[kbolino]

Though it may never prove viable, hydrogen from electrolysis of water creates no emissions other than oxygen. While large ships may be able to use nuclear power directly, the risks of a tenfold increase in small floating nuclear reactors in civilian hands are not trivial. Hydrogen provides a way to keep power generation centralized, secure, and efficient while also distributing it where needed. But this assumes an extreme excess of power, something only possible with nuclear or large-scale renewables, which may never come to pass in sufficient quantities or without even bigger problems.

[Galaxeblaffer]

I don't disagree that nuclear propulsion has a long way to go before becoming safe enough for civilian use and might require extra security and regulation forever, but we need to do something other than dumping bunker fuel waste into the sea and atmosphere even coubting in several terrorist attacks and accidents it might still end up killing less people using nuclear. As you say going with hydrogen and by extension ammonia, which is probably the only viable way to fuel ships, will require unthinkable amounts of renewables which is mostly intermittent energy meaning hydrogen electrolysis plants will have to be ramped up and down as the wind blows or the sun shines. this makes the efficiency calculations even worse for an already super wasteful undertaking. i guess we could potentially use nuclear but why waste 80-90% of the energy created by converting and then using hydrogen ? My opinion is that we need to throw much more money at making nuclear fission safer.

[kbolino]

I think most of the heat and fuel will get wasted anyway. A nuclear-powered cargo ship is going to be closer to a nuclear-powered submarine than a nuclear-powered aircraft carrier, in terms of the number of people being supported and the needs beyond propulsion. That means running the reactor below its optimal fuel efficiency, and then intentionally discarding a portion of its heat most of the time to retain some spare capacity for demand spikes and emergencies. Of course, small-scale nuclear reactors are an active area of research, but AFAIK the results haven't been promising yet. So it's an avenue, but not the only one.

Running a reactor full-bore on shore, pumping ocean water, desalinizing it, electrolyzing it to hydrogen, cryogenically separating nitrogen from air, and combining the result to produce ammonia, might look really inefficient in isolation, but might be efficient enough when compared against the alternative. When the costs of logistics, personnel, and capital are accounted for, it might even come out ahead. That having been said, the amount of waste oxygen this would produce is immense, and could create risks of its own. Ammonia is also more likely to produce NOx when combusted in air than hydrogen is, so that's not great, either.