And that's why GE has decided that the solution is to keep such reactors small, to minimize safety concerns as well as the size of systems, among other design changes. For example, the piping in the PRISM that carries the liquid sodium coolant around the reactor has two layers. As soon as the inner pipe that actually carries the sodium springs a leak, sensors in the outer layer shut down its flow. "We have learned from the past," Loewen says.
In fact, the PRISM is based on a government research project design that successfully operated in Idaho for decades, under the name the Experimental Breeder Reactor II, or EBR-II. And, although sodium does not play well with air or water, it gets along with metals. "At EBR-II, you could still see [engineering] chalk marks inside the [reactor] vessel when it was drained," Loewen says. That suggests an impressive degree of stability, given that such chalk marks are rapidly eroded in the hundreds of light-water reactors in operation today.
That compatibility with metals is also why GE has chosen to make an alternative nuclear fuel as well, as part of its "Advanced Recycling Center" concept that includes PRISM reactors. Rather than the pellets of uranium oxide used in other fast reactors and conventional reactors as fuel, GE would fabricate metal alloy fuels, with the plutonium or uranium mixed with zirconium metal. That might also allow GE's metal fuels to incorporate the full spectrum of radioactive elements in spent nuclear fuel. "From the experience of making just one oxide, it's tough," Loewen argues. "Add in all the other elements and it's a science project." Such "mixed oxide" fuel has not proved popular in the broader nuclear reactor fleet, although France continues to pursue it, with the U.S. soon to follow.
The challenge is that the metal fuel gets hot—and unlike oxide-based fuels, when it heats, it swells. If the fuel expands too much, it can crack the surrounding cladding, and that presents a big problem. GE's solution is to put in less fuel: "Let's not put in 100 percent of the volume, let's put in 75 percent," Loewen says.
Metal's ability to transfer heat more efficiently means that PRISM's ultimate heat sink—where the 500-degree C heat from the liquid sodium gets dumped—is air rather than water. Natural circulation in the reactor alone is enough to remove all the heat generated by the radioactive decay of the elements in the reactor fuel. "You don't need any human action," Loewen notes. "You don't need valves to open or any automatic safety systems. That's the most significant safety feature."
Fit for purpose
Of course, there is a simpler solution to the U.K.'s plutonium problem: bury it. The PRISM proposal, however, would transmute the plutonium before burying it, as an additional level of security. "We're going to take plutonium oxide that's a powder, turn it into fuel form, put it in the reactor, make it more radioactive, and then put that into the ground," Loewen admits, which would also render it unfit for nuclear weapons. "That's what the customer is asking for."