Burns, blindness, asphyxiation and agonising death can all be caused by even moderate concentrations of ammonia — a highly toxic, hydrogen-derived chemical that is expected to become a major zero-carbon shipping fuel and international H2 carrier.
Coupled with the damage that ammonia could cause to marine ecosystems — one expert tells Recharge a major spill at sea could “sterilise cubic miles of ocean” — it is little wonder that some observers are feeling uneasy at the prospect of NH3 being shipped around the world, especially in the massive volumes being discussed.
According to a recent report by the International Renewable Energy Agency, one quarter of the world’s hydrogen will be internationally traded in 2050, in a 1.5°C scenario, with 45% of that shipped as ammonia — roughly 69 million tonnes of H2. That’s almost the equivalent of all the (mainly grey) hydrogen being produced globally today.
Irena also states that in its 1.5°C pathway, the demand for ammonia would increase fourfold by mid-century to 688 million tonnes — with 197 million tonnes of that used as a shipping fuel and 127 million tonnes as a hydrogen carrier.
Ammonia — which is mainly used today for fertiliser production and oil refining — is widely preferred as both a shipping fuel and H2 carrier because it far more energy dense by volume than compressed or liquefied hydrogen, and is much easier to store than the latter, which requires cryogenic temperatures of minus 253°C. Ammonia gas, however, can be stored as a liquid at room temperature under a pressure of just 10 bar, or under atmospheric pressure at minus 33°C.
Safety at sea
One of the problems with dramatically scaling up the shipping of ammonia is that as more vessels take to the high seas, the likelihood of a serious incident grows.
Already, the maritime industry has an “unenviable” safety record, according to the sector’s own Together in Safety group, which flagged a rise of 5% in shipping casualties and incidents to 2,815 in 2019 compared to 2018.
“A lot of vessels sink,” David Cebon, professor of mechanical engineering at the University of Cambridge, tells Recharge, noting that 40% of all shipping cargo today consists of fossil fuels.
“Two dozen ships go down every year and if 40% of them were carrying ammonia to replace fossil fuels as the way to transfer energy, then that would be ten ships a year. [Even] if you had half or a quarter of that… you’ve still got a lot of ammonia ending up in the ocean.
And make no mistake, ammonia is nasty. It is a caustic, water-seeking chemical that attaches itself to moisture — including in a person’s eyes, mouth, lungs, throat or skin.
Exact exposure limits vary from regulator to regulator but even very low concentrations in the air of just 30 parts per million (ppm) — possible from a very small leak, or even sometimes regular operation — can cause breathing difficulties if a person is exposed to it for more than 15 minutes. At 100ppm, eye and throat irritation will occur, and at 500-700 ppm any exposure at all becomes dangerous, with a risk of burns to throat, skin, lungs and eyes, potentially causing blindness. At higher concentrations, the risk of death increases significantly.
Severity of injury depends on the exact concentration of ammonia and length of exposure but in a worst-case scenario it can cause the respiratory tracts to burn and swell shut, suffocating the victim.
It doesn’t end there. When ammonia spills — for example, in a collision at sea that causes the tank to rupture — and especially if hits water, it reacts with condensation in the air to form a heavy, toxic mist that does not dissipate.
An ammonia gas cloud is “very difficult to handle”, explains George Mallouppas, associate scientist at the Cyprus Marine and Maritime Institute (CMMI), noting that while a port or industrial area can be evacuated to avoid a spill cloud, it is “nearly impossible” to do so on a vessel at sea.
And while ammonia is significantly less flammable than liquefied natural gas (LNG), it is nevertheless flammable, so any leak at sea would be accompanied by a risk of explosion or fire.
A serious ammonia spill would also be devastating under the waves. Ammonia dissolves in water to form ammonium hydroxide, which is highly toxic to marine life.
“A concentration of 1 ppm of ammonia in water is sufficient to kill many marine organisms,” Paul Martin, co-founder of the Hydrogen Science Coalition, tells Recharge. “A ship sinking with a cargo of ammonia could potentially sterilise cubic miles of ocean.”
The effects would be worse if the spill happened in a bay or port, where naturally-occurring ammonia levels are already raised due to high levels of decomposing organic matter on the seabed and low water exchange, says Manos Moraitis, assistant scientist at the CMMI’s Marine and Coastal Ecosystems Centre.
“Ammonia is considered one of the most common causes of fish kills in aquatic ecosystems,” he tells Recharge. “The impact of ammonia leakage in ports, estuaries and semi-enclosed gulf bays will be most severe.”
Good safety record to date?
On land, there have been several high-profile ammonia accidents, not least the devastating 1976 road accident in Texas, where NH3 fumes from a crashed ammonia tanker killed five people and injured 178 more.
And as recently as last month, Carlsberg was fined £3m ($3.6m) by UK regulators for an ammonia leak from one of its breweries, which killed one and seriously injured another.
But it is worth noting that there have been relatively few ammonia incidents at sea so far, despite the fact that NH3 has, for many years, been shipped by sea at volume — around 8% of total production, according to Martin’s estimates.
The IHS Markit database on shipping accidents reports just four ammonia-related incidents as of 2021, of which only one related to a leak of NH3 being transported as a cargo, an event with no injuries, fatalities or environmental damage.
The remaining three, however, were fatal accidents — all on fishing vessels where ammonia used as an on-board refrigerant either leaked or exploded.
But Cebon warns that the probability of more accidents as a result of the expected expansion in ammonia transport will be amplified by the lack of effective regulation at sea — as well as official incompetence, inaction or corruption.
One is the example of the Safer oil tanker, currently rusting off the coast of Yemen and at risk of leaking 1.14 million barrels of oil into the sea due to a political impasse over who is responsible for it at a time of civil war.
A tanker harboured near a populated area would be a tempting target for terrorism, and could produce deaths and injuries on a catastrophic scale
And the explosion of ammonium nitrate in Beirut, Lebanon, in August 2020 — which killed 218 people due to official negligence — demonstrates the catastrophic consequences of a single failure relating to hazardous substances, even if it is handled carefully and safely by the vast majority of those who use it.
Human error is, according to Together in Safety, a contributing factor in 75-96% of incidents at sea, as extended periods onboard and time away from loved ones erode crews’ effectiveness.
This risk factor is further exacerbated by the need for well-trained staff with specialist knowledge on how to handle ammonia — which some shipping companies may not be able or willing to splash out on.
Martin goes even further. “The big risk of ammonia as a bulk energy carrier is terrorism,” he says. “A tanker harboured in a location near a populated area would be a tempting target and could produce deaths and injuries on a catastrophic scale.”
Risks grow when using ammonia as fuel
Another problem for ammonia’s relatively clean safety record is that its use as a fuel is relatively new and untested — and is inherently more risky.
Unlike cargo ammonia, which is carried onboard in a sealed tank, ammonia used as a fuel will be integrated into the operation of the vessel.
This means pipes carrying fuel from the tank to the fuel preparation room (the space containing pumps, compressors, etc) to the engine — maximising the surface area that could leak or sustain damage.
It means ammonia burning in an engine that could leak or explode. It means an engine room with crew working in it, who could all be exposed to leaking fuel. It means fugitive ammonia emissions from the engine escaping out of the exhaust.
The good news is that LNG has already made the transition from cargo to fuel relatively safely, so there is a model to follow for other liquefied gases.
The bad news is that, as classification society DNV makes clear in its recent Ammonia as a Marine Fuel Safety Handbook, LNG-fuelled vessel design is not appropriate for ammonia, so a whole new approach is needed — to both design and operation.
“The crew involved must be properly skilled, upskilled or reskilled in order to meet the specific requirements of handling a new fuel with physical and chemical properties and behaviour significantly different than the established marine liquid fuels,” explains Professor Elias Yfantis, senior scientist at CMMI’s Marine and Offshore Science, Technology and Engineering Centre.
DNV suggests a number of mitigation measures to make ammonia safe to load, transport and burn at sea.
Firstly, it suggests locating the ammonia tank away from ship and cargo operations, and “carefully” considering placement to minimise risk from external events, such as grounding or collision. It also recommends gas-tight secondary barriers around the tank, pipes and the engine, to contain any potential leaks and stop them leaching into areas where crew might be operating.
In addition, it calls for air-lock exits for enclosed spaces where ammonia and people would co-exist, so that crew members can quickly escape and trap any escaping NH3. And full personal protection equipment (PPE) is a given, along with an effective management system for so-called “boil-off gas” — pockets of ammonia gas that would form as the cold liquid ammonia in the fuel tank warms up.
At port, loading and unloading of ammonia should be carried out as remotely as possible, and carried out with well-trained staff, using specialist hose couplings that reduce the risk of leaks. DNV adds that fuel-loading hoses should be purged with inert gas to flush out any residual ammonia.
And in the event that any of these systems fail? Mallouppas is confident that leak detection systems on the market today are up to the job of alerting crew to potential leaks, allowing them to intervene before the situation escalates.
“There are commercially available sensors and clever systems that are mature enough to detect leaks quickly enough,” he says. “The response time would be as quick as any other system that detects leaks. There are also new systems that will need to receive approval.”
In general, CMMI, which recently published a paper on the use of ammonia as an energy carrier, is of the view that while ammonia is indeed a dangerous substance to ship, the combination of regulation and innovation will be enough to mitigate the risks.
“CMMI believes ammonia to be an effective energy carrier provided that technologically matured, applicable, safe and cost-effective solutions will be fully available, properly validated and approved by the classification societies,” it tells Recharge.
Martin is less optimistic when asked about the likelihood of effective regulation. “I’m confident that we could do so, but I’m also confident that we would not do so until after a significant loss of life forced the imposition of regulations,” he says.
It is hard not to draw comparisons with nuclear power. Like ammonia, nuclear has the potential to decarbonise vast swathes of the energy system and has a relatively good safety record — but this does very little to calm public anxiety about the potentially catastrophic consequences that could occur.
Like nuclear power, ammonia also has no room for manoeuvre on safety. It will require a gargantuan effort and significant investment from shipping companies, engine-makers and naval architects — as well as politicians and regulators — to ward off even one devastating accident.
According to the latest Alternative Fuels Insight report from DNV, not a single ammonia-powered vessel has been ordered to date — although it expects the first orders soon.
The technology, however, is not quite there yet, with the first ammonia engines still being developed.
For instance, MAN Energy Solutions aims to have a commercially available two-stroke dual-fuel ammonia engine for large ocean-going vessels in 2024, while Finnish supplier Wärtsilä plans to retrofit its first two-stroke ammonia engine into a vessel by 2025 — after securing €10m in funding from the European Commission.
Japanese shipping company Mitsui OSK Lines signed a memorandum of understanding with MAN in 2021 to purchase ammonia engines from the German company, but there was no deadline on the arrangement.
“Ammonia has generated a lot of interest, especially from the deep-sea ship segments, and it has a lot of potential — but developing an engine that is powered by ammonia has been a challenge,” Peter Kirkeby, principal specialist on dual-fuel engines at MAN, was quoted as saying in a DNV report published in February.
“One of the biggest hurdles is how to burn ammonia efficiently to extract the maximum amount of power while making sure the engine is still a compact design.”
By nature, NH3 burns much more slowly than diesel oil, and needs a higher temperature to start “autoignition”. This means that sustaining combustion is harder than with other fuels.
“And, of course, you also need to ensure that the engine allows for the usual performance peaks that come with acceleration, etc,” added Kirkeby. “We are planning for a final fuel mix that would contain around 95% ammonia and 5% of a pilot fuel such as marine gas oil. In the future this could even be biofuel.”
In the short term at least, ammonia seems to be losing out in the alternative-fuels shipping market.
DNV says that 54 methanol-powered ships are currently in service or have been ordered, compared to eight hydrogen vessels and zero ammonia ones.
Methanol (CH3OH) has an even higher volumetric density than ammonia — 4.33kWh per litre, compared to 3.5 — and it is also far easier to handle, being a liquid at room temperature under atmospheric pressure. However, methanol does contain carbon, and to be “carbon-neutral” would require a biogenic source of CO2 for carbon capture, green hydrogen as its base chemical, as well as renewable energy to power the energy-intensive manufacturing process.
According to Recharge’s sister newspaper, TradeWinds, three big-name shipowners — Denmark’s Maersk, Germany’s MPCC and France’s CMA CGM — have already ordered 26 methanol dual-fuel vessels between them. And three more major operators — Cosco, Eastern Pacific and Pacific International Lines — are eyeing methanol propulsion for their next round of new container ships.
Dual-fuel engines that can run on both methanol and traditional fossil fuel are the favoured option at present because methanol is not yet available as a shipping fuel at any kind of scale.
“Owners are waking up to the fact that the next alternative — ammonia — is not that easy,” one ship broker told TradeWinds, pointing to ammonia’s toxicity and the need to remove nitrous oxide — a potent greenhouse gas produced when burning ammonia — from the exhaust stream.