Marine nuclear propulsion

A reactor has three advantages over a diesel engine that matter little on land and a great deal at sea: energy density, endurance, and the fact that it needs no air to run. That last one settled everything, because a nuclear submarine never has to surface to charge batteries. This is why the first pressurised water reactor was not built for an electricity grid but for the USS Nautilus, which first steamed under nuclear power in 1955, with that design only later brought ashore at Shippingport in 1957. A large share of Western PWRs descends from that naval lineage. The military scale accumulated since then is enormous and not widely known: more than 160 vessels sail today with over 200 reactors aboard, and marine operation adds up to more than 14,000 reactor-years, more than the entire power fleet of several countries combined.

The civilian side is a much shorter story. Exactly four nuclear merchant ships were built, and none is operating today. The American NS Savannah, launched in July 1959 and in service from 1962 to 1972, carried a 74 MWt Babcock & Wilcox PWR driving 20,300 shaft horsepower; she sailed 450,000 nautical miles and called at 45 foreign and 32 domestic ports, with room for 8,500 short tons of cargo and 60 passengers. Germany's Otto Hahn entered service in October 1968 with a 38 MWt reactor and was the technical success of the group: 650,000 nautical miles under nuclear power in nine years, and a first refuelling in 1972 after 250,000 miles on 1.7 tonnes of low-enriched uranium. Japan's Mutsu, at 36 MWt, barely sailed at all. And the Soviet Sevmorput, launched in 1986 and delivered at the end of 1988, is the largest of the four: a container and lighter carrier with a 135 MWt KLT-40, 33,980 tonnes deadweight and capacity for 1,328 containers, refuelled only twice in its entire life.

What happened to the Mutsu deserves precision, because it lodged in public memory in a way the facts do not support. On 1 September 1974, during her first power ascent and with the reactor at just 1.4% of capacity, a leak of fast neutrons and secondary gamma rays was detected through the shield: a streaming problem caused by a poorly designed shielding configuration, one that Westinghouse had flagged during its design review without the warning being acted on. No radioactive material escaped, but the Japanese press reported it as though some had, and fishermen at Ōminato blockaded the ship's return to her home port for more than fifty days; she got back only on 15 October, after an agreement was reached. The shield was rebuilt between 1978 and 1982 and the home port moved to Sekinehama in 1983. It was a shielding-engineering and management failure, not a radiological accident.

None of the four failed on reactor physics: economics and law stopped them. The Savannah needed roughly $2 million a year more in operating subsidy than an equivalent Mariner-class freighter, carried a crew a third larger and specially licensed, and her streamlined hull made loading the forward holds slow just as ports were automating; the project was shelved in 1971 with fuel oil at $20 a tonne, and by 1974, at $80, the numbers would have worked. The Otto Hahn was never cleared to transit either the Panama or the Suez Canal, and visited 33 ports in 22 countries, most of them once and under special permit; her reactor was shut down in 1979 and she was converted to diesel. Behind those refusals sits a legal gap that is still open: the Brussels Convention on the Liability of Operators of Nuclear Ships, adopted on 25 May 1962, never entered into force, largely because it also covered warships and the states with nuclear military fleets did not ratify it. With no civil liability regime in effect, every port call is negotiated one port at a time.

What has worked steadily is the niche where the diesel alternative is poor: ice. Russia has run nuclear icebreakers since the Lenin in 1959, and is now building the Project 22220 series with two RITM-200 reactors of 175 MWt each delivering 60 MW to three propellers, breaking ice 2.8 metres thick, with a 40-year design life and refuelling every six to seven years. That same lineage produced the Akademik Lomonosov floating plant, with two KLT-40S units on a 21,500-tonne barge, and the optimised floating units with RITM-200M reactors that refuel every twelve years across a sixty-year life. Today's interest in civilian marine propulsion comes from shipping decarbonisation and rests on those compact, low-enriched reactors: the transatlantic corridor study launched by Lloyd's Register, Maersk and the ports of Felixstowe and Charleston is not trying to solve the reactor, which is solved, but precisely the two problems that sank the Savannah and the Otto Hahn, port access and the insurance regime.