You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

The Trouble with MOX

Why is the U.S. intent on imitating one of Fukushima’s scariest flaws?

Last August, workers at Japan’s now infamous Fukushima Daiichi nuclear plant loaded the first batch of mixed-oxide fuel, or MOX, into one of their reactors. The event went largely unnoticed in the United States; but in Japan it was deeply controversial. Unlike traditional nuclear fuel, which is pure uranium, MOX is a far more dangerous blend of both uranium and plutonium (the latter is among the most carcinogenic substances on Earth). Dogged by bitter public opposition, Fukushima’s operators had spent a decade fighting for permission to deploy their MOX cache, which had been languishing at the plant ever since arriving by ship from Belgium in 1999. They finally won the battle, but only after a scandal toppled Fukushima’s anti-MOX governor.

In a grim bit of foreshadowing, just after MOX was loaded into Fukushima’s reactor no. 3, an alarm light flickered on, indicating a problem with the emergency core-cooling system. Operators decided it was just a glitch. Of course, no one knew then that Japan would be ravaged by an earthquake and a tsunami, knocking out the plant’s main power supply—or that the back-up cooling would fail, leaving workers scrambling to cool Fukushima’s reactors by any means possible.

Today, several Fukushima reactors are teetering near meltdown—a nerve-wracking prospect made all the more scary by the presence of MOX in one of them. This raises an unsettling question about U.S. policy: Why, after rejecting MOX for decades, is the U.S. government currently in the midst of a massive, $10 billion project to supply this dangerous fuel to nuclear power plants around the country?


Like a handful of other nations, Japan had embraced MOX as a way to reuse the plutonium from spent nuclear fuel (traditional uranium fuel produces some plutonium during fission, though not nearly as much as is contained in MOX). Washington found MOX appealing for a very different reason: It was supposed to solve the vexing riddle of what to do with the vast quantities of weapons-grade plutonium we produced during the cold war. In the ’90s, the Department of Energy (DOE) studied dozens of options—among them lobbing it into space—before finally settling on a hybrid approach. A portion of the plutonium would be mixed with other radioactive waste, embedded in glass, and buried, a process known as immobilization. The rest would be converted into MOX and used as fuel. The idea was that, if one program foundered, due to either technical problems or public protest, there would be another on which to fall back.

Then, in 2000, the United States and Russia signed an agreement committing each country to dispose of 34 metric tons of weapons-grade plutonium—the equivalent of 17,000 warheads. Russia planned to convert all 34 tons into MOX. It pushed the United States to do the same on the grounds that, theoretically, if tensions between the two countries ever flared again, immobilized plutonium could be retrieved and used to make bombs. The plutonium in spent MOX fuel, by contrast, is altered through fission, making it more difficult (though not impossible) to use in existing weapons. “This was supposed to be permanent disposition, meaning this material was never again available for weapons use,” explains Anne Harrington, deputy administrator for defense nuclear nonproliferation at the National Nuclear Security Administration. “The feeling was that immobilization didn’t do that.”

There were other factors working in MOX’s favor, too. One was cost: Funding two separate programs to dispose of plutonium, rather than just focusing on MOX, was expensive. Meanwhile, the French nuclear giant Areva, which manufactures the bulk of the world’s MOX supply, had been contracted (along with the Baton Rouge-based Shaw Group Inc.) to build the U.S. MOX plant in South Carolina. An arm of the French government, Areva has a powerful Washington lobbying operation: Over the last decade, it has spent more than $10 million lobbying American officials. It pushed hard for the MOX-only route and managed to forge deep inroads with the nuclear-energy-friendly Bush administration. In fact, two Areva lobbyists sat on President Bush’s energy-transition-advisory team. In 2002, Bush’s Secretary of Energy Spencer Abraham finally pulled the plug on immobilization in favor of MOX. He has since gone on to chair Areva’s board of directors.

Whatever the reasons for the shift, relying on MOX to dispose of weapons-grade plutonium has some serious downsides. For one, producing and distributing it involves multiple stages of processing and transport. Critics argue this could create opportunities for theft or sabotage by terrorists or rogue states. What’s more, according to nuclear physicists, using MOX in reactors that weren’t designed for that purpose—and only one plant in the United States includes reactors that were—increases the chances of a serious accident, albeit marginally.

Worse, in the event that a MOX-fueled reactor melts down, releasing large quantities of radioactive material, it could be far deadlier than a reactor containing traditional uranium-based fuel. In 1999, the DOE calculated the risk for reactors that were candidates to receive MOX fuel (there is a different slate of candidates now, but the models overlap) and estimated that, in some accident scenarios, a MOX reactor would cause between 3 and 14 percent more cancer deaths than its uranium-fueled counterparts. Edwin Lyman, a nuclear physicist and senior staff scientist at the Union of Concerned Scientists, predicts an even wider gap. Using computer modeling, he found that a core-melt accident involving a pressurized-water reactor loaded with 40 percent MOX fuel—as is the plan in the United States—could cause up to 30 percent more cancer fatalities.


Under our agreement with Russia, we were supposed to start disposing of weapons-grade plutonium at a rate of two tons a year in 2007. But the MOX program has hit numerous stumbling blocks. The cost of building the MOX plant has ballooned from $1 billion to $5 billion, with the total project costs topping $10 billion. Officials now predict it will be 2016 before the plant is finished. After that, it will take at least another two years before it produces the first test fuel—which would still need to be approved by the Nuclear Regulatory Commission, a process that could take yet more years.

Even once these hurdles are cleared, the DOE still has to work out the kinks in the plutonium pipeline. The agency is supposed to be building a second facility to extract the plutonium from the pits inside nuclear warheads and turn it into feedstock for the MOX plant. But, in late 2009, it announced that it was scrapping the plans it had been working from for more than a decade and starting over. While the new plans and timeline have yet to be finalized, an investigation by the Government Accountability Office found the facility was unlikely to be finished in time to supply the MOX plant. (The DOE says it has identified alternate sources of plutonium to bridge the gap for the first several years.)

Meanwhile, the DOE still has to line up utilities to buy the fuel. In 1999, North Carolina-based Duke Energy signed on. But it backed out two years ago after irregularities surfaced during a weapons-grade MOX test in one of its reactors. Since then, the DOE has struggled to find other buyers, even though it is offering to provide MOX for less than the going rate for uranium (a notable fact given that uranium costs roughly a quarter as much to produce). Utilities, it turns out, are wary of taking on the risk. “It’s a small possible benefit for a considerable amount of hassle,” explains Frank von Hippel, a nuclear physicist who codirects the Program on Science and Global Security at Princeton. “MOX has to be guarded like weapons material. There’s the potential public relations woes. Utilities will probably need to have their operating licensing modified, which provides opportunities for interventions by groups that oppose nuclear power.”

Given the lack of interest from private utilities, perhaps it’s not surprising that the DOE would turn to the federally owned Tennessee Valley Authority (TVA). TVA is exploring using MOX in three General Electric Mark I reactors at its Brown’s Ferry site in Alabama, as well as two Westinghouse ice-condenser reactors at its Tennessee-based Sequoyah plant. The problems with the Mark I have made headlines recently because it’s the type in use at Fukushima, but the ice condensers are also deeply flawed. The core issue in both cases is that the containment system—the outer shell, which is supposed to prevent radioactive material from escaping in an accident—is exceptionally thin and susceptible to rupture. As early as 1972, American regulators considered banning both models because they were prone to radioactive leaks. More recently, Sandia National Labs found that the chance of the Mark I’s containment system failing early on during a core meltdown was a staggering 42 percent. “We’re talking about reactors with very poor designs,” Victor Gilinsky, a former Nuclear Regulatory Commissioner who evaluated the licensing application for both plants, told me.

Besides TVA, the DOE has one potential customer: Energy Northwest in Washington state is looking into using MOX in its sole reactor—a General Electric Mark II. Unfortunately, it’s vulnerable to the same containment woes. “Why put this more dangerous fuel in reactors that are prone to containment failure?” asks Lyman of the Union of Concerned Scientists. “It boggles the mind.”


MOX program supporters contend that, in this country, the chances of a severe nuclear accident with a large-scale plutonium or uranium release are extremely remote. They also note that modifications to many U.S. Mark I and ice-condenser reactors have made them less prone to radioactive leaks. “The risks are miniscule,” says Ken Bromberg, the National Nuclear Security Administration’s assistant deputy administrator for fissile materials disposition.

Nevertheless, the grim spectacle in Japan has stirred doubts about U.S. MOX plans. In late March, after plutonium was found in the soil at Fukushima Daiichi, Representative Edward Markey, a Massachusetts Democrat and a leader on energy issues, called for an “immediate review of the safety issues associated with MOX fuel fabrication and use here in the United States, including whether it makes sense to move forward.”

Unfortunately, it’s not clear what other options we have. Over the last decade, while we’ve been pouring resources into MOX, research into other techniques like immobilization has stalled. Switching courses now would mean sinking more money into unproven technologies. That isn’t a particularly appealing option. Then again, as the events at Fukushima show, neither is doubling down on MOX.

Mariah Blake is a writer living in Washington, D.C. Her work has appeared in Mother Jones, Foreign Policy, and The Washington Monthly. This article originally ran in the April 28, 2011, issue of the magazine.

Follow @tnr on Twitter.