The shipping industry is unusual among the heavy emitters because the technical path to decarbonizing it is reasonably well-understood, the policy backdrop is unusually clear, and yet the industry as a whole is moving more slowly than the policy and the technology would suggest. I find this interesting enough to be worth unpacking, because the bottleneck turns out to be in a place that most people wouldn't guess.

Some context. Global shipping moves about 80% of world trade by volume. Container ships, bulk carriers, tankers, and the rest of the merchant fleet —roughly 100,000 vessels — burn about 300 million tonnes of bunker fuel a year. Bunker fuel is the residual from refining, the cheapest and dirtiest fossil fuel still in commercial use. Total emissions from shipping are roughly 3% of global fossil fuel consumption, concentrated in a relatively small number of large ships. Per ton of CO₂,shipping is among the cheapest sectors to decarbonize, but only if you can replace the fuel.

The International Maritime Organization, which is the UN body that regulates international shipping, committed in 2023 to net-zero emissions from international shipping around 2050, with intermediate targets in 2030 and 2040. Most of the major shipping companies — Maersk, MSC, CMA CGM — have made their own commitments and are starting to order ships designed for alternative fuels. This is a slow industry, but it's moving in one direction.

The interesting question is which fuel. There are basically four candidates, and the comparison is instructive.

Battery-electric is out, except for very short routes. A large container ship crossing the Pacific consumes roughly 40,000 tonnes of bunker-fuel-equivalent energy. To replace that with lithium-ion batteries would require the entire ship to be batteries and would still come up short on range. Short-haul ferries and harbor craft can run on batteries. Ocean shipping can't.

Biofuels work technically, but the supply constraints are severe. To replace global bunker fuel demand with biofuels would require committing land area roughly equivalent to a large fraction of global agriculture, which isn't feasible. Biofuels will play a role on the margin, but they can't be the primary path.

Methanol is the front-runner today, mainly because it's the easiest toretrofit. Methanol can be made from biomass or from hydrogen and captured CO₂. Maersk has ordered a fleet of methanol-capable ships. The challenge is that green methanol is expensive —currently three or four times the cost of bunker fuel — and the CO₂ feedstock supply is its own bottleneck.

Ammonia is, I think, the most likely long-term winner for the large fraction of the fleet. Ammonia is made by combining nitrogen with hydrogen at high pressure, and the hydrogen can come from electrolysis. It has nine times the volumetric energy density of compressed hydrogen, is liquid at moderate pressure, and ships have infrastructure that already exists because the fertilizer industry ha sbeen moving ammonia around the world for a century. The ports know how to handle it. The bunkering infrastructure can be retrofit. The engines exist, in early commercial form.

Every path forward, except for biofuels, runs through hydrogen. Methanol synthesis needs hydrogen. Ammonia synthesis needs hydrogen. Synthetic diesel needs hydrogen. The shipping industry's transition is, when you trace it back, a transition to electrolytic hydrogen as an input to whatever fuel turns out to be the carrier.

The buildout required to support this is significant. Modeling out a realistic scenario where shipping transitions over the next 25 years gives you somewhere between 200 and 600 gigawatts of new electrolyzer capacity, just to serve the shipping fuel market. That's more than the entire current global installed base of electrolyzers, by a lot. It's comparable in scale to the entire global solar PV manufacturing capacity today.

This is where the membrane comes in, and where the bottleneck I mentioned at the beginning actually lives. The electrolyzer industry, at current efficiency and current iridium loading, can't physically support that scale of build out. The [iridium supply constraint](/articles/materials) caps the industry at roughly 400,000 stacks per year worldwide at conventional loadings. Hitting200+ gigawatts of new capacity over two decades, just for shipping, requires either dramatically more iridium supply (which isn't available) or dramatically less iridium per stack (which is a membrane problem). Same logic for current density: building enough electrolyzer capacity in the available time requires stacks that produce more hydrogen per square meter of active area, and current density is gated by the membrane.

So the path to decarbonizing shipping, as a chain of dependencies, looks like this: shipping needs ammonia, which needs electrolytic hydrogen, which needs millions of square meters of new electrolyzer capacity, which is constrained by iridium loading and current density, which are constrained by membrane performance. Five steps. The last one is a thin plastic substrate.

This is the kind of chain that I think is hard to see from the application end. If you're thinking about shipping decarbonization, you're probably thinking about engine technology and bunkering infrastructure and IMO regulations and shipping company commitments. You're probably not thinking about a thin polymer film in a coating line in Rochester, New York. But the path runs through the polymer, because the polymer is the platform underneath the industry that has to scale to supply the fuel.

I find this kind of dependency chain clarifying because it illustrates the point of working on a platform technology. We don't sell to shipping companies. We probably won't sell to shipping companies. But the shipping industry's decarbonization, when it eventually happens, will run through customers we do sell to — the electrolyzer manufacturers who are starting to ship at gigawatt scale and who can't get to terawatt scale on the existing membrane chemistry.

The shipping industry is the kind of demand pull that doesn't show up on the membrane manufacturer's customer list directly, but that shapes the whole trajectory of the market we serve. There are a lot of demands like this. The shipping industry is one. The chemicals industry is another. They aren't customers. They're the reason the customers exist.

‍