A ship picks its fuel the way an aircraft picks its wing: every other design choice follows from it. Heavy fuel oil ruled the sea for a century because it packs about 40 MJ into every kilogram and roughly 36 MJ into every liter, costs less than the steel around it, and sits in an unpressurized tank at room temperature. Every alternative on offer loses on at least one of those three axes, and the question for an owner ordering a ship today is which loss is survivable for the trade the ship will run. This article is the hub for the alternative and zero-carbon fuels cluster: it sets the fuels side by side on the five tests that decide a fuel choice, then routes down to the deep-dive articles that carry each fuel’s detail. The whole comparison sits under decarbonization and alternative fuels and feeds the well-to-wake accounting in well-to-wake fuel pathways and the carbon price in the IMO Net-Zero Framework and GFI.
The five tests recur for every fuel, so it helps to name them once. First, energy density: how much energy the fuel carries per kilogram and, the figure that actually sizes the ship, per liter. Second, the carbon footprint on two boundaries, tank-to-wake (what comes out of the funnel) and well-to-wake (the full chain from extraction or synthesis to combustion). Third, toxicity and fire safety, which decide the containment and the crew protection and route through the IGF Code or the IGC Code. Fourth, bunkering readiness: whether a ship can refuel where it trades. Fifth, cost and availability, which decide whether any of the rest matters in a charterer’s spreadsheet. Hold those five and the ladder below reads as one comparison, not eight separate stories.
Energy density: the constraint that sizes the ship
Energy density is where the alternatives separate first, because a ship is a volume-limited box. Fuel oil sets the benchmark at about 40 MJ/kg lower heating value and roughly 36 MJ per liter, so a tank that holds a fuel oil voyage is small relative to the cargo. Every alternative gives up volumetric energy, and the giveback ranges from modest to severe. LNG holds about 50 MJ/kg by mass, more than fuel oil, but its cryogenic liquid density of around 450 kg per cubic meter drops its volumetric energy to roughly 21 MJ per liter, so an LNG tank is about 1.7 to 2 times the volume of the fuel oil it replaces, before the insulation around it.
Methanol and ammonia fall further. Methanol carries only about 19.9 MJ/kg, half of fuel oil by mass, and around 15.8 MJ per liter, so a methanol tank runs roughly 2.5 times the size of a fuel oil tank for the same energy. Ammonia sits near 18.6 MJ/kg and about 12.7 MJ per liter. Hydrogen is the extreme case in both directions: about 120 MJ/kg by mass, the highest of any chemical fuel, but as a cryogenic liquid at minus 253 degrees Celsius its density is only around 71 kg per cubic meter, giving roughly 8.5 MJ per liter, less than a quarter of LNG. The DNV and SEA-LNG comparison work puts liquid hydrogen, ammonia, and methanol at about 34, 51, and 63 percent of the volumetric energy of LNG, which is the cleanest one-line summary of why deep-sea hydrogen is hard; the hydrogen volumetric energy density calculator works the gas-versus-liquid storage trade for a given tank. Batteries are off this scale entirely: a marine lithium-ion pack stores about 0.5 MJ/kg, roughly one-eightieth of fuel oil by mass, which confines them to short routes.
| Fuel | LHV (MJ/kg) | Volumetric energy (MJ/L) | Storage state | WtW CO2e potential |
|---|---|---|---|---|
| Heavy fuel oil (baseline) | ~40 | ~36 | Ambient liquid | Baseline |
| LNG | ~50 | ~21 | Cryogenic liquid, -162 C | Limited cut; deep with bio/e-methane |
| Methanol | ~19.9 | ~15.8 | Ambient liquid | Deep cut only if green/bio |
| Ammonia | ~18.6 | ~12.7 | Liquid, -33 C or pressurized | Near-zero if green |
| Hydrogen (liquid) | ~120 | ~8.5 | Cryogenic liquid, -253 C | Near-zero if green |
| Biofuel (HVO) | ~44 | ~34 | Ambient liquid | Up to ~90% lower (feedstock) |
| Biofuel (FAME) | ~37 | ~33 | Ambient liquid | Feedstock-dependent |
| Battery (Li-ion pack) | ~0.5 | ~1 to 2 | Electrochemical | Zero TtW; grid-dependent WtW |
| Nuclear (reactor) | n/a (fission) | n/a | Sealed reactor core | Near-zero operating CO2 |
The figures in the table are lower heating values & approximate liquid volumetric densities at the storage state shown; an individual fuel batch and a specific tank geometry shift them by a few percent, and the per-fuel detail and well-to-wake numbers live in the well-to-wake fuel pathways hub. The dominant hazard and the deployment maturity track the storage state directly. Fuel oil, HVO, and FAME are flammable ambient liquids and are mature or drop-in available today, FAME usually as a blend because it oxidizes in storage. LNG is a mature cryogenic fuel whose flammability pairs with methane slip; methanol is commercial and carries a toxic, low-flashpoint hazard. Ammonia is acutely toxic and at early-commercial stage, hydrogen is highly flammable with a wide flammability range and remains pilot-stage, the Li-ion battery’s thermal-runaway risk confines it to short routes, and the reactor’s radiological hazard keeps fission naval-only. Two patterns carry the design lesson. Mass-dense fuels can still be volume-poor, which is why hydrogen tops the kilogram column and bottoms the liter column. And the volumetric column, not the mass column, is the one that costs an owner cargo space, so a fuel that halves volumetric energy either halves the range or eats the payload.
Tank-to-wake against well-to-wake: two different scoreboards
A fuel’s carbon footprint depends entirely on where you draw the boundary, and conflating the two boundaries is the most common error in fuel comparison. Tank-to-wake counts only what the ship emits from its own funnel: the CO2, methane, and nitrous oxide that leave the stack. Well-to-wake counts the whole chain: extraction or synthesis, processing, liquefaction or compression, transport to the bunker port, and combustion. The two can disagree sharply. Green ammonia and green hydrogen emit no CO2 at the funnel, so their tank-to-wake CO2 is zero, but their well-to-wake footprint depends on whether the electricity that made them came from renewables or from a coal grid. The IMO Net-Zero Framework and FuelEU Maritime both regulate on the well-to-wake basis, which is why the funnel number alone no longer decides compliance, and the IMO Net-Zero Framework and GFI sets out the greenhouse-gas fuel intensity metric that does the pricing.
LNG shows the gap most clearly. Burned in an engine, LNG emits about 25 percent less CO2 per unit of energy than fuel oil, because methane’s hydrogen-to-carbon ratio is higher. But methane that slips through unburned is a greenhouse gas with a 100-year global warming potential of 28 times CO2 under the IPCC Fifth Assessment Report (29.8 for fossil methane under the Sixth), and 84 to 86 times over a 20-year horizon. A low-pressure dual-fuel engine can leak a few percent of its methane unburned during the combustion cycle, and that slip erodes much of the tank-to-wake CO2 advantage once it is counted as CO2-equivalent. The arithmetic and the engine-by-engine slip rates are worked in the methane-slip cluster, and the IRENA and Lloyd’s Register pathway studies both treat the well-to-wake LNG figure as a limited reduction rather than a deep cut for fossil-sourced gas.
The fuels that reach near-zero well-to-wake all depend on a green production route. Green methanol made from captured CO2 and green hydrogen, green ammonia made from renewable electricity, and synthetic e-methane all approach zero net CO2, but only when the inputs are genuinely renewable; the same molecules made from fossil feedstock save little. Biofuels carve their cut at the feedstock: HVO and FAME from waste oils can cut well-to-wake CO2 by up to roughly 90 percent against fossil diesel, but a biofuel from a crop grown on cleared land can be worse than fossil once land-use change is counted. This is the reason the regulators moved to a well-to-wake metric: it stops a ship claiming a zero at the funnel while the emissions sit upstream.
Toxicity, fire, and the IGF Code
A new fuel is a new hazard, and the regulatory architecture sorts the hazards into two regimes. The International Code of Safety for Ships using Gases or other Low-flashpoint Fuels, the IGF Code, was adopted as IMO resolution MSC.391(95) and entered force on 1 January 2017. It governs ships that burn fuels with a flashpoint below 60 degrees Celsius, LNG and methanol foremost, and it adds Part G to SOLAS chapter II-1. The Code is goal-based rather than prescriptive: it sets functional requirements for fuel containment, bunkering, gas detection, ventilation, ignition-source control, and machinery-space arrangement, and lets the design meet them in different ways. Gas carriers that burn their own cargo are exempt, because they already comply with the International Code for the Construction and Equipment of Ships Carrying Liquefied Gases in Bulk, the IGC Code, whose containment and safety provisions cover the cargo and its use as fuel together.
The hazards differ by fuel, and the design follows the hazard. LNG is flammable and cryogenic, so the risk is fire and cold burns plus the asphyxiation hazard of a vapor cloud; the IGF Code’s double-barrier containment and gas-detection rules address it. Methanol is liquid at ambient temperature, which simplifies storage, but it is toxic if ingested, burns with a nearly invisible flame, and has a low flashpoint, so it needs cofferdams, nitrogen inerting, and fire detection tuned to its flame. Ammonia is the hard case: it is acutely toxic to humans at low airborne concentrations, well below the level at which it is flammable, so the dominant risk is a toxic release to the crew rather than a fire. The IGF Code did not originally cover ammonia or hydrogen, so the IMO issues interim guidelines for ammonia-fueled and hydrogen-fueled ships, and the toxicity-driven detail of containment and crew protection is the subject of ammonia as marine fuel. Hydrogen adds its own problem: a very wide flammability range and a low ignition energy, so leak detection and ventilation dominate its design.
The practical effect is that the safety regime, not the chemistry alone, sets the cost of adoption. A methanol retrofit is comparatively contained because the fuel is a room-temperature liquid; an ammonia installation carries a toxic-gas management burden that touches the whole accommodation and machinery layout; a hydrogen installation carries cryogenic and wide-flammability burdens together. Each fuel’s article in this cluster works the safety case in full, and the common thread is that the IGF Code or its interim equivalents turn the abstract hazard into a concrete set of barriers an owner must build and a class society must approve.
The fuels, one rung at a time
LNG and methane slip
LNG is the most mature alternative and the one with the most ships afloat. It is fossil methane cooled to minus 162 degrees Celsius, burned in dual-fuel engines that can switch to oil, and it cuts funnel CO2 by about 25 percent against fuel oil while eliminating sulfur and most particulates. Its weakness is methane slip: a low-pressure Otto-cycle dual-fuel engine passes a few percent of its methane unburned, and because methane’s 100-year GWP is 28 times CO2 under AR5 (29.8 for fossil methane under AR6), that slip eats much of the well-to-wake benefit. Fossil LNG is therefore a transition fuel that buys air-quality gains and a partial CO2 cut, with the deep decarbonization deferred to bio-LNG or synthetic e-methane. The full treatment, including the LNG fuel system and the bunkering chain, sits in LNG as marine fuel, the LNG core-properties calculator carries the density, calorific, and tankage figures, and the slip arithmetic feeds the well-to-wake LNG pathway.
Methanol and methanol engines
Methanol is the alternative gaining ground fastest in the boxship trades after LNG, and its appeal is handling: it is a liquid at ambient temperature and pressure, so it stores in conventional-style tanks with cofferdams rather than cryogenic systems, and a methanol bunkering operation looks much like an oil bunkering. It carries only about half the energy of fuel oil per liter, so the tank runs roughly 2.5 times larger, and fossil-route methanol cuts little well-to-wake CO2; the deep cut needs green or bio-methanol. Two-stroke and four-stroke methanol-burning engines are in commercial service, and the combustion, pilot-fuel ignition, and engine families are covered in methanol as marine fuel, the engine detail in methanol marine engines overview, and the methanol core-properties calculator sets the density and energy figures against fuel oil.
Ammonia
Ammonia is the leading candidate for deep-sea zero-carbon operation precisely because it has no carbon to emit and is far easier to store than hydrogen, liquefying at minus 33 degrees Celsius or under modest pressure at ambient temperature. It already moves worldwide as a fertilizer feedstock, so a production and shipping base exists, and green ammonia made from renewable hydrogen approaches zero well-to-wake CO2. The barrier is acute toxicity: ammonia poisons humans at airborne concentrations far below its flammable range, so the entire safety case is built around preventing and detecting a toxic release, handled under IGF Code principles with IMO interim guidelines. Ammonia combustion also needs careful control of nitrous oxide, itself a strong greenhouse gas, and of unburned ammonia. The full safety, combustion, and infrastructure case is in ammonia as marine fuel, and the ammonia core-properties calculator gives the storage and energy figures that size the tank.
Hydrogen and fuel cells
Hydrogen is the lightest fuel by mass and the hardest to store by volume, which splits its marine use in two. As a cryogenic liquid at minus 253 degrees Celsius it carries only about 8.5 MJ per liter, less than a quarter of LNG, so liquid-hydrogen deep-sea propulsion remains a pilot-stage proposition limited by tank volume and boil-off. Where hydrogen does win is in fuel cells on short-route and harbor craft, where a proton-exchange-membrane fuel cell converts hydrogen to electricity at high efficiency with only water at the tailpipe, sidestepping the combustion losses and the NOx of an engine. The storage chemistry and the deep-sea barrier are in hydrogen as marine fuel, the liquid-hydrogen core-properties calculator carries the cryogenic density and energy figures, and the fuel-cell types, projects, and class rules are in hydrogen marine fuel cells overview.
Biofuels: FAME, HVO, and bio-LNG
Biofuels are the only alternatives that drop into existing engines and tanks with little or no modification, which makes them the fastest available cut for the current fleet. HVO, hydrotreated vegetable oil, is a paraffinic renewable diesel with about 44 MJ/kg and roughly 34 MJ per liter, close enough to fuel oil to run neat in many engines, and from waste feedstock it can cut well-to-wake CO2 by up to about 90 percent. FAME, fatty-acid methyl ester biodiesel, carries about 37 MJ/kg and is usually used as a blend because it oxidizes and absorbs water in storage. Bio-LNG is biomethane that substitutes for fossil LNG in the same engines and tanks. The catch across all three is feedstock: the well-to-wake cut depends entirely on the source, and a crop-based biofuel grown on cleared land can lose its advantage once land-use change is counted. The feedstock accounting and the engine compatibility are in biofuels in shipping.
Battery-electric and hybrid
Batteries store about 0.5 MJ/kg at pack level, roughly one-eightieth of fuel oil, so a battery sized for an ocean voyage would outweigh the cargo. That energy density confines full-electric propulsion to short crossings that recharge at every berth and to harbor craft, where the fixed route and frequent charging make the weight workable. Norway leads the fully electric ferry fleet, with crossings that charge in the few minutes alongside, and the operational and charging model is covered in battery-electric ferries overview. On larger conventional ships the battery’s role is different: a hybrid pack shaves load peaks, provides spinning reserve, and lets the engines run nearer their efficient point, the subject of battery-hybrid propulsion. The well-to-wake footprint is zero at the ship but depends entirely on the grid that charges the pack.
Nuclear
Nuclear propulsion sidesteps the energy-density problem completely: a fission core needs refueling on a timescale of years, not days, and emits no operating CO2. The technology is proven at sea, but only in navies and a handful of icebreakers, because the merchant barriers are commercial and regulatory rather than technical: port acceptance, liability and insurance, security, crew licensing, and waste handling. The reactor types, the naval and icebreaker fleets, and the merchant prospects are in naval nuclear propulsion overview. For ocean freight it remains a long-horizon option, watched closely because it is the only candidate that solves range without a volume or refueling penalty.
Bunkering readiness: can the ship refuel where it trades
A fuel that a ship cannot buy at its trading ports is a fuel the ship cannot use, and bunkering availability sorts the ladder differently from energy density. LNG bunkering is established at a growing list of major ports with dedicated bunker vessels, which is part of why LNG leads the orderbook. Methanol bunkering is expanding because it can largely reuse liquid-fuel handling, and new methanol bunker vessels are entering service. Ammonia and hydrogen bunkering are at the pilot and first-of-kind stage, so an owner ordering an ammonia ship is partly betting on the bunker network arriving in step with the fleet. Biofuels enjoy the widest reach because they move through the existing oil bunkering chain. Battery charging needs shore power and high-capacity connections at each terminal, which ties the fully electric ferry to a fixed route with built charging infrastructure.
The bunkering picture is also a chicken-and-egg problem that the orderbook data tracks. DNV’s Alternative Fuels Insight platform records the dual-fuel orderbook and the bunkering-vessel orders alongside it, and the pattern in 2026 is that the established fuels, LNG and methanol, draw the orders while ammonia and hydrogen wait for infrastructure to mature. An owner reads that data the way a charterer reads a freight curve: it tells which fuel will be buyable at scale over the ship’s 25-year life, not just at delivery. The same well-to-wake intensity that decides compliance under the IMO Net-Zero Framework and GFI also shapes which fuels attract the bunkering investment, because a fuel that cannot meet the intensity target will not draw the infrastructure.
How a bunkering operation differs by fuel
The physical act of taking fuel aboard changes with the fuel’s state, and the difference drives both the cost and the safety case. An oil bunkering is a room-temperature liquid transfer through a hose, and the crew’s main worries are spill containment and the gauge. A methanol bunkering looks much like it because methanol is also an ambient-temperature liquid, so it largely reuses the oil-bunkering equipment and procedure, with added nitrogen inerting and vapor control for methanol’s toxicity and low flashpoint. That continuity is half of why methanol scaled into the container trades faster than the cryogenic fuels: the bunkering operation did not have to be reinvented.
An LNG bunkering is a different operation. The fuel arrives at minus 162 degrees Celsius, so the transfer uses insulated lines, cooled-down equipment, and a managed boil-off, and the operation follows the truck-to-ship, ship-to-ship, or terminal-to-ship modes that ports built out over the past decade. The crew manages cryogenic burns, a denser-than-air cold vapor that pools, and the tank pressure. Liquid hydrogen pushes every one of those constraints harder at minus 253 degrees Celsius, with a wider flammability range and a lower ignition energy, which is the operational reason hydrogen bunkering sits at the first-of-kind stage rather than in routine service.
Ammonia bunkering carries the toxic-release hazard rather than the fire hazard as the dominant risk, because ammonia poisons the crew at airborne concentrations far below the flammable range. So an ammonia transfer needs gas detection, exclusion zones, vapor-return lines, and crew protection sized for a toxic leak, and the IMO interim guidelines for ammonia-fueled ships govern the arrangement until the IGF Code is amended to cover it. Battery charging is not a bunkering at all but a high-power electrical connection: a short-route ferry takes a megawatt-scale charge in the minutes alongside, which needs shore-side grid capacity, an automated connector, and a charging window built into the schedule. A fully electric crossing that loses its shore connection loses its propulsion, so the charging infrastructure is part of the vessel’s safety case in a way an oil bunker never was. The detail of each operation lives in the fuel’s own article, and the common thread is that the bunkering operation, not just the combustion, sets the safety regime an owner must build, and a class society must approve before the first transfer.
Cost and availability: the figure that decides
Cost is where the alternative-fuel decision is usually settled, and on a straight energy-cost basis every alternative starts behind fuel oil. The green versions cost most: green ammonia and green hydrogen carry the cost of renewable electricity plus electrolysis plus synthesis, and green methanol adds the cost of captured CO2. The Lloyd’s Register and UMAS techno-economic work and the IRENA pathway studies both put the green fuels well above fossil fuel oil per unit of energy at present, with the gap closing as renewable electricity and electrolyzer costs fall and as the carbon price on the fossil baseline rises. That second half matters: the alternatives do not have to beat fuel oil’s raw price, only fuel oil’s price plus the carbon cost the regulators now attach to it through the IMO Net-Zero Framework, FuelEU Maritime, and the EU Emissions Trading System.
The right way to read the cost gap is per tonne of CO2 abated, not per tonne of fuel, because the regulators price carbon and the owner is buying compliance as much as energy. A fuel that costs three times fuel oil per unit of energy but cuts well-to-wake CO2 by 90 percent buys its abatement at a calculable cost per tonne, and that figure is what competes against the carbon price the IMO Net-Zero Framework, FuelEU Maritime, and the EU Emissions Trading System now attach to the fossil baseline. As those carbon prices rise on a known schedule and renewable-electricity and electrolyzer costs fall, the crossover point at which the green fuel is cheaper all-in moves toward the present, which is the economic mechanism the pathway studies model. The same logic explains why a transition fuel that cuts a quarter of the carbon at a small premium can still pencil out today while a deep-cut fuel waits for its supply chain and its price to mature.
Availability compounds the cost question. There is not yet enough green hydrogen, green ammonia, or green methanol produced worldwide to fuel a meaningful share of the fleet, so even an owner willing to pay faces a supply constraint in the near term. Biofuel availability is limited by sustainable feedstock, which caps how far waste-based HVO and FAME can scale before crop-based sources with worse land-use profiles fill the gap. This is the reason the realistic forecast is a fuel mix rather than a single winner: each fuel scales against a different constraint, and a deep-sea bulker, a short-sea ferry, and an inland push-boat optimize the five tests differently. The per-fuel cost and intensity detail, broken out pathway by pathway, sits in well-to-wake fuel pathways, and the blended-fuel well-to-wake calculator gives the WtW intensity of a dual-fuel mix for the compliance arithmetic.
How the fuel converts: engine, fuel cell, and the efficiency penalty
Energy density sizes the tank, but the conversion device decides how much of that stored energy turns into thrust, and the two alternatives split on this. A marine diesel or dual-fuel engine is a heat engine, and a large two-stroke crosshead converts roughly 50 percent of the fuel’s lower heating value into shaft work, the highest thermal efficiency of any commercial prime mover. A four-stroke medium-speed engine sits a few points lower. So a fuel burned in an engine, LNG, methanol, ammonia, or a biofuel, loses about half its chemical energy to heat, which is why a slip of a few percent of unburned methane matters against a 50 percent baseline rather than a 90 percent one.
A fuel cell changes the arithmetic. A proton-exchange-membrane cell running on hydrogen converts the fuel electrochemically rather than thermally, sidestepping the Carnot limit, and reaches roughly 50 to 60 percent electrical efficiency at the stack, with the system efficiency lower once the balance of plant draws its share. The tailpipe is water and the device has no NOx because there is no flame, which is the structural reason hydrogen pairs with fuel cells on the routes where its volume penalty is survivable. The fuel-cell types and their marine projects are worked in hydrogen marine fuel cells overview, and the engine-side combustion of each fuel sits in its own article.
The conversion device also carries part of each fuel’s emissions problem. A low-pressure Otto-cycle dual-fuel engine slips methane during the valve overlap and the combustion cycle, and a high-pressure diesel-cycle engine slips far less but costs more and needs a high-pressure gas system. Ammonia combustion produces nitrous oxide, itself a greenhouse gas with a 100-year global warming potential around 265 times CO2 under AR5, plus unburned ammonia at the stack, so an ammonia engine needs aftertreatment tuned to both. The slip and the aftertreatment are engine-specific, not fuel-specific, which is why the same molecule scores differently depending on the machinery it feeds.
Matching the fuel to the ship type
No fuel wins across the fleet because a deep-sea bulker, a short-sea ferry, and a harbor tug optimize the five tests differently, and the orderbook reflects that split rather than a single transition. A 14,000-TEU container ship on a Far East to Europe rotation runs for weeks between bunker ports and is volume-constrained, so it favors the fuel that gives the most range per cubic meter of tank: LNG today, with methanol the fast-growing second because its ambient-liquid handling keeps the fuel system simple. The same ship cannot run on batteries at any plausible pack size, and liquid hydrogen would eat too much cargo space for the range.
A short-sea ferry inverts the constraints. It runs a fixed route, returns to the same berth many times a day, and can recharge or refuel at known points, so a battery pack sized for one crossing plus a margin is workable, and the battery-electric ferries overview traces the Norwegian crossings that proved the model. A harbor craft or an inland push-boat sits in the same regime: short legs, frequent return to base, and a fixed charging or bunkering point, which is also where hydrogen fuel cells find their first commercial footing.
Between the two extremes sit the trades that will carry the hard decarbonization argument. A deep-sea tanker or bulker on a long ocean leg needs a zero-carbon fuel with deep-sea range, which points at green ammonia for its storability or green methanol for its handling, both gated on green-fuel supply and bunkering reach rather than on engine readiness. A cruise ship adds the toxicity constraint of thousands of people aboard, which weighs against ammonia and toward methanol or LNG with a battery layer. The hybrid pack that shaves load peaks and provides spinning reserve, covered in battery-hybrid propulsion, layers onto any of these, so the realistic fleet of the 2030s is a mix keyed to trade, not a single winner.
Reading the orderbook: what the uptake data shows
The DNV Alternative Fuels Insight platform tracks the alternative-fuel orderbook and the in-service fleet ship by ship, and it is the cleanest public read on which fuels owners are actually betting on rather than discussing. The pattern through 2026 is consistent: LNG dual-fuel leads the alternative-fuel orderbook by a wide margin, methanol dual-fuel has moved into a firm second after a surge of container-ship orders, and ammonia and hydrogen appear as a small but growing count of first-of-kind and pilot vessels rather than a fleet. The same platform tracks the bunker-vessel orderbook alongside the fuel-ship orderbook, which is the supply-side signal that tells whether a fuel will be buyable at scale.
That orderbook is a 25-year bet, not a delivery-day choice, which is the discipline it imposes on an owner. A ship ordered today delivers in two to three years and trades into the 2050s, across the period when the IMO Net-Zero Framework and GFI tightens its greenhouse-gas fuel intensity targets and the carbon price on the fossil baseline climbs. So the fuel choice is a forecast of which fuel will be both buyable and compliant for two decades, and the dual-fuel engine is the hedge: a ship that can burn fuel oil and methanol, or fuel oil and LNG, keeps its options open while the green-fuel supply builds. The IRENA and Lloyd’s Register pathway studies both frame the transition as a staged shift through transition fuels rather than a single jump, which is the same logic the orderbook shows in metal.
How the cluster articles fit together
This hub is the comparison; the detail lives one level down in eleven articles, each carrying one fuel’s full case. The two most mature liquid and gas routes are in LNG as marine fuel and methanol as marine fuel, with the methanol engine families in methanol marine engines overview. The carbon-free candidates are in ammonia as marine fuel, hydrogen as marine fuel, and the fuel-cell detail in hydrogen marine fuel cells overview. The drop-in route is in biofuels in shipping. The electric routes are in battery-electric ferries overview and battery-hybrid propulsion, and the long-horizon option in naval nuclear propulsion overview. The safety regime that governs the low-flashpoint fuels runs through IGF Code.
Read in the order of the five tests, the eleven articles trace each fuel from its energy density through its carbon boundary, its safety regime, its bunkering reach, and its cost. The cluster sits under decarbonization and alternative fuels as the fuel-choice branch of the wider decarbonization story, and it connects across to well-to-wake fuel pathways for the intensity accounting and to the IMO Net-Zero Framework and GFI for the carbon price that now drives the whole decision. No single fuel wins all five tests, which is exactly why the choice is a trade-specific judgment and not a universal answer.
Limitations
This article compares the marine fuel options on their governing trade-offs; it is not a substitute for a fuel-specific feasibility study, a class society’s design approval, or the actual rules of the IGF Code, the IGC Code, and the IMO interim guidelines for ammonia and hydrogen. The energy-density figures are lower heating values and representative liquid volumetric densities at the storage state shown; a specific fuel batch, tank geometry, and insulation arrangement shift both the mass and the volume figures, and a real fuel-system sizing must use the supplier’s and the engine maker’s data, not these round numbers.
The well-to-wake CO2-equivalent assessments depend entirely on the production pathway and the methodology and global-warming-potential horizon chosen, and they change as regulators update the default factors; the figures here describe the direction and magnitude of the differences, not certified compliance values. Methane-slip rates, ammonia nitrous-oxide formation, and biofuel land-use effects are all engine-, fuel-, and feedstock-specific, and a compliance calculation must use the current regulatory factors under the IMO Net-Zero Framework, FuelEU Maritime, or the EU ETS for the specific ship and voyage. Bunkering availability, fuel price, and green-fuel supply move continually, so the orderbook and uptake observations reflect the published DNV Alternative Fuels Insight position and will shift; an owner’s decision must rest on current market and infrastructure data for the ship’s intended trade.
See also
- Decarbonization and alternative fuels: the parent hub for the whole decarbonization story.
- Well-to-wake fuel pathways: the intensity accounting fuel by fuel.
- IMO Net-Zero Framework and GFI: the greenhouse-gas fuel intensity metric that prices the fuels.
- LNG as marine fuel: the most mature alternative and methane slip.
- Methanol as marine fuel: the ambient-liquid alternative gaining ground.
- Methanol marine engines overview: the engine families that burn methanol.
- Ammonia as marine fuel: the carbon-free candidate and its toxicity case.
- Hydrogen as marine fuel: the storage barrier to deep-sea hydrogen.
- Hydrogen marine fuel cells overview: fuel-cell types, projects, and rules.
- Biofuels in shipping: FAME, HVO, and bio-LNG as drop-in cuts.
- Battery-electric ferries overview: short-route fully electric operation.
- Battery-hybrid propulsion: peak-shaving and spinning reserve on conventional ships.
- Naval nuclear propulsion overview: the long-horizon zero-carbon option.
- IGF Code: the safety regime for low-flashpoint fuels.