Would a New Class of Nuclear Reactors Have Withstood the Tsunami?

After the Japanese military spent much of last week struggling to cool spent rods at the damaged Fukushima Daiichi Nuclear Power Station, resorting to spraying them with fire hoses in some cases, the situation seems to have calmed recently as power was restored to at least two of the plant’s reactors. Though just as Prime Minister Naoto Kan stated that he could see “a light at the end of the tunnel” of the disaster, workers were evacuated as plumes of smoke poured from two of the damaged reactors.

The folks over at IV Insights, the blog associated with Nathan Myhrvold‘s Intellectual Ventures, point out that it was the complete loss of power that disabled the cooling systems protecting the plant’s reactors. Which raises the question: Is there nuclear technology that could withstand such a catastrophe? Possibly. TerraPower, an Intellectual Ventures spin-off that also boasts Bill Gates as an investor, is working on a new reactor design called a traveling wave reactor that uses fast reactor technology, rather than the light water technology used at the Fukushima Daiichi plant.

The two biggest advantages of the fast reactor design is that it requires no spent fuel pools and uses cooling systems that require no power to function, meaning the loss of power from the tsunami might not have crippled a fast reactor plant so severely. Read more here.


There are also smaller reactors with no moving parts. These would be buried in neighborhoods and essentially forgotten for ten years, when they'd need to be refueled. They're not large, they can't overheat and the material inside can't be weaponized. Problem is no one has figured out a way to make money off them.

But as to conventional reactors, there's always a combination of events that will take them down. That combination may be rare, kind of like the fifth largest earthquake ever recorded followed by a tsunami quite a bit higher than the protective barrier around the plant, but there will always be a black swan that can take them down. In the case of the Japanese reactors, had the backup generators and fuel supplies been up high, like the spent fuel ponds, the reactors wouldn't have had so much trouble. But that's hindsight.


Any idea why spent fuel rods were/are stored on site? In the US it's because license review for the long term storage site at Yucca Mountain has been halted. Has there been similar foot-dragging in Japan?

Jim Bob

Every reactor has the same difficulties that the Japan reactors met up with.
A shut-down reactor still has a great deal of thermal inertia and need for continuing cooling, and the fission products continue to generate a heat load for many days after shut down.

Loss of power to run secondary cooling systems machinery, such as intake water pumps, cooling towers, steam condensers and the like would still be necessary. It is not enough to have circulation around the reactor core that does not require power.

Dale Sheldon-Hess

The passive cooling systems on modern reactors, such as the Westinghouse AP1000, can supply sufficient cooling to handle decay heat for the first 72 hours after shutdown, with zero power.

Obviously, it takes more than 3 days for the reactor to cool entirely, and it took more than 3 days to reconnect external power at Fukishima, but most of the damage was done in the first 3 days, so it absolutely would have reduced the damage (and remember: so far, no one has died, and only the workers on site have taken any appreciable amount of radiation.)

(Disclosure: I used to be a (non-safety system) software engineer for Westinghouse nuclear.)


What's so new about liquid metal fast reactors. They have been around from the beginning.


This blog post certainly raises the question: does anyone know what begging the question means?


I find the idea using Thorium as a fuel source much more interesting. I highly recommend looking up both Thorium and Liquid Fluoride Thorium Reactor via your search engine of choice.

The website "Energy from Thorium" is particularly enlightening.


Modern reactors having cooling systems designed such that the coolant will circulate without power. It would not be able to cool a running reactor, but would be able to shut down the reactor.

Numerous other designs exist with claims to be safer. Guess what? Technology advances in the 20-30 years since most of the world's reactors were built.

Pekka Taipale

By the way, there were a number of reactors, implemented with existing technology years ago, that did withstand this tsunami in Japan - Onagawa and Tokai plants. To be more precise, they did take some damage, but the backup power mechanisms were able to handle the damage caused by tsunami.

Somehow the whole discussion has run out of hand. We don't have actual fatalities due to nuclear problems (although plant workers were killed by the direct impact of earthquake). We have certainly 15 000 and likely 20 000 people killed by the tsunami, including 1800 homes washed away by the burst dam of a hydroelectric plant. And most people just talk about the nuclear reactors, while people still have been saved from the rubble of the quake and tsunami.

Martin Uebele

What I miss in the discussion about the technical feasability of building save nuclear stations is the human element in it. Apparently, it was known that these stations were not up to date from a safety point of view, since their back-up generators are not placed in water tight rooms. Still, they were used. Also, it just came out that the owners did not act according to security guidelines before the earthquake, and were officially made to correct that. (See for example www.ibtimes.com/articles/123314/20110316/japan-nuclear-safety-warnings-iaea-wikileaks-earthquake-fukushima.htm). Also, there were massive human failures which allegedly led to the disaster of Tchernobyl, not only flaws of the reactor design. So, even if the technology was theoretically safe, what does it help if people don't use it in a safe way?

And on the allegedly skewed media coverage: what about the expected future fatalities. Just think through the two possible scenarios of either nuclear meltdown or successful cooling of the reactors, multiply their probabilities with the respective numbers of excess deaths over the next twenty years and compare the sum with the people who are endangered by missing food and heating - what's the result?


Bob Meyrowitz

Huh? No person of the public received a dose of radiation that would even make them sick, and certainly not anywhere near a level which has been shown to correlate with an increase in cancer risk (which is approximately an acute dose of 100 mSv). There won't be any 'excess' deaths due to Fukushima, just like there weren't any due to Three Mile Island. The OP's point is that the media is overreacting on issues of safety with nuclear power, when the real risk is tsunamis themselves.

Martin Uebele

The point I want to make is that with nuclear accidents it is far more difficult to say what is going to happen next which stimulates people's imagination and therefore interest. Of course the people starving and freezing are worrysome but it is quite clear what is going to happen to them in the next days and weeks which doesn't feed media's demand for news.

According to Japanes government speaker Yukio Edano radiation levels in "some places more than 30 km away from the plant may have been higher than 100 mSv today." (Financial Times Germany online, 23 March 11). So if there is at least some reasonable level of probability that the situation may turn much worse than it is in the moment, I don't consider the media coverage to be an overreaction but an adequate reflection of risk.

Andrew Krause

YES! And it's infuriating to those of of who are aware of it, but the IEC Fusor design championed by the late, great Dr. Robert Bussard would have just shut down. The polwell-based design designated the "WB" series is basically an improved version of the farnsworth-hirsch fusor that high school kids have been taking to science fairs for more than a decade now (and which has been around since the 50's). Unlike those devies, the WB series can - no has - achieved stable plasma and sustained energy output on the order of milliseconds - something not yet achieved with takomak designs (IETR) or by even more exotic designs out of MIT or Sandia. (If you're a plasma... a millisecond is as good as year.)

More importantly - these devices, while proven on deuterium-deuterium reactions that produce fast neutron radiation - would be capable of aneutronic fusion using proton-boron reactions. The alpha/beta radiation emitting particles that are left over from p-Br reactions have half-lives measure in hours. (Another plus, alpha and beta radiation can directly drive atomic batteries to produce electricity, eliminating the need to steam-driven turbines. This is how we've powered deep space probes for decades. Voyagers 1 and 2 are still ticking away out there in the void running off their atomic batteries.)

I need the good folks at Intellectual Ventures to get with the good folks at SpaceDev (yes, that spacedev, the guys helping to build Virgin Galactic's fleet of sub-orbital joy rides). The entire body of product from Bussard's research - including the WB-8 reactor model, which I believe produced 10 kev on it's final run in November of 2010 - was transferred there when Energy Matter Conversion Corporation ran out of grant money. I think they'll find that they're $7mn away from WB-8 that can indefinately sustain a plasma, and $200m away from a commercial 100MW reactor plant.

Again, the advantages here are:
1. Proven reactor design - high school kids are building these
2. Proven lineage - research on IEC fusor technology has been going on since the 1930's, so the theoretical framework is in place to support advancement
3. Direct conversion of nuclear energy to electrical energy
4. No cooling system required at all - heat is waste
5. Capable of running with no neutronic radiation, only short-lived alpha and beta radiation
6. Can use just about any fusable fuel all the way down to Boron, which is something we have plenty of.
7. Proliferation proof - cannot produce enriched weapons grade material

Thanks... I'll get off my soapbox now.


Matt Podolsky

"cool spent rods at the damaged Fukushima Daiichi Nuclear Power Station, resorting to spraying them with fire hoses in some cases, the situation seems to have calmed recently as power was restored to at least two of the plant’s four reactors"

Correction: Fukushima Daiichi has six reactors, not four. (Fukushima Dai-ni has four reactors.)


That picture is of the Trojan Nuclear Power Plant's cooling tower. It was demolished in 2006.


There has been a "walk away safe" reactor design that has been around for decades. It is called CANDU. Of course it isn't American and it doesn't have Bill Gates and Microsoft as beneficiaries so it can't be any good.


One of the most frightening aspects of this accident is the lack of passive stability as pointed out in the article. Stability is always a problem with high energy density fuels/sources. Natural gas explodes quite happily, oil burns or spills, coal mines explode, collapse, or burn for millenia, and hydroelectric dams has been known to fail spectacularly. Even laptops have a nasty habit of bursting into flames. But none of these render areas uninhabitable quite the way that a nuclear accident can. Fortunately, there are a number of designs with significantly improved passive stability, which other posters have pointed out. The problem with all these designs is that a cooling leak... say after a massive earthquake... would similarly lead to meltdown (although secondary and tertiary containment is vastly improved in more modern reactors). Although it may exist, I have seen no evidence that the TerraPower reactor would maintain core integrity with only passive solid thermal conductivity (not relying on a cooling fluid) And while there's a lot to say in favor of their design (and I do mean a lot) I'm not comfortable concluding that it solves all the problems of nuclear power!



The TWR does however use liquid sodium as a coolant. If you think that leaking coolant is an issue in a light water reactor, try dealing with a substance that reacts violently on contact with air and water. The experimental fast breeder reactors built in Japan and Russia (which also use sodium coolant) have had major issues in this regard.

I also understand that it is possible to design a LWR that has sufficient passive cooling capacity to survive a failure of the active cooling system as long as it is shut down. This would seem to be a better option.