Tritium on Ice: The Dangerous New Alliance of Nuclear Weapons and Nuclear Power (review)

In Tritium on Ice, Kenneth Bergeron takes on three difficult tasks. He provides a factual account of the Department of Energy's (DOE) search for a method to produce tritium, the essential ingredient for "boosting" plutonium nuclear weapons. He offers a scathing critique of the decision, made by Secretary of Energy Bill Richardson in 1998, to achieve tritium production not in a weapons complex facility but in one or more commercial power reactors operated by the Tennessee Valley Authority (TVA). And he sets out to explain the role of tritium in nuclear detonations, the types of reactors that have been proposed for tritium use, and the methods of evaluating reactor safety. The three tasks require different methods: analysis of bureaucratic process, exposé, and explanation of technological processes in a manner comprehensible to a general reader.

Bergeron succeeds admirably on each count. He traces obscure decision-making processes, conveys an impassioned warning about the dangers to national security posed by ill-conceived policies, and imparts a concise understanding of nuclear weapons and reactor technology.

The thesis of Tritium on Ice is that there is an "ebb and flow regarding U.S. attention to the dangers inherent in nuclear technology. Intense vigilance is followed by a slow slide into complacency" (p. 23). Bergeron argues that, since the end of the cold war, the slide into complacency has led to very dangerous policies, and he brings to light disturbing developments that have tended to escape popular attention. Tritium, the radioactive isotope of hydrogen essential to modern nuclear weapons, has not been produced in the United States since 1988. But with arms reductions and "mining" the isotope from decommissioned weapons, there is sufficient tritium to last until well after 2020, by Bergeron's estimate. During the 1980s, before the cold war ended and nuclear disarmament began, the Department of Energy adopted a rational process for deciding which of several competing technologies should be instituted to manufacture tritium at DOE sites, where the technology would be part of the weapons complex and protected by normal security procedures. During the Clinton administration, however, Secretary Richardson disregarded the carefully constructed analysis of technological options and opted for the cheaper and more expedient path of producing tritium in commercial TVA reactors, to begin by 2006.

The trouble with this decision, as Bergeron demonstrates, is manifold. First, there is a longstanding precedent in the United States, dating back to Eisenhower's time, that separates nuclear weapons work from commercial nuclear power facilities. The TVA decision will dramatically and dangerously violate that precedent. Moreover, under the nonproliferation agreements that prevent new nations from joining the nuclear club, there are many provisions separating commercial reactors from weapons applications. In ignoring such provisions, the United States will make it easier for potential proliferators to argue for using commercial reactors to produce strategic nuclear materials.

Another difficulty concerns the poor safety record and management of TVA compared to other utilities using nuclear power. TVA's treatment of whistle-blowers, for example, illustrates its disregard for safety, for quality assurance, and for good management practices in general. Furthermore, as a commercial utility, TVA has no experience in dealing with classified nuclear technology. It is almost inevitable that the classified aspects of tritium production will leak to potential proliferators abroad. Still another risk is that TVA will not be able to protect against another kind of leak: that of dangerously radioactive tritium into the environment. In addition, the reactors chosen for tritium production, with an ice-basket arrangement for quenching possible heat excursions, have the weakest containment structures of any commercial reactors. Bergeron explains the technical reasons for this weakness and the technopolitical reasons for the choice of the design. The irony is that none of these risks need be incurred, because tritium will not be in short...