Is Safe Nuclear Power Possible?

Fukushima explosion

If the recent events in Japan were a movie, we'd say that the plot was too outlandishly catastrophic to be true—first an earthquake, then a tsunami, then a nuclear accident. Watching footage of entire neighborhoods being shoveled inland by roiling water, and knowing that those buildings and vehicles contained people, was horrifying.

But while the citizens of Japan continue to struggle with their own personal hell, many of us who live elsewhere are already beginning to ask "what if?" Any nation that sits in an active seismic zone, as Japan does, cannot avoid quakes and tsunamis. But what about the damaged reactors? Are we now going to have to abandon nuclear power as an alternative to fossil fuels? And more importantly, what if a safer kind of reactor had been in place when the quake and tsunami struck?

I put that question to Kirk Sorensen, my go-to guy for information about alternative forms of nuclear power and the web host of What, if anything, might have happened if liquid-fluoride thorium reactors, or LFTRs (pronounced "lifters") had been used instead of regular uranium-based light-water reactors?

"Short answer," he explains, "My personal guess is that there would have been no concern at all about them after the quake."

That's partly because LFTRs don't run at high pressure like normal water-cooled plants do, and also because they have more reliable ways of controlling the nuclear inferno that continues to burn inside most reactors after they shut down. "A major problem at Fukushima was that the tsunami knocked out the emergency power system that was supposed to pump water through the plant to keep it cool," Sorensen says. In contrast, safer LFTR designs automatically shut themselves down even if emergency power is lost.

The harsh reactions in the Fukushima Daiichi complex blasted the surrounding water coolant and generated a pressure-cooker of flammable hydrogen gas, which eventually exploded. No such problem with LFTRs, it seems. "They run at low pressures and they have no water in the core itself, only stable molten salts," Sorenson says, "so there's no risk of radioactive steam explosions."

Also on the list of pluses of these so-called "green nukes" is the relatively benign nature of their wastes, the abundance of thorium worldwide, and the unsuitability of their contents for bomb-making. Nonetheless, no energy source is completely safe—even hydro dams sometimes collapse—and experts who design protective shield walls to house LFTR systems face a serious dilemma. What kinds of accidents should they be built to withstand? To survive an earthquake, for example, flexibility is key. But what about an aircraft collision (think terrorism). In that case, the barrier should be as rigidly resistant to impact as possible.

For an engineer like Sorensen, problems such as this are interesting puzzles begging to be solved. When I asked him what the safest possible design for a reactor might be, he quickly dreamed up a radical-sounding idea. "Small, water-tight LFTRs loosely tethered to the ocean floor would be immune to earthquakes, waves, and air attacks," he said. When I chuckled at the notion of a floating Three Mile Island, he pointed out that these would rest far below the surface, and added that more reactors have been used inside submarines than on land.

"Powerful submersible LFTRs could generate as much electricity as larger, land-based reactors do," he says. "They could be spread along a suitable coastline, with high-voltage power lines running offshore to connect with them. If there had been several LFTRs in submersibles when the Japanese earthquake struck, I suspect they would have survived the tsunami and might be powering the grid right now, helping to get the country back on its feet."

OK, but what's a "suitable coastline?" Wouldn't the next tsunami or typhoon simply smash a LFTR-sub and hurl it ashore, trailing a spaghetti tangle of sizzling cables behind it? I turned to my go-to guy for oceanography, Larry Cahoon at the University of North Carolina-Wilmington, for answers.

His verdict was that shallow-water coasts probably wouldn't work. "I've seen total bottom disturbance in 100 feet of water from a mere category 1 storm, and tsunamis create very strong lateral currents near shore. You'd also have to create a no-traffic zone, as Murphy’s Law for Boats states that two man-made objects on the sea will inevitably collide. But the ultimate problem is that when the danger is greatest you have the least control," he explains. "On land you can get in and out of a plant no matter what; at sea you have no chance once it gets rough. The Navy does run reactors on aircraft carriers and subs. but they send their ships out to sea when a big storm threatens, meaning shore power would be cut. I do like thorium reactors, though. Too bad if this disaster taints all things nuclear."

I don't know how realistic Sorenson's daydreams of a safe nuke future are, though they sound reasonable (at least for deep-water settings) and I've heard similar things from other experts, as well. Of course, even experts can be mistaken. Just this January, the World Nuclear Association updated an online report titled "Nuclear Power Plants and Earthquakes." The first line read "Japanese, and most other, nuclear plants are designed to withstand earthquakes, and in the event of major earth movement, to shut down safely," and later passages described in reassuring detail how Japan's nukes have weathered numerous quakes measuring up to 7.2 on the Richter scale. Two months later, a magnitude 9 quake has now teamed up with a tsunami, and that report doesn't seem so comforting any more.

But risk is a fact of life, and there's much at stake in our increasingly urgent quest for alternative energy sources. Failure to switch to non-fossil fuels during this century could trigger a global super-greenhouse lasting hundreds of thousands of years, and running out of affordable petroleum would be no picnic either. Simply put, I hope that we can still give LFTRs a chance to prove themselves. If the tragedy in Japan makes us toss them aside now, then its aftershocks may echo even more profoundly around the world and far into the deep future.

Curt Stager is an ecologist, paleoclimatologist, and science journalist with a Ph.D. in biology and geology from Duke University. His new book is DEEP FUTURE: The Next 100,000 Years of Life on Earth (St. Martin's Press, March 2011).

Read more coverage of the Japan earthquake.

Add New Comment


  • Andrew Krause

    LFTRs still involve radioactive, toxic metals. Thorium, once disassociateed out of its halide form, is rather explosive in air, if I remember my chemistry right.

    Simple, safe polywell fusor designs would be inexpensive, completely safe (safer than a coal fired plant, in fact), able to be colocated in residential areas, are proliferation proof and can run on cheap, plentiful and aneutronic fuels like boron-11 or lithium. Fusors are so easy to build that high school students have been taking them to science fairs for nearly a decade. The Polywell WB8 prototype developed by EMCC and housed at SpaceDev (yes, that SpaceDev) is only a few engineering steps away from proving the viablity of a modular, 100MW design.

  • Robert Bernal

    Molten salt reactors are proven to be safer back at ORNL. No backup power reqired! If this had happened to a LFTR (liquid fluoride THORIUM reactor), there would not have been the hydrogen explosion (no water needed) and there would have been just a spilling of radioactive salts that can't get any hotter than it already is in its molten state. I'm sure, even though it is not pressurized, a good containment shield would have withstood even a terrorist attack, much less a large earth quake.
    The only drawback is IF water gets in... Seal the dome and keep away from the ocean!
    We should therefore ban ALL LWR's in favor of this most superior design!

  • Tse-Sung Wu

    There's a lot to like about Thorium as a reactor fuel, not the least of which is proliferation. Indeed, according to a article on this topic, US efforts were stopped when it was "too civilian" and not of sufficient military value (i.e., weapons): it is apparently very difficult, expensive and risky to make weapons grade material out of the waste of a Th-fired reactor. Which, in addition, is also very efficient, creating much less waste by mass than our current designs. So you get very little waste from which it is difficult to secretly and safely fabricate weapons-grade material.

    It seems to me the fundamental problem with nuclear reactors is that reactor cores continue to generate heat when all other systems are shut down. Most designs require an active cooling system to keep them safe. So you have a situation where the source of heat is passively reliable (or inexorable, depending on your perspective) and the cooling system is active and requires all manner of power, control, infrastructure etc., as we are seeing in Japan. A natural gas fired plant is the opposite: you need to actively maintain combustion by pumping in natural gas; combustors are cooled by water, but do not create huge risk if that active water-based system is shut down. Air will eventually cool them down.

    Think about it this way: with a nuclear plant, ALL of its fuel is in the reactor. Imagine a coal fired plant in which 30 yrs of coal caught on fire. That's sort of what happens in a malfunctioning nuclear plant.

    The question is: How can we design a nuclear reactor that passively stops generating heat, and is cooled passively, that is, by the environment it is likely to be in during an accident? Gas cooled reactors were thought to be a good design precisely because the rate of heat transfer of He, the gas typically used in such reactors, is closer to that of air, which would likely displace He in the case of an accident and loss of coolant. Water conducts heat 24x as much as air, and so what we're seeing in Japan is the reactor core and spent fuel rods nearby exposed to air, due to significant malfunction of the cooling system. The air can't take away the heat nearly as quickly as water, and so the rods are melting.

    Th seems so promising b/c it's abundant, very efficient, and has low proliferation risk (though I'm not sure about dirty bombs). However, what happens if there is a leakage of the molten fluoride salts in which the Th is suspended or "dissolved"? And what are the risks associated with these salts? What happens when they spill out? Are they corrosive? With what materials are they reactive? What kind of radiation is released when the salts, laden with fuel, go into the environment? What chances to they have at vaporizing? At getting into groundwater? When the reactor gets too hot, the design suggests that the expansion of the molten salts will stop the reaction- hence it's considered self-regulating. Does that mean if the salts are concentrated, say, by a spill or some accident, that the reaction will start up again?

    Here is, for instance, a DOE report on remediating a molten fluoride salt nuclear reactor:

    Let's use this disaster as an opportunity to learn to build a better nuke, if it's possible.

  • Louann Oravec

    NO. The weather and conditions are too unpredictable. Every time man tried to outsmart nature it comes back and bites him on the behind. Plus our presidents have not been too smart, I would not trust some of them to operate a light switch in a dog house.

  • Colin Megson

    Kirk is way off beam thinking offshore siting of LFTRs is likely to be acceptable. We want decentralised 100 MWe units, so securely designed, that they can be sited in and around local communities to save the 10% transmission losses and give us twice the bang for our bucks, by providing district heating, where appropriate.

    In security terms, the underground sited units need to be unmanned, capable of withstanding earthquakes and aircraft impact, with systems which disable or even kill anyone illegitimately penetrating the containment.