisiana, with capacity of approximately 1.4 million tonnes per year, is the leading North American blue-ammonia development. Yara International’s Sluiskil and Pilbara projects, JERA’s Aichi project in Japan, and Aramco’s Jubail and Jafurah Saudi developments represent the global pipeline. By mid-2026, certified blue ammonia production capacity reached approximately 2.5 million tonnes per year, against a global ammonia market of 185 million tonnes.

The MEPC.391(82) Annex 1 default for blue ammonia depends on the certified capture rate and is typically issued through a third-party verifier under ISCC, REDcert or CertifHy frameworks, with the certificate naming the capture rate and the storage chain. The compliance value under FuelEU is intermediate. Better than grey ammonia, worse than green ammonia, and a meaningful but not transformational improvement relative to VLSFO. A blue ammonia cargo at 45 gCO2eq/MJ WtW delivers an approximately 51 percent intensity reduction relative to VLSFO, which substantially exceeds the FuelEU 2030 trajectory. The shipowner running blue ammonia therefore has compliance headroom that can be banked or pooled, subject to the certified-volume availability.

The other axis of evaluation is upstream methane leakage. A blue ammonia plant that captures 90 percent of production-stage CO2 but draws on a natural gas supply with 2.5 percent upstream methane leakage delivers a worse WtW intensity than a 70 percent-capture plant on a 0.3 percent leakage supply, because the methane-leakage term applied at GWP100 of 28 outweighs the marginal capture benefit. The MEPC.391(82) Annex 1 framework requires the upstream leakage value to be either the certified project-specific figure or a regional default, and the blue ammonia marketing claim therefore has to be verified at the certificate level rather than asserted at the project level.

Green ammonia: electrolysis + Haber-Bosch (RFNBO)

Green ammonia is the renewable-fuel grade produced from green hydrogen (manufactured by electrolysis of water using renewable electricity) and atmospheric nitrogen (from a cryogenic air separation unit). The two feedstocks are combined in a Haber-Bosch synthesis loop similar to the conventional pathway, with the difference that the hydrogen atom originates outside the fossil chain. The product ammonia is molecularly indistinguishable from grey ammonia.

The electrolyser is the dominant capital and energy term. Three principal electrolyser technologies are commercially available in 2026: alkaline electrolysis (mature, low capital cost, electrical efficiency of approximately 51 to 60 percent on an LHV-of-hydrogen basis), proton-exchange-membrane (PEM) electrolysis (faster ramp-rate, more capital-intensive, electrical efficiency of approximately 56 to 65 percent), and solid-oxide electrolysis (still pre-commercial at marine scale, highest theoretical efficiency at 70 to 80 percent with high-temperature heat integration). A typical green-ammonia plant in 2026 uses alkaline or PEM electrolysers feeding a buffer hydrogen storage that smooths the renewable-supply variability, then a continuously operating Haber-Bosch loop that draws hydrogen at design rate.

The WtT intensity is driven by three terms: the renewable-electricity intensity supplied to the electrolyser, the synthesis-loop electricity intensity, and the embedded emissions of the plant infrastructure (electrolyser stacks, Haber-Bosch reactor, ASU, refrigeration). A well-designed green-ammonia plant with renewable electricity at less than 18 gCO2eq/kWh and a modern alkaline or PEM electrolyser achieves a WtT intensity of approximately 3 to 12 gCO2eq/MJ. With ammonia’s TtW CO2 at zero and a small N2O term of 1 to 3 gCO2eq/MJ, the WtW intensity for high-quality green ammonia lands in the 3 to 15 gCO2eq/MJ band.

The grid-intensity sensitivity is severe. A green-ammonia plant on a renewable-grid mix at 50 gCO2eq/kWh (still well below most national grid averages but typical for a hybrid renewable-plus-grid configuration) carries a WtT intensity of approximately 25 to 35 gCO2eq/MJ, which erodes the green-ammonia advantage substantially and may fail the RED III 70 percent GHG-saving threshold. A plant on a coal-heavy grid at 800 gCO2eq/kWh (typical for a Chinese or Indian grid average) carries a WtT intensity of approximately 350 to 450 gCO2eq/MJ, which is materially worse than grey ammonia. The geographical and additionality rules in the RED III RFNBO methodology are designed to prevent this outcome by requiring the renewable electricity to be additional, temporally correlated and geographically correlated.

The RED III delegated act on RFNBOs sets the conditions for the renewable-electricity input. The electricity must be additional (sourced from a renewable plant commissioned no earlier than 36 months before the electrolyser), temporally correlated (matched on an hourly basis from 1 January 2030, monthly until then), and geographically correlated (in the same bidding zone or a neighbouring zone with no congestion). The delegated act also mandates a minimum 70 percent GHG saving versus a fossil comparator of 94 gCO2eq/MJ, which means certified green ammonia cannot exceed approximately 28.2 gCO2eq/MJ WtW. The MEPC.391(82) Annex 1 framework adopts a similar threshold structure for RFNBO eligibility, although the temporal and additionality criteria are less prescriptive than RED III.

The MEPC.391(82) Annex 1 default for green ammonia depends on the certified electricity intensity. Renewable-grid green ammonia at less than 18 gCO2eq/kWh carries a default WtW intensity of approximately 8 gCO2eq/MJ, and hybrid configurations with grid backup carry defaults of approximately 15 to 25 gCO2eq/MJ. The FuelEU Annex II values track these figures with minor methodological differences. Green ammonia qualifies for the FuelEU RFNBO 2x multiplier, which doubles the GHG-intensity benefit of an RFNBO bunker for the period 2025 to 2033 inclusive, and is the strongest compliance lever in the marine fuel mix outside negative-WtT bio-LNG.

Green ammonia production reached approximately 250,000 tonnes per year of certified output by mid-2026, against the global ammonia market of 185 million tonnes. The leading operating projects include the Yara Heroya green-ammonia plant in Norway (commissioned 2024), the Iberdrola Puertollano plant in Spain, the CF Industries Donaldsonville green-ammonia retrofit, and the NEOM Green Hydrogen Project in Saudi Arabia (under construction, with first ammonia production targeted for 2027). The price spread to grey ammonia stands at roughly 3 to 5 times on a tonne basis in 2026, reflecting the renewable-electricity cost, the small production base, and the policy-driven demand from FuelEU and the IMO net-zero framework.

MEPC.391(82) Annex 1 defaults per grade

The IMO 2023 Guidelines on Lifecycle GHG Intensity of Marine Fuels, adopted as resolution MEPC.391(82) at the 82nd session of the Marine Environment Protection Committee, codify the per-fuel default emission factors that feed the marine GHG fuel standard under the IMO net-zero framework. Annex 1 of the Guidelines provides default WtT, TtW and WtW values for each fuel and pathway, with the ammonia grades treated as multiple distinct lines covering the SMR, ATR, partial-oxidation-of-coal and electrolysis pathways.

The tabular summary for ammonia pathways in Annex 1, reduced to indicative values, is as follows:

The TtW CO2 is reported as zero for all ammonia grades because the molecule contains no carbon, which is a structural feature of ammonia versus methanol or LNG. The TtW N2O term covers engine-side nitrous oxide slip, which is set as a default of 1 to 3 gCO2eq/MJ for type-tested ammonia dual-fuel engines with N2O abatement (SCR-based or upstream catalyst). Engines without abatement, or operating outside the type-test envelope, can incur N2O slip terms of 10 to 30 gCO2eq/MJ, which would dominate the WtW total for any green-ammonia bunker.

The Annex 1 defaults apply when the bunker supplier does not provide a project-specific certified value. A supplier with a third-party certified WtT figure under ISCC, REDcert, CertifHy, or an equivalent scheme can substitute the certified value for the default, and the verifier under the FuelEU and IMO compliance frameworks accepts the certified value subject to mass-balance and chain-of-custody verification.

The ammonia-specific Annex 1 entries also include unburned-ammonia slip terms. Ammonia dual-fuel engines exhibit slip in the range of 0.1 to 1.5 percent of fuel input depending on engine architecture, load and combustion strategy, with the latest MAN ME-LGIA two-stroke designs achieving slip below 0.5 percent in the type-test envelope. Unburned ammonia is not a direct GHG (zero GWP), but it is a precursor to N2O in the atmosphere through nitrogen-cycle chemistry, and it is a NOx and PM2.5 precursor through reaction with NOx and SOx in ambient air. The MEPC.391(82) framework treats unburned ammonia as a marginal contribution to the WtW total of approximately 0.1 to 0.5 gCO2eq/MJ, with the larger contribution captured through the N2O slip term.

FuelEU Annex II treatment

The FuelEU Maritime regulation, Regulation (EU) 2023/1805, takes effect for ships above 5,000 GT calling at EU and EEA ports from 1 January 2025. The regulation imposes a declining limit on the fleet-average WtW GHG intensity of energy used on board, starting at approximately 89.34 gCO2eq/MJ in 2025 (a 2 percent reduction relative to the 2020 baseline of 91.16 gCO2eq/MJ) and tightening to 77.94 gCO2eq/MJ in 2030, 62.9 gCO2eq/MJ in 2035, 39.0 gCO2eq/MJ in 2040, 26.6 gCO2eq/MJ in 2045, and 18.2 gCO2eq/MJ in 2050.

Annex II of the regulation provides default WtW emission factors for each fuel and production pathway. The ammonia entries in Annex II mirror the MEPC.391(82) Annex 1 structure with minor numerical differences:

The Annex II defaults are adjusted by the company through a verifier-approved certificate where a project-specific WtT value is documented. The certificate must be issued by a recognised voluntary scheme (ISCC EU, REDcert, CertifHy, 2BSvs, or equivalent), must trace the feedstock through the supply chain on a mass-balance basis, and must conform to RED III sustainability criteria for green ammonia under the RFNBO framework.

The compliance arithmetic combines Annex II defaults across the bunker mix, applies the RFNBO multiplier where eligible, and produces the fleet-attained GHG intensity that is compared to the year’s limit. A surplus is banked, pooled or transferred. A deficit triggers the FuelEU penalty per /wiki/fueleu-penalties-pooling-multipliers, set at 2,400 EUR per tonne of VLSFO-equivalent energy required to close the gap, increased by 10 percent per consecutive year of non-compliance. The penalty is structurally larger than the spot premium of green ammonia once the multiplier is applied, which explains the policy-driven demand for low-carbon ammonia grades from 2025 onwards.

The pilot-fuel stream contributes to the FuelEU calculation alongside the main ammonia fuel. A typical 5 to 10 percent MGO pilot at the conventional MGO Annex II default of 91.2 gCO2eq/MJ adds approximately 4.5 to 9 gCO2eq/MJ to the engine-side blended intensity, which is meaningful for a fuel that targets sub-15 gCO2eq/MJ overall on the green-ammonia branch. The pilot-fuel substitution to bio-MGO or HVO pilot offers a 4 to 5 gCO2eq/MJ improvement on the blend for a 5 percent pilot fraction, and is an active development area for the 2027 to 2030 horizon.

RED III sustainability criteria for green ammonia

The Renewable Energy Directive Recast III (Directive (EU) 2023/2413, RED III), adopted in October 2023, sets the sustainability and GHG-saving criteria that determine whether a green-ammonia cargo qualifies for FuelEU Annex II treatment as a renewable fuel. Ammonia is treated as a derived RFNBO under the RED III framework, with the certification chain running from the electricity supplier through the electrolyser operator, the Haber-Bosch synthesis operator, the bunker supplier and the verifier.

The green-ammonia (RFNBO) criteria, set by the RED III delegated act on renewable hydrogen (Commission Delegated Regulation (EU) 2023/1184), require:

• 70 percent GHG saving versus the fossil comparator of 94 gCO2eq/MJ. This caps certified green ammonia WtW intensity at approximately 28.2 gCO2eq/MJ.
• Additionality of the renewable electricity supply, sourced from a renewable plant that came into operation no earlier than 36 months before the electrolyser. A grandfathering window applies until 2028 for plants in operation before 2028. The objective is to ensure that the renewable electricity attributed to RFNBO production is genuinely incremental rather than displaced from another use.
• Temporal correlation, matching electrolyser hourly consumption to renewable-plant hourly production from 1 January 2030. A monthly correlation rule applies in the transition period from 2024 to 2029, which loosens the constraint and allows producers to commission projects under interim rules.
• Geographical correlation, sourcing the electricity from the same bidding zone or a neighbouring zone without grid congestion. The bidding-zone framework is administered by ENTSO-E for the European synchronous grids and by equivalent operators in other jurisdictions, and the cross-border correlation rule is administered through the verifier.
• Synthesis-loop electricity attribution, requiring that the electricity supplied to the Haber-Bosch synthesis loop and the ASU also satisfy the additionality and correlation criteria if the producer wants to claim full RFNBO status across the integrated plant.

The Commission Delegated Regulation (EU) 2023/1185 sets the GHG calculation methodology for RFNBOs, with default emission factors for the renewable electricity source (zero for additional, correlated renewable supply), the electrolyser efficiency (project-specific, verified by certificate), the synthesis-loop electricity (matched to the same RFNBO criteria), and the supply-chain transport of the ammonia from the production site to the bunker terminal (small contribution, typically 0.5 to 1.5 gCO2eq/MJ). The verifier sums these terms to produce the certified WtT intensity that supersedes the Annex II default.

The certification chain runs from the electricity supplier through the electrolyser operator, the Haber-Bosch operator, the storage operator, the bunker supplier and the verifier. Each step carries a Proof of Sustainability (PoS) or Proof of Compliance (PoC) document under one of the recognised voluntary schemes (ISCC EU, REDcert EU, CertifHy, 2BSvs, or equivalent). The mass-balance accounting allows the renewable molecules to be commingled with conventional ammonia in storage and transport infrastructure provided the certified volumes are tracked by document rather than by physical segregation.

The RED III criteria are the most demanding among the major RFNBO frameworks, and a green-ammonia plant that satisfies the RED III criteria typically also satisfies the equivalent thresholds in the IMO MEPC.391(82) framework, the Japanese green ammonia certification scheme, the Korean K-ETS clean-hydrogen framework and the US 45V hydrogen production tax credit. Cross-recognition agreements between the schemes are still developing in 2026, with bilateral arrangements between the EU and Japan and between the EU and the UK in active negotiation.

RFNBO 2x multiplier and additionality/correlation rules

Green ammonia that satisfies the RED III RFNBO criteria qualifies for the FuelEU Maritime 2x multiplier under Article 5(7) of Regulation (EU) 2023/1805. The multiplier doubles the contribution of the RFNBO bunker to the GHG-intensity calculation for the period from 1 January 2025 through 31 December 2033 inclusive. The mechanism is designed to accelerate RFNBO uptake during the supply-ramp phase by amplifying the compliance value of an early-mover bunker, and is applied to the energy content of the RFNBO fuel rather than to its emission factor.

The arithmetic operates as follows. A vessel that bunkers 1,000 GJ of green ammonia at 8 gCO2eq/MJ WtW under the multiplier is treated, for FuelEU intensity purposes, as if it had consumed 2,000 GJ at 8 gCO2eq/MJ. The numerator (total emissions) doubles, but so does the denominator (total energy), so the intensity contribution of the RFNBO line stays at 8 gCO2eq/MJ. The multiplier effect appears in the average intensity calculation when the RFNBO is bunkered alongside conventional fuels: the doubled RFNBO energy enlarges the share of the low-intensity stream in the fleet-average denominator and pulls the attained intensity downwards more aggressively than a face-value calculation would suggest.

For a vessel bunkering 90 percent VLSFO at 92 gCO2eq/MJ and 10 percent green ammonia at 8 gCO2eq/MJ on an energy basis, the face-value attained intensity is:

With the 2x multiplier on the green-ammonia line, the green-ammonia energy is treated as 20 percent of the total (10 percent plus an additional 10 percent virtual), which expands the denominator. The effective attained intensity for compliance becomes approximately:

The 7 gCO2eq/MJ improvement from the multiplier is the lever that justifies the green-ammonia premium for an early-mover shipowner. The mechanism is detailed at /wiki/fueleu-rfnbo-multiplier and worked through arithmetically at /calculators/fueleu-rfnbo-multiplier.

The additionality rule prevents a green-ammonia producer from buying renewable electricity from an existing wind or solar plant that was already supplying the grid. The rule requires the renewable plant to be commissioned no earlier than 36 months before the electrolyser, which in practice means new build wind, solar or geothermal capacity dedicated (or attributed) to the RFNBO project. The temporal correlation rule prevents a producer from running the electrolyser at night on coal-fired grid power and claiming the daytime solar generation as the RFNBO source through annual averaging. The geographical correlation rule prevents a producer from claiming wind generation in Norway as the source for an electrolyser in Spain, even if both are in the European synchronous grid, unless the producer can demonstrate uncongested transmission capacity through the relevant cross-border interconnectors.

These three rules operate jointly and are administered through the verifier under the recognised voluntary schemes. The compliance burden is non-trivial, and the certification cost adds approximately 2 to 5 EUR per MWh of renewable electricity to the green-ammonia delivered cost. The cost is small relative to the FuelEU compliance value (which can exceed 200 EUR per tonne of ammonia equivalent on the multiplier-adjusted basis), but it is a real frictional cost that must be carried in the project economics.

Engine landscape: MAN B&W ME-LGIA, Wartsila 25 ammonia, WinGD X-DF-A

The ammonia-fuelled fleet in 2026 runs on three principal engine families, all of which accept the same liquid ammonia regardless of grade and all of which require a small pilot-fuel injection to initiate compression ignition.

MAN Energy Solutions ME-LGIA is the dominant two-stroke ammonia dual-fuel engine. The ME-LGIA is a Diesel-cycle engine derived from the ME-C two-stroke base platform with a high-pressure liquid-ammonia injection system replacing the heavy fuel-oil injectors. Liquid ammonia is supplied to the engine at approximately 50 bar from the bunker tank, then boosted to approximately 700 bar at the injection event, atomises into the cylinder, and is ignited by a co-injected pilot fuel (typically 5 to 10 percent of total fuel energy on a marine gas oil basis). The Diesel cycle preserves combustion phasing similar to fuel-oil operation, and the ammonia slip is typically below 0.5 percent of fuel input on the latest type-test data. The platform was first ordered in 2022 with first deliveries to ammonia-fuelled bulkers and gas carriers from 2024 onwards. The engine size range covers 6S60ME-LGIA (small to medium-bore) through 7S95ME-LGIA (very-large-bore for VLCC and ULCS class), with the platform extension to all major ME-C base sizes underway.

Wartsila 25 ammonia is the four-stroke ammonia dual-fuel engine for medium-speed propulsion and auxiliary applications. The Wartsila 25 ammonia is based on the Wartsila 25 four-stroke platform with an ammonia fuel-supply skid, low-pressure ammonia injection, and a pilot-fuel injection system. The four-stroke architecture suits ferries, ro-ro vessels, supply vessels, ammonia carriers (running on cargo boil-off), and auxiliary generators. Ammonia slip is typically in the 0.5 to 1.5 percent range, slightly higher than the two-stroke ME-LGIA because of the four-stroke combustion-chamber geometry and the lower injection pressure. The platform reached commercial release in 2024 with the first installations on ammonia carriers and offshore support vessels, and Wartsila has announced a 32-bore variant (Wartsila 32 ammonia) for the next development phase.

WinGD X-DF-A is the two-stroke ammonia dual-fuel engine from Winterthur Gas and Diesel, a co-development with Mitsui E&S and CSSC. The X-DF-A is based on the X-DF base platform (two-stroke low-speed) with an ammonia fuel supply system and pilot injection. The engine adopts a Diesel-cycle ammonia injection at high pressure similar to the MAN ME-LGIA architecture, with combustion characteristics tuned for low N2O slip and acceptable ammonia slip. The X-DF-A reached commercial release in 2025 with the first vessel orders for VLGC and ammonia carrier applications. The platform competes head-to-head with the MAN ME-LGIA on the very-large-bore segment.

Other entrants include the Hyundai Heavy Industries HiMSEN ammonia four-stroke, the J-ENG UE-LSGiA two-stroke ammonia engine, and a number of medium-speed marine engines from Japanese builders targeting the ammonia-carrier and bulker segments. The dual-fuel landscape is detailed at /wiki/ammonia-marine-engines-overview.

The unified characteristic across all engine families is that ammonia is a compression-ignition fuel with a pilot requirement rather than a spark-ignition fuel like LNG in an Otto-cycle dual-fuel engine. Ammonia’s auto-ignition temperature of approximately 651 degrees Celsius is too high to ignite reliably from compression alone in a marine-bore cylinder, and the laminar flame speed of approximately 0.07 m/s is one of the lowest of any practical fuel, which makes flame propagation slow and combustion phasing sensitive to the pilot-fuel injection. The pilot fuel is conventionally marine gas oil at 5 to 10 percent of fuel energy, and the substitution to bio-MGO or HVO renewable diesel is the next decarbonisation step beyond the main fuel switch.

Pilot fuel requirements

The compression-ignition combustion of ammonia in a marine dual-fuel engine requires a pilot fuel because of three properties of ammonia. First, the auto-ignition temperature of approximately 651 degrees Celsius is high relative to a marine-bore compression-ignition cylinder, which typically reaches 500 to 650 degrees Celsius at top dead centre and would not reliably ignite ammonia from compression alone. Second, the laminar flame speed of approximately 0.07 m/s is approximately one-fifth of methane and one-tenth of methanol, which means flame propagation across the combustion chamber is slow and the heat-release profile must be initiated by a faster-burning ignition source. Third, the minimum ignition energy of ammonia is high relative to hydrocarbon fuels, and the spark-ignition route is impractical at marine cylinder volumes.

The pilot fuel solves these problems by introducing a small quantity of a high-cetane fuel (typically marine gas oil DMA grade) at top dead centre, which auto-ignites on compression and produces a hot ignition kernel that initiates ammonia combustion. The pilot fraction ranges from approximately 5 percent of fuel energy on the latest MAN ME-LGIA designs at high load to approximately 10 percent on lower-load operation or smaller-bore engines, with 7 percent representing a typical operating average. The pilot is injected through a separate pilot-fuel injector pump and supply line, and the pilot quantity is controlled by the engine management system to maintain combustion stability across the operating envelope.

The pilot contribution to the total WtW intensity is calculated by energy-weighted averaging of the ammonia stream and the pilot stream. For a typical operating point with 93 percent ammonia energy and 7 percent MGO pilot energy, where the ammonia is green ammonia at 8 gCO2eq/MJ and the MGO pilot is conventional at 91.2 gCO2eq/MJ, the blended WtW intensity is approximately:

The pilot adds approximately 6 gCO2eq/MJ to the green-ammonia baseline, which is meaningful for a fuel that targets sub-15 gCO2eq/MJ overall. The same arithmetic applied to grey ammonia at 121 gCO2eq/MJ with the same 7 percent MGO pilot yields approximately 119 gCO2eq/MJ for the mix, where the pilot contribution is marginal because both streams are fossil and similar in intensity.

The pilot fuel supply is subject to the same certification requirements as the main fuel. A vessel claiming a green-ammonia RFNBO compliance benefit must document the pilot-fuel WtW intensity against the FuelEU Annex II default for the pilot stream, which is the conventional MGO default unless the vessel bunkers certified bio-MGO or HVO renewable diesel as the pilot. The substitution of conventional MGO pilot with HVO or bio-MGO pilot offers a 4 to 5 gCO2eq/MJ improvement on the blended intensity for a typical pilot fraction, and is an active development area for the 2027 to 2030 horizon.

N2O slip and the dominance in TtW intensity (AR5 GWP100 = 273)

Nitrous oxide (N2O) is a non-CO2 greenhouse gas with an AR5 100-year global warming potential of 273 (AR6 updated this to 273 with minor methodological refinements; AR4 used 298). N2O is formed in ammonia combustion through several reaction pathways: direct oxidation of unburned ammonia in the lean post-flame region, NH-radical reactions with NO at intermediate temperatures, and partial reduction of NO in the cooling boundary layers. The resulting N2O concentration in the engine exhaust depends on combustion temperature, residence time, equivalence ratio, ammonia slip and the presence of NOx-control catalysts.

The structural concern for ammonia as a marine fuel is that the high GWP100 of N2O means even a small slip fraction can dominate the TtW intensity. A simple arithmetic example illustrates the scale. Ammonia at 18.6 MJ/kg LCV and a TtW emission factor for a 0.5 percent N2O slip (mass basis, expressed as N2O per kg of ammonia consumed) yields:

A 0.5 percent mass slip of ammonia converted to N2O therefore adds 73 gCO2eq/MJ to the TtW intensity, which by itself approaches the WtW intensity of grey ammonia and dominates any green-ammonia bunker claim. The MEPC.391(82) and FuelEU frameworks therefore require the engine-side N2O emission factor to be either the type-test certified value or a default that reflects modern abatement-equipped engines.

The actual N2O slip from a modern ammonia dual-fuel engine is far lower than the 0.5 percent illustrative example, but the engineering design point is set by exactly this concern. Type-test data from the MAN ME-LGIA, the Wartsila 25 ammonia, and the WinGD X-DF-A indicate engine-out N2O slip in the range of 0.05 to 0.20 percent of fuel input on a mass basis, which translates to a TtW N2O contribution of approximately 7 to 30 gCO2eq/MJ before abatement. With a downstream selective catalytic reduction (SCR) reactor optimised for N2O destruction (typically a vanadium-titanium catalyst at 280 to 350 degrees Celsius), the post-abatement N2O slip is reduced by 60 to 90 percent, leaving a residual contribution of approximately 1 to 3 gCO2eq/MJ that is the value reported in the MEPC.391(82) Annex 1 default.

The combustion-strategy choices that minimise N2O slip include high-temperature combustion (which thermally decomposes N2O above approximately 950 degrees Celsius), late-cycle injection that reduces the lean post-flame residence time, and combustion-chamber geometries that minimise wall-quench effects. The MAN ME-LGIA two-stroke architecture, with its long combustion-chamber length and high in-cylinder temperature peak, exhibits favourable N2O behaviour relative to the four-stroke architectures. The Wartsila 25 ammonia and similar four-stroke designs typically need a heavier post-treatment catalyst stack to reach the same final N2O level.

The emissions monitoring requirement is non-trivial. NOx and N2O concentrations in the exhaust must be measured during the type-test (at certified test cells) and verified periodically in service through portable analyser sampling or onboard continuous emission monitoring. The MEPC.391(82) framework references the IMO NOx Technical Code for the type-test methodology and adds a specific N2O measurement protocol for ammonia-fuelled engines. The FuelEU framework relies on the MEPC.391(82) value as the verified engine factor.

The unburned-ammonia slip is a separate concern. Ammonia itself has a GWP of zero on a 100-year basis (it is short-lived in the atmosphere, oxidising to N2 within hours), so unburned ammonia does not directly contribute to the WtW intensity. It does contribute to local air-quality impacts (PM2.5 formation through reaction with sulfate and nitrate aerosols), and a small fraction is converted to N2O in the atmosphere, which is captured indirectly in the lifecycle accounting through a separate term. The MEPC.391(82) Annex 1 treats unburned ammonia at approximately 0.1 to 0.5 gCO2eq/MJ for type-tested engines, with the larger contribution captured through the engine-side N2O slip term.

Toxicity, safety, IGF Code Part B / MSC.1/Circ.1687

Ammonia is a toxic, low-flashpoint (when vaporised in air), corrosive and pungent gas. Acute inhalation exposure at 25 ppm causes respiratory irritation, at 100 ppm causes severe eye and respiratory irritation, at 300 ppm represents the immediately-dangerous-to-life-and-health (IDLH) threshold, at 1,500 to 2,000 ppm causes pulmonary edema and respiratory failure, and at 5,000 ppm or above is rapidly fatal. The toxicity profile is materially worse than methanol or LNG and is closer in operational impact to chlorine or hydrogen sulfide on a per-tonne-released basis. The crew protection envelope, the ventilation design, the leak-detection sensitivity, and the emergency-response procedures must all be sized to this toxicity.

The IMO regulatory framework for ammonia-fuelled ships is provided by the IGF Code (International Code of Safety for Ships using Gases or other Low-flashpoint Fuels) Part B framework, supplemented by the Interim Guidelines for the safety of ships using ammonia as fuel issued as MSC.1/Circ.1687 in May 2024 at the 108th session of the Maritime Safety Committee. The Interim Guidelines apply pending the formal incorporation of ammonia-specific provisions into the IGF Code, which is on the IMO Maritime Safety Committee work programme with target adoption in 2027 to 2028.

The principal safety provisions in MSC.1/Circ.1687 are:

• Toxicity-driven hazardous-area classification that treats ammonia release as a toxic-gas-cloud hazard rather than purely a flammability hazard. The release-rate calculations, the exclusion-zone sizing, and the muster-point arrangements are sized to a worst-credible-release scenario at IDLH concentration thresholds.
• Double-walled fuel piping with leak detection and inert-gas blanketing in the annular space, identical in concept to the LNG IGF Code provisions but with tighter leak-detection thresholds (typically 25 ppm trip).
• Ammonia-rated materials for the fuel-supply chain. Carbon steel and most stainless steels are compatible with dry liquid ammonia, but copper, brass and zinc are aggressively attacked. The piping, valves, gaskets and seals must be specified for ammonia service.
• Gas-detection arrays with low-ppm sensitivity (typically 25 ppm warning, 100 ppm shutdown, 300 ppm muster) at the bunker manifold, the fuel-room, the engine room and the accommodation HVAC intake. Detector technology is electrochemical, infrared or photoacoustic depending on the location and response-time requirement.
• Crew personal protective equipment including dedicated ammonia-rated gloves, eye protection, splash suits, and self-contained breathing apparatus (SCBA) for response to release events. Standard fuel-oil PPE is inadequate for ammonia exposure.
• Emergency-release-system (ERS) couplings at the bunker manifold to allow rapid vessel-to-shore disconnect under release conditions, with vapour return and inert-gas purging built into the disconnect sequence.
• Tank vent treatment with thermal oxidation, scrubbing or remote venting to a safe location. Direct atmospheric venting of ammonia from the bunker tank is not permitted.
• Crew training at a level higher than the IGF Code Part A baseline, with a dedicated ammonia-fuel competence module covering toxicity, exposure response, and emergency procedures.

The bunker delivery operation is materially heavier than the methanol or LNG equivalent. The exclusion zone around the bunker manifold during transfer typically extends 50 to 100 metres versus 20 to 30 metres for methanol, and the ship-to-ship bunker barge must be equipped with a vapour-return line, inert-gas blanketing, and a closed-loop emergency-shutdown system. The Singapore Maritime and Port Authority and the Rotterdam Port Authority have issued port-specific procedural guidance for ammonia bunkering that supplements MSC.1/Circ.1687, and the IACS classification societies (DNV, Lloyd’s Register, ABS, ClassNK, BV, RINA) have issued class-specific rules for ammonia-fuelled ships and bunker vessels.

The crew-safety case is the principal regulatory hurdle for early commercial ammonia bunkering. The first commercial ammonia STS bunker took place in Singapore in 2024 under a closed-port-area regime, and routine commercial operations at Rotterdam, Houston and Yokohama are expected to ramp from 2026 to 2028 as port authorities issue full operational permits.

Bunkering supply chain and storage

The ammonia bunkering supply chain in 2026 covers a small number of established hubs with reliable supply and a growing network of pilot and demonstration sites. The supply chain depends on the grade.

Grey ammonia bunkering operates at ammonia-import terminals worldwide. CF Industries (US Gulf, capacity of approximately 8 million tonnes per year), Yara International (Norway, Trinidad and Australia, capacity of approximately 8.5 million tonnes per year), Trammo (global trader and shipowner, with 14 ammonia gas carriers), and Mitsui (Japanese ammonia trader and project developer) supply the marine market through dedicated bunker barges or through STS transfer from ammonia gas carriers. The bunker price for grey ammonia in Singapore, Rotterdam and Houston in 2026 ranged from approximately 600 to 800 USD per tonne on an ammonia basis, equivalent to approximately 1,200 to 1,600 USD per tonne on a VLSFO-energy-equivalent basis.

Blue ammonia bunkering is concentrated at specific supply nodes. CF Industries supplies blue ammonia at Donaldsonville from 2026 under the Blue Point project, with offtake contracts to Maersk, NYK and other early-mover ammonia-fuelled vessel operators. Yara supplies blue ammonia at Sluiskil (Netherlands) and at Pilbara (Australia). Aramco and Mitsui developed a blue-ammonia chain from Saudi Arabia to Japan through long-term offtake agreements. The certified blue ammonia volumes available to the marine bunker market reached approximately 800,000 tonnes in 2026, against a marine ammonia demand of approximately 600,000 tonnes (driven by the first ammonia-fuelled bulkers, gas carriers and pilot vessels).

Green ammonia bunkering is currently limited to a handful of sites. Yara delivers green ammonia from Heroya (Norway) to Northern European bunker barges. CF Industries delivers green ammonia from Donaldsonville (US Gulf). Iberdrola supplies green ammonia from Puertollano (Spain) to South European bunker operations. The NEOM Green Hydrogen Project in Saudi Arabia (under construction) is targeted to deliver green ammonia from 2027 to a global ammonia-shipping network. The certified green ammonia volumes available to the marine bunker market reached approximately 250,000 tonnes in 2026.

Storage is split between cryogenic and pressurised modes. Cryogenic storage at minus 33 degrees Celsius and atmospheric pressure is the standard for large terminal and shipboard tanks above approximately 1,000 m3, because the liquid density at minus 33 degrees (682 kg/m3) is higher than the pressurised liquid density at ambient temperature (618 kg/m3 at 9 bar), and the tank wall thickness for a low-pressure cryogenic vessel is lower than for a high-pressure ambient vessel of equivalent volume. Pressurised storage at ambient temperature and 9 bar is used for smaller tanks (below approximately 200 m3) and for road and rail tankers, where the simpler mechanical envelope outweighs the density penalty. Semi-pressurised storage at intermediate conditions (minus 5 degrees and approximately 5 bar) is used for medium-scale terminals.

The shipboard ammonia bunker tank on a marine application is typically a fully pressurised cylindrical or bilobe vessel in stainless steel or low-temperature carbon steel, with a vacuum-insulated or perlite-insulated outer jacket, a boil-off compressor to maintain pressure within design limits, and a connected fuel-supply system that delivers liquid ammonia to the engine at the required pressure. The bunker operation transfers liquid ammonia at the corresponding storage condition (cryogenic or pressurised) through a vapour-return-equipped manifold to the ship tank, with inert-gas blanketing of any vapour space and continuous gas detection at the manifold.

The supply chain bottleneck through 2030 is the certified low-carbon ammonia volume. Total certified blue and green ammonia production reached approximately 1.05 million tonnes in 2026 against a marine ammonia demand of approximately 600,000 tonnes, which means low-carbon supply is currently in surplus. The supply demand balance shifts as the ammonia-fuelled fleet expands (over 200 vessels on order or in operation by 2026, with delivery profile through 2030), and the medium-term constraint is the green-ammonia volume rather than the blue-ammonia volume.

Regulatory hurdles for first commercial ammonia bunkering

The first commercial ammonia bunkering operations at major ports require an integrated regulatory clearance covering the IMO framework, the flag-state administration, the port-state authority, the classification society, the local environmental regulator and the local emergency-response services. The pathway has been navigated successfully at Singapore (first STS ammonia bunker 2024), and is in active development at Rotterdam, Houston, Yokohama, Pilbara, and Brisbane.

Singapore issued the technical reference for ammonia STS bunkering through the MPA in 2023, conducted the first commercial STS ammonia bunker at the closed Raffles Reserved Anchorage in 2024 under a closed-port-area regime, and is targeting full operational ammonia bunkering at the open anchorages from 2027 onwards. The MPA framework covers the bunker barge approval, the receiving vessel approval, the operator competence, the exclusion-zone management, the emergency-response chain, and the periodic regulatory review.

Rotterdam issued the Port of Rotterdam Authority’s ammonia bunkering procedure in 2024, conducted the first ammonia bunker trials at the Maasvlakte 2 terminal in 2025, and is targeting commercial bunker availability from 2026 to 2027. The Rotterdam framework integrates the Dutch flag-state regulations, the Port Reception Facilities directive, the Seveso III chemical-establishment thresholds, and the local emergency-response coordination through the Rotterdam Port Authority.

Houston, Pilbara, Yokohama, and Brisbane are at varying stages of regulatory approval, with first commercial ammonia bunker operations targeted for 2026 to 2028 depending on the local approval pathway. The US Coast Guard issued the Navigation and Vessel Inspection Circular for ammonia-fuelled vessels in 2025, the Australian Maritime Safety Authority issued the Marine Order for ammonia-fuelled vessels in the same window, and the Japanese Ministry of Land, Infrastructure, Transport and Tourism issued bunker guidelines for the major Japanese ports.

The cross-cutting hurdles are: emergency-response capability for a worst-credible toxic release at a port that may have residential populations within 1 to 5 kilometres, insurance and liability allocation for incidents at the bunker manifold, training and certification of bunker barge crews and shore terminal operators, and the legal framework for cross-border STS transfers in international waters versus port-state waters. The regulatory pathway is comparable in complexity to the early LNG bunkering rollout from 2010 to 2018, and the timeline expectation is similar (10 to 12 years from first commercial bunker to broad availability across major ports).

Formula, assumptions, and limits

Formula

The well-to-wake intensity of an ammonia grade in MEPC.391(82) and FuelEU Annex II accounting takes the form:

where g indexes the production grade (grey, blue, green), EF_WtT,g is the well-to-tank intensity for that grade as certified or as the Annex 1 / Annex II default, the TtW CO2 is zero by chemistry (no carbon in NH3), the N2O slip term captures the engine-out N2O converted to gCO2eq through GWP100, and the slip and NOx terms cover unburned ammonia and indirect NOx-to-N2O atmospheric chemistry.

The TtW N2O contribution is calculated as:

where GWP100 of N2O is 273 (AR5 / AR6), m_N2O slip is the mass of N2O released per unit mass of ammonia consumed, m_NH3 is the unit mass of ammonia consumed, and LCV_NH3 is the lower heating value of ammonia (18.6 MJ/kg).

For a blend of grey and certified-green ammonia, the blended WtW intensity is:

where x_green is the energy fraction of certified green ammonia in the bunker mix.

The pilot-fuel adjustment for the engine-side energy mix is:

where p is the pilot energy fraction (typically 0.05 to 0.10).

Derivation

The WtT term aggregates upstream emissions from feedstock extraction or sourcing, hydrogen production (steam-methane reformer, autothermal reformer, or electrolyser), Haber-Bosch synthesis, product refrigeration and storage, transport to the bunker terminal, and bunker delivery. Each step is quantified in gCO2eq per MJ of ammonia delivered to the ship’s manifold, and the values are summed.

The TtW CO2 term for ammonia is set to zero by chemistry. Ammonia (NH3, molecular weight 17.03) contains zero carbon atoms, so combustion produces N2 and H2O as the principal products with no CO2. The carbon-free combustion is the structural feature that distinguishes ammonia from methanol and LNG on the TtW side, and it is the reason ammonia is treated as a zero-carbon fuel at the engine even when the WtT intensity is high.

The TtW N2O term is derived from the mass balance of nitrogen in the combustion chemistry. Ammonia oxidation produces a mixture of N2 (preferred), NO (intermediate, requires SCR for compliance with the IMO NOx Technical Code), and trace N2O (a kinetic byproduct of incomplete oxidation). The N2O yield depends on the combustion temperature, residence time, equivalence ratio and post-treatment configuration. Type-test data on modern marine ammonia engines indicates engine-out N2O at 0.05 to 0.20 percent of fuel input on a mass basis, reduced by 60 to 90 percent through SCR catalyst optimisation, leaving a post-abatement N2O contribution of approximately 1 to 3 gCO2eq/MJ on a GWP100 basis with N2O GWP of 273.

The slip term covers the small fraction of ammonia that passes through the engine unburned. Unburned NH3 has a GWP100 of zero (it is short-lived in the atmosphere), but it contributes to local air-quality impacts through PM2.5 formation, and a small fraction is converted to N2O atmospherically. The MEPC.391(82) framework treats this contribution at approximately 0.1 to 0.5 gCO2eq/MJ for type-tested engines.

Assumptions

The framework assumes:

• The bunker certificate accurately documents the production pathway and the WtT intensity, with full chain-of-custody traceability under a recognised voluntary scheme.
• The mass-balance accounting under ISCC EU, REDcert, CertifHy or equivalent is implemented with verified physical chain of custody through commingled storage.
• The engine-side N2O slip conforms to the type-test certified value, which is verified in service through periodic emission-monitoring requirements.
• The SCR catalyst stack is operational and within its design temperature window. A vessel running with the SCR bypassed or degraded is outside the type-test envelope and incurs higher N2O emissions that are not captured in the default WtW value.
• The pilot-fuel stream is documented separately and contributes its own WtW intensity to the energy-weighted average.
• The renewable-electricity supply for green ammonia satisfies the RED III additionality, temporal and geographical correlation criteria.

Worked example

A vessel bunkers 1,000 tonnes of certified green ammonia at a WtW intensity of 8 gCO2eq/MJ (including 2 gCO2eq/MJ from post-abatement N2O slip), with a 7 percent MGO pilot at the conventional MGO Annex II default of 91.2 gCO2eq/MJ. The blended engine-side WtW intensity is:

The vessel’s compliance value under FuelEU 2026 with a target of approximately 88.0 gCO2eq/MJ is 88.0 minus 13.82 = 74.18 gCO2eq/MJ of surplus per MJ of fuel consumed. With the RFNBO 2x multiplier applied to the green-ammonia fraction, the effective surplus rises further for the ammonia portion of the bunker. The arithmetic is detailed at /calculators/fueleu-rfnbo-multiplier.

A grey-ammonia counterfactual at 121 gCO2eq/MJ with the same 7 percent MGO pilot yields a blended intensity of approximately 119 gCO2eq/MJ, which is in deficit by approximately 31 gCO2eq/MJ versus the 2026 target. The deficit triggers the FuelEU penalty mechanism per /wiki/fueleu-penalties-pooling-multipliers, at a substantially larger penalty per tonne than the equivalent grey-methanol case because of the higher gap.

A blue-ammonia case at 45 gCO2eq/MJ with the same pilot yields approximately 48 gCO2eq/MJ blended, which is in surplus by approximately 40 gCO2eq/MJ versus the 2026 target. The blue-ammonia case demonstrates that even without the RFNBO multiplier, a high-quality CCS-based ammonia stream delivers strong FuelEU compliance value.

Edge cases and limits

• Coal-gasification grey ammonia from China carries a WtW intensity of approximately 200 gCO2eq/MJ, which is more than double the VLSFO baseline. A vessel that bunkers coal-derived ammonia under FuelEU is in catastrophic deficit and accumulates penalties at the maximum rate. The pathway is unviable under any current EU compliance framework, and is excluded from RFNBO and certified-blue offtake markets.
• N2O abatement failure: a vessel with a bypassed or degraded SCR catalyst incurs engine-out N2O slip at 0.10 to 0.20 percent, which adds 15 to 30 gCO2eq/MJ to the TtW intensity. The default Annex 1 value assumes operational abatement, and a verifier audit can require the company to apply a higher N2O factor if abatement is not maintained.
• Methane leakage in blue-ammonia gas supply: a blue-ammonia plant on a 2.5 percent upstream methane leakage gas supply delivers a worse WtW intensity than a 0.5 percent leakage supply by approximately 15 to 20 gCO2eq/MJ, which can erode the entire CCS benefit. The certificate must document the gas-supply leakage value.
• Synthesis-loop electricity attribution for green ammonia: a green-ammonia plant that runs the Haber-Bosch synthesis loop on grid electricity (rather than on the same renewable supply as the electrolyser) incurs the grid carbon intensity in the WtT calculation. The RED III RFNBO methodology requires the synthesis-loop electricity to satisfy the same additionality and correlation criteria.
• Pilot fuel stream: a vessel that bunkers green ammonia but uses conventional MGO as pilot retains the fossil pilot intensity on the energy-weighted average. The transition to bio-MGO or HVO pilot fuel is the next decarbonisation step beyond the main fuel switch.
• Mass-balance versus book-and-claim: blue and green ammonia certification under RED III follows mass-balance, which requires physical chain of custody through commingled storage and transport. Book-and-claim allocation is not currently accepted under FuelEU Annex II.
• N2O measurement uncertainty: portable N2O analysers have a measurement uncertainty of approximately 10 to 20 percent at the low concentrations typical of post-abatement engine exhaust. The verifier under FuelEU and IMO compliance accepts the type-test value subject to periodic in-service verification, with allowance for measurement uncertainty in the verification protocol.

Regulatory basis

• IMO MEPC.391(82), 2023 Guidelines on Lifecycle GHG Intensity of Marine Fuels (LCA Guidelines), Annex 1 default emission factors per fuel and pathway.
• Regulation (EU) 2023/1805 (FuelEU Maritime), Annex II default WtW emission factors and Article 5(7) RFNBO multiplier.
• Directive (EU) 2023/2413 (RED III), sustainability and GHG-saving criteria for RFNBOs.
• Commission Delegated Regulation (EU) 2023/1184 (RFNBO additionality, temporal and geographical correlation rules).
• Commission Delegated Regulation (EU) 2023/1185 (RFNBO GHG calculation methodology).
• IMO MSC.1/Circ.1687, Interim Guidelines for the safety of ships using ammonia as fuel.
• IMO IGF Code Part B (low-flashpoint and toxic fuels framework).
• IMO NOx Technical Code (engine type-test methodology including N2O measurement protocol).
• IPCC AR5 / AR6 GWP100 values, with N2O at 273.

Common errors

• Treating ammonia as a zero-emission fuel because it contains no carbon. The TtW CO2 is zero, but the engine-side N2O slip and the WtT intensity can both be substantial, and the WtW intensity of grey ammonia exceeds VLSFO.
• Applying AR4 GWP of 298 for N2O instead of AR5 / AR6 GWP of 273. The MEPC.391(82) and FuelEU frameworks use the AR5 / AR6 value.
• Omitting the pilot-fuel contribution from the WtW calculation. The 5 to 10 percent MGO pilot adds 4.5 to 9 gCO2eq/MJ to the engine-side blended intensity.
• Applying the RFNBO 2x multiplier to blue ammonia. Blue ammonia does not qualify as an RFNBO because the hydrogen feedstock is fossil-derived; only green ammonia with full RED III RFNBO certification qualifies.
• Assuming book-and-claim allocation is accepted under FuelEU Annex II. The current rule requires mass-balance with physical chain of custody.
• Using the GWP20 methane factor of 84 instead of GWP100 of 28 for upstream methane leakage in the WtT calculation. Both MEPC.391(82) and FuelEU Annex II use GWP100.
• Ignoring the synthesis-loop electricity in the green-ammonia WtT calculation. The Haber-Bosch loop draws 0.6 to 0.8 MWh per tonne of ammonia and is a meaningful term if it is supplied from a non-RFNBO grid source.
• Conflating grey ammonia from natural gas SMR (110 to 130 gCO2eq/MJ WtW) with grey ammonia from coal gasification (175 to 215 gCO2eq/MJ WtW). The two pathways are both labelled grey but have very different lifecycle footprints.

See also

• /wiki/ammonia-as-marine-fuel
• /wiki/ammonia-marine-engines-overview
• /wiki/fueleu-intensity-formula-breakdown
• /wiki/fueleu-rfnbo-multiplier
• /wiki/marine-gfs-methodology
• /wiki/imo-net-zero-framework
• /wiki/per-fuel-wtw-methanol-grades
• /wiki/per-fuel-wtw-bio-lng
• /wiki/per-fuel-wtw-vlsfo-mgo
• /calculators/fuel-wtw-ammonia
• /calculators/fuel-wtw-blend
• /calculators/fuel-wtw-methanol
• /calculators/fueleu-rfnbo-multiplier
• /calculators/fueleu-ghg-intensity
• /calculators/gfi-attained

Related calculators

• e-Diesel / FT e-Fuel - Well-to-Wake
• VLSFO - Well-to-Wake
• LPG - Well-to-Wake
• LNG - Well-to-Wake by engine pathway
• HFO - Well-to-Wake
• Ammonia Fuel - NOx + N₂O Slip
• MGO / MDO - Well-to-Wake
• Hydrogen - Well-to-Wake by pathway

Related formulas

• fuel wtw ammonia
• Ammonia Fuel - NOx + N₂O Slip
• fuel wtw vlsfo
• fuel wtw nuclear
• fuel wtw mgo
• fuel wtw methanol
• fuel wtw lpg
• fuel wtw lng