A U.S. flag waves near the cooling towers of the Three Mile Island nuclear power plant, where the U.S. suffered its most serious nuclear accident in 1979.
A U.S. flag waves near the Three Mile Island nuclear power plant, where the U.S. suffered its most serious nuclear accident in 1979.
(Jonathan Ernst/Courtesy Reuters)

By now everyone has seen the videos of explosions at Japan’s crippled Fukushima Daiichi nuclear plant and the aerial photos of what looks like the result of a World War II bombing. The Fukushima accident has revived the long-dormant issue of nuclear safety, and these indelible images will no doubt accompany all future debates over nuclear energy. Lately, policy discussions have focused on expanding the role of nuclear power; they must now shift to making sure the existing nuclear plants, and especially the older ones, meet strict contemporary standards.

Of the 13 nuclear reactors along Japan’s coast that were directly impacted by the earthquake and tsunami, it was the four oldest ones that failed completely. The more modern units sustained damage but rode out nature’s onslaught, even if just barely, despite facing forces far greater than what they were designed to withstand. Postmortems will likely disclose that the older plants were designed to lower standards than more recent ones and were not adequately upgraded. Such findings will raise questions about older reactors elsewhere, including in the United States.

The operating Fukushima reactors shut down immediately upon sensing the March 11 earthquake, but radioactive fission products in the reactors’ uranium fuel continued to generate heat after the shutdown and therefore required continuous cooling. But cooling was unavailable as the reactors lost all electric power, including from backup emergency diesel generators. Without adequate cooling, fuel temperatures rose to dangerous levels. The zirconium tubing surrounding the fuel reacted with hot steam to produce hydrogen, which mixed with air and exploded, destroying the surrounding structures.

As the heating continued, at least part of the fuel in the reactors likely melted and released radioactive material, some of which then escaped through the breached protective structures and spread into the surrounding air. The multiple failures of safety systems in the four reactors at Fukushima went beyond any emergency scenario the Tokyo Electric Power Company had ever envisioned, forcing it to improvise solutions, including using fire pumps filled with seawater to cool the reactors.

The disaster at Three Mile Island in 1979 illustrates how quickly reactor fuel can reach melting temperatures if it is not cooled. There, half the reactor core melted in the first two hours of the accident. The melting stopped when an operator turned on emergency cooling, which had mistakenly been turned off. In fact, the active phase of the accident was over before regulators in Washington were informed. (I was the senior commissioner in charge of the Nuclear Regulatory Commission during the first day of the accident.) Altogether, as bad as it seemed at the time, the Three Mile Island accident was a far less serious event than the one at Fukushima. It is significant that we did not learn the degree of melting until several years later, when the reactor vessel’s radioactivity had decayed sufficiently to open it. Similarly, the world will not learn what happened inside the Fukushima reactors for years, assuming things do not get worse.

The storage pools for radioactive spent fuel at Fukushima have also posed a threat. They contain more dangerous radioactive products than do the reactor vessels. One drained pool allowed spent fuel to heat up; similar overheating threatened an adjoining unit, and Japanese officials have been desperately trying -- using helicopters, fire pumps, and even riot police water cannons -- to keep the spent fuel rods under water. It appears that they have finally succeeded in reconnecting a power line to one of the reactors, which will help pump much-needed cooling water.

The reason for the frantic efforts to stabilize the reactors and spent fuel pools is that if the situation were to deteriorate further, molten and vaporized fuel could release dangerous radioactive fission products such as Cesium-137. If the radioactive materials are deposited in sufficient concentrations, they could make surrounding areas essentially uninhabitable. That is what happened around Chernobyl in Ukraine and even in parts of Belarus and Russia that were affected by the fallout. There, restrictions on use of contaminated land were not strictly enforced, but in a Western country or Japan, they would be -- at enormous economic cost. U.S. reactors typically store much more spent fuel than did the Fukushima plants, and a similar disaster here could therefore be devastating.

In the coming months, it will be necessary to conduct a thorough investigation involving both Japanese and international experts in order to help nuclear operators around the world absorb the lessons more effectively.

The United States should use the occasion to reexamine the safety of its operating nuclear reactors, especially the older ones. More than half of the United States’ 104 reactors were licensed more than 30 years ago and constructed under safety codes less demanding than the ones applied to new reactors today. These older plants have been upgraded to some extent but not in any systematic manner. Their seismic analyses, for example, reflect old earthquake data that underestimated the degree of earth motion.

Approximately 20 U.S. “boiling water reactors” -- which boil water in their pressure vessels -- are essentially identical to the failed Fukushima reactors. Most U.S. “pressurized water reactors” -- which do not generate steam in the reactor itself -- have formidable reinforced-concrete containment domes (although typically not as formidable as those in Europe). But nine of these pressurized reactors have smaller and weaker containment structures that rely on buckets of shaved ice to quench steam and thus reduce containment pressure in the event of an accident. Officials in Washington must take a hard look at these nine reactors -- four in North Carolina, three in Tennessee, and two in Michigan.

Unfortunately, U.S. regulators have been overly accommodating to the industry they supervise. The Nuclear Regulatory Commission has been handing out 20-year extensions to plants, whose original licenses were for 40 years. This includes the country’s oldest operating plant, New Jersey’s Oyster Creek, which went into operation in 1969 and now holds a license to operate until 2029. These extensions tend to be granted after NRC reviews that are heavily weighted toward accepting the validity of past technical conclusions. Rather than simply green-lighting old nuclear plants, officials need to reverse the burden of proof and examine more carefully whether past acceptances of old safety systems remain valid today.

The Obama administration’s first reaction to the Fukushima accident was overly defensive. But U.S. President Barack Obama took a step in the right direction on March 17 by asking the NRC to undertake “a comprehensive review of the safety of our domestic nuclear plants in light of the natural disaster that unfolded in Japan.” There should be no illusions, however, about the likely result of this review.

The NRC regulation that covers safety upgrades (called “backfits” in regulatory jargon) is strongly biased against any costly improvements. The last time the NRC launched such a review of the U.S. nuclear plants, in 1996, the industry mobilized then-Senator Pete Domenici (R-N.M.), who was the chairman of the NRC’s appropriation subcommittee. Domenici promptly threatened a stunned NRC Chairperson Shirley Jackson with a deep budget cut unless she reversed her approach and made the agency more industry-friendly. Jackson did exactly that: she fired some of the top officials and toned down the NRC’s criticism of industry. The staff got the message. Domenici brags about this episode in his 2004 book, A Brighter Tomorrow, and goes on to write that he has “been very impressed with the NRC.” So has the nuclear industry, which has always preferred self-regulation to government oversight.

The United States needs firm government safety regulation because public and private incentives differ when it comes to nuclear power. The industry’s economic interests generally dictate safe operation, but sometimes complacency sets in, or a company’s management tries to cut corners. The safety performance of U.S. nuclear plants, as measured by the NRC, has in fact improved over the years. But doubts remain over how the NRC deals with outliers, such as Ohio’s Davis-Besse, which had a close brush with a very serious accident in 2003 shortly after having received top grades from the NRC in all 18 safety categories shortly before.

The Fukushima accident, especially because it happened in a wealthy industrialized country and involved plants of U.S. design, is a warning that should be heeded. Washington needs to reexamine not only U.S. reactors’ safety systems but also the fundamentals of the U.S. approach to nuclear safety. That means looking at how the NRC balances the costs and benefits in deciding whether to require safety upgrades, and whether some of the most vulnerable plants should be operating at all.

The NRC has a highly skilled staff for dealing with day-to-day technical issues, but the agency has trouble confronting larger safety issues, especially those that threaten the prospects of commercial nuclear power or raise questions about its previous license approvals. The five appointed NRC commissioners currently monopolize all authority -- both state and federal -- over nuclear power plant safety. A state government cannot reject a nuclear facility within its borders on grounds of safety; only the NRC can do so.

In the wake of the Japanese nuclear crisis, it is time for Washington to reconsider this arrangement and allow states to decide for themselves.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now
  • VICTOR GILINSKY is a physicist and an energy consultant. He served on the Nuclear Regulatory Commission from 1975 to 1979 and was the senior commissioner in charge during the first day of the Three Mile Island accident in 1979.
  • More By Victor Gilinsky