Blog

Traveling Wave Reactors: Wave Goodbye

By Kennedy Maize

Washington, March 25, 2010 — Hype in the energy world has long history, going back to many generations of perpetual motion machines and the like (cold fusion for example). Nuclear hype is one of the most presistent forms, from electricity “too cheap to meter,” to atomic-powered bombers, to cars with nuclear-powered engines, to the use of hydrogen bombs to dig canals, and so forth.
The latest nuclear hype, driven by some largely incoherent press reports from Japan, is the “travelling wave reactor,” (TWR) which allegedly is going to revolutionize nuclear electric generation. These reactors will be small, modular, fueled by nuclear waste, and, of course, cheaper than anything we have seen today. Yipee!
Now, about that bridge I would like to sell you in Brooklyn.
In reality, the TWR is a hoary variant of liquid metal fast breeder reactor concepts, resurrected by a technology firm with ties to Bill Gates. There’s really not much new here. But the Gates name gives it unwarranted buzz.
The traveling wave reactor dates back to 1958, and the days of the swooning love for anything atomic (the Shippingport, Pa., reactor started generating commercial electric power in 1957 and U.S. nuclear submarines, led by the Nautilus, were cruising under the seas). The idea  of traveling reactions is intriguing, but the hype is typical and the obstacles appear overwhelming.
The concept, first formulated by physicist Saveli Feinberg, envisions a nuclear reactor that uses “fertile,” meaning unenriched U238 which is capable of taking in neutrons, with a smear of highly-enriched U235 on the end, enough to generate a critical mass. The fissioning of the U235 would generate neutrons, making a walking wave through the uranium core, producing fissile plutonium, which would then fission, resulting in heat and neutrons to sustain the chain reaction, a process that could take decades before it played itself out. In theory.
Atomic physicists have played with the concept for more than 40 years, but have never gotten beyond what my friend and nuclear engineer Bob Pollard called “paper reactors.” They always work perfectly. Lately, a spinoff (of a Microsoft spinoff) called TerraPower has caused a bit of a splash, including a rather breathless article in the Wall Street Journal, touting talks between the U.S. firm and Toshiba, a Japanese energy giant.
TerraPower is a unit of Intellectual Visions, based in Everett, Wash., formed by former Microsoft intellectual guru Nathan Myrhvold and bankrolled by Microsoft founder Bill Gates. Press accounts said TerraPower is “in discussions”with Japan’s Toshiba Corp. (owner of Westinghouse, a conventional nuclear power plant company), over developing the traveling wave reactor.
Japan has long been a big-time booster of breeder reactors. Its Monju plant, the exemplar of the technology, caught fire in 1995, as a result of leakage of liquid sodium coolant, and has not operated since. It strikes me that “in discussions” is an entirely meaningless phrase, connoting far less than meets the eye.
The theoretical TWR reactor, as pitched by its U.S. salesfolk at TerraPower, offers a number of hypothetical advantages over conventional light-water reactors. The TWR machine is smaller than the conventional LWR dinosaurs, coming in at between 100-300 MW per unit. That provides the possibility to add units as demand develops (that’s called “modularity,” one of the chief advantages of gas-fired technology); the TWR theoretically can burn depleted uranium, a largely benign nuclear waste byproduct of uranium enrichment (but not spent light-water reactor fuel, which is the real waste problem); it would require less enriched uranium than conventional nuclear reactors.
Problems? Plenty. First, the traveling wave reactor doesn’t exist anywhere. It’s entirely hypothetical. That means it will require years, maybe decades, to see it if works, requiring scarce Department of Energy (taxpayers) dollars. Nobody in the real world would fund this research at the scale required to get a legitimate test reactor running.
Liquid sodium coolant is another scary problem. It tends to catch fire, and has in prototypes of liquid metal-cooled breeder reactors, such as Japan’s Monju project. The 280-MW plant began construction in 1985, went critical in 1995, closed shortly thereafter as a result of a sodium leak and fire. It has not operated since.
Economics for the technology are completely unknown, and perhaps unknowable. Natural uranium and spent uranium is cheap, but so is enriched uranium, and the arbitrage between them is a crap shoot, given the unknown technology of the traveling wave reactor. The capital costs of the TWR plants intuitively look much higher than conventional nukes.
The Energy Daily, a sister publication of POWER, recently reported that Energy Secretary Steven Chu said new breeder reactor technologies that can use nuclear waste as a fuel “would need decades of research and development.” Chu added that breeder and nuclear fuel reprocessing technologies would mean that “eventually there will be [spent] fuel with no economic value” that needs “final disposition.” In other words, there is no way to avoid the requirement to deal with high-level nuclear waste, regardless of breeders reactors or reprocessing.

CNN has described Mryhvold, former Microsoft chief technology officer who left the company in 2006, as a polymath, with degrees in space physics and geophysics, a PhD in theoretical and mathematical physics, a master’s in mathematical economics, a master photographer, a gourmet cook, and a noted paleontologist. One wag (me) described him as the intellectual equivalent Ron Popeel’s Vegamatic kitchen appliance of the 1970s, which could thinly slice “a firm, unripened tomato.”

Mryhvold’s major contributions to Microsoft, before he left with large bags of cash, were the execrable operating systems of the 1990s, Windows 3.1 and Windows 95. Based on my experience, I can only classify these software endeavors as cybergarbage. Not encouraging when it comes to nuclear power technology.

The TWR has come under intense skepticism by physicists. Check out the Physics Forums web site for a taste of derision. I’ll add my own, as a journalist and non-physicist: I’m dubious about any advanced nuclear technology that relies on liquid sodium as a coolant. It’s nasty stuff, and it hasn’t yet worked, to the best of my knowledge, anywhere in the world.