Abstract
The Tennessee Valley's Watts Bar nuclear power plant.
In an unprecedented decision, Energy Secretary Bill Richardson has designated three Tennessee Valley Authority (tva) commercial nuclear reactors–Watts Bar Unit 1 and Sequoyah Units 1 and 2–as the “preferred” facilities to produce tritium for U.S. nuclear weapons.
The decision, announced December 22, breaks the longstanding practice of not producing bomb material in civilian reactors. Nonetheless, Richardson said, the “Tva is well suited for this work because meeting national security needs is part of its statutory charter, and its reactors are government facilities.”
Backing his judgment was an extensive 1998 interagency review, which found that “any nonproliferation issues are manageable and should not preclude selection of commercial reactors.”
Although nonproliferation concerns weighed heavily in Richardson's decision, he said cost was a significant factor during the long and contentious debate over competing proposals. Richardson asserted that the commercial light-water reactor option was “the best deal for taxpayers, and it has the flexibility we need to meet our present and future tritium needs. It's the only option that doesn't require a large capital expenditure.”
Tritium gas, an isotope of hydrogen, enhances the yield of nuclear weapons. But it decays at the rate of about five percent per year, making replenishment a continuing concern. The Energy Department formerly produced tritium in reactors at the Savannah River Site in South Carolina, but the last Savannah River reactor was shut down in 1988.
Since that time, tritium taken from dismantled weapons has been purified and recycled back into the current stockpile. But Energy estimates that the United States will need new tritium by 2005, if the stockpile remains at the START I level. Even if the United States reduces its arsenal to a START II level, it will need new sources of tritium by 2011.
The leading options for tritium production were building a linear accelerator, which would produce tritium by bombarding helium 3 gas with neutrons; completing the TVA's unfinished and mothballed Bellefonte reactor; or buying tritium from existing TVA reactors.
The Congressional Budget Office estimated last August that the accelerator option would cost about $9.5 billion in 1999 dollars over a 40-year period. In contrast, the Bellefonte option would cost about $2.5 billion. Using existing reactors would cost about $1.8 billion.
Richardson's decision to use the existing TVA reactors may be a boon to taxpayers, but it's bad for the nation's nonproliferation agenda, critics said. For decades the United States has encouraged other nations to follow its example of not using civilian nuclear plants for military purposes.
Richardson's announcement caused several organizations dedicated to preventing nuclear proliferation to cry foul. Robert Tiller of Physicians for Social Responsibility, said: “Our leaders are upset that Iraq might use civilian facilities to produce components for weapons of mass destruction, but the United States is now committed to do that very thing–turning a civilian reactor into a bomb facility.”
Congressional critics include Cong. Edward Markey, a Democrat from Massachusetts and a long-time non-proliferation advocate, and Cong. Lindsey Graham, a Republican from South Carolina whose district includes the Savannah River Site, the most likely place where an accelerator would have been built. Markey was especially colorful:
“We have breached the 53-year separation between Atoms for Peace and atoms for war. We cannot credibly preach nuclear temperance from a bar stool. If we are mixing our civilian and military nuclear programs in the United States, then the rest of the world will not pay much heed to our admonitions to … refrain from using their civilian nuclear energy programs for military purposes.”
This strange-bedfellows political marriage gave birth last year to the Graham-Markey amendment, which was attached to the 1999 Defense Authorization Bill. That measure underlined the policy of abstaining from military tritium production in civilian nuclear reactors. The amendment passed the House in a voice vote, but died in September.
Congressional observers expect to see a similar amendment this year. And it is likely to garner support from many nonproliferation organizations, including the Nuclear Control Institute. Paul Leventhal, president of the institute, believes that “it would be better from a nonproliferation perspective to prohibit production in civilian reactors except in a national emergency declared by Congress,” which is “now the case with plutonium under the Atomic Energy Act.”
Litigation possible
The Tennessee Valley Authority had encouraged Energy to select the moth-balled Bellefonte plant in Hollywood, Alabama. But the hefty price tag for completing the plant persuaded Energy to buy tritium from existing plants.
The Watts Bar plant, located near Spring City, Tennessee, began commercial operation in 1996 after a more-than-two-decades construction period that was plagued with serious safety and design problems. The plant owns the dubious distinction of being the most expensive–some $8 billion–nuclear plant to be built anywhere. And it has the most whistle-blower allegations of any civilian U.S. nuclear plant.
Whistleblowers are still coming forward. A December 26, 1998, Associated Press report examined the alleged problems with Watts Bar's ice condenser system, a crucial emergency cooling system.
Curtis Overall, an ice condenser specialist, found 171 broken screws while cleaning the plant's system in 1995. Luis Reyes, the Nuclear Regulatory Commission's regional administrator, reports that the NRC is satisfied that the screws do not affect plant safety and that further inspections would be conducted this spring during Watts Bar's planned refueling.
Having targeted Watts Bar for safety violations, anti-nuclear and environmental activists are setting their sights on further protests–and possible litigation. For instance, Ralph Hutchison of the Oak Ridge Environmental Peace Alliance explains that his organization is “committed to doing everything it can to prevent tritium production.”
If Energy proceeds with the tritium plan, Hutchison envisions protest demonstrations and probable legal action. Although Hutchison concedes that “the grounds for a possibly successful NEPA [National Environmental Policy Act] challenge are not apparent,” James Riccio, staff attorney for Public Citizen's Critical Mass Energy Project, counters that “litigation is good for two things: obfuscating and delaying, and now the issue is political, not legal.”
Other TVA reactors also suffer from poor safety histories. Watts Bar's designated backup plants for tritium production, Sequoyah Units 1 and 2, located along the Tennessee River at Soddy Daisy, Tennessee, sit in the back of the nuclear safety class, according to recent independent studies.
Both reactors fell into Public Citizen's “25 worst performers” category in its 1996 “Nuclear Lemons” report, which ranked the nation's reactors. A 1998 study by the Union of Concerned Scientists found that Sequoyah Unit 1 was “the lowest performer of any of the focus group plants.”
Roads not taken
Los Alamos National Laboratory lobbied hard to have its linear accelerator design accepted and built in South Carolina.
When that design was critiqued by the Congressional Budget Office as too expensive, accelerator advocates appealed to many in the nonproliferation community, saying their method would not incur the proliferation problems that would plague a civilian reactor program.
What's the rush?
Energy says the tritium supply will begin to dip below the START I arsenal's needs around 2010. This arsenal will presumably contain some 8,400 deployed warheads. (In addition to deployed warheads, Energy insists on keeping a “hedge stockpile” of several thousand warheads filled with tritium as well as a five-year reserve of tritium.)
Under START II, the United States and Russia are to decrease their deployed strategic warheads to a maximum of 3,500 each. Factoring in 1,000 tactical warheads, the total U.S. stockpile will number 4,500. At this level, Energy could delay tritium production until after 2020, according to an analysis that Frank von Hippel, an arms control expert at Princeton, and I conducted. That analysis assumes the United States needs neither a hedge stockpile nor a five-year reserve of tritium.
Under START III, the total number of U.S. warheads could fall to the 2,500-3,500 range. Given that, von Hippel and I believe that Energy could delay tritium production until after 2025 or 2030. Under longer-term reduction agreements (involving Russia, China, France, and Britain), the U.S. arsenal might sink into the hundreds, delaying the need for tritium to late in the century, perhaps 2075. (In 1997, the National Academy of Sciences said the United States could maintain a credible deterrent with a “few hundred warheads.”)
–C. D. F.
Although research and development continues on the accelerator, Joan Rohlfing, Energy's top official for nonproliferation matters, says the department does not currently plan to build it. Nonetheless, the proposed accelerator has powerful friends, particularly Sen. Pete Domenici, a Republican from New Mexico, who thinks accelerator technology will provide a long-term solution for the production of tritium and medical isotopes. He also believes that it could be used to dispose of radioactive waste through transmutation.
In addition to the civilian reactor and accelerator programs, Energy also considered the Hanford nuclear site's Fast Flux Test Facility (FFTF) as a potential tritium factory. However, Richardson said in his December 22 press conference, “After examining the different options, I have decided that the Fast Flux Test Facility will not play a role in producing tritium, but I will decide about any civilian missions for the facility by spring.”
Some advocates of cost containment ask: If cost reduction analyses drove much of the tritium debate, did Energy select the cheapest option? Richard Garwin, a prominent champion of purchasing tritium from foreign sources, has often estimated that buying tritium from Russia, for example, would cost much less than the civilian reactor or accelerator options.
In response, many American and Russian politicians have suggested that purchasing tritium from Russia would be a politically explosive issue with Russian nationalists and hardline communists, as well as with many American politicians on the right end of the spectrum. Nevertheless, a congressional source recently outlined to me how a purchase deal could possibly unfold.
The overriding concern regarding possible purchase is to ensure a secure supply of tritium. If a foreign seller–possibly Russia or more probably some other tritium vendor–were to cut off the supply, it would take Energy two to three years to start up its own production facility and extract tritium without dipping below the amount required for the arsenal.
The congressional source estimated that it would take two years to gear up a civilian reactor, so he suggests that in the first year the United States could buy enough tritium for three years, which would add another year's supply as a safety margin. In each subsequent year, the United States would need to purchase just enough tritium to meet its current demand. Such a plan, he says, would minimize proliferation problems while buying time for the accelerator design to be perfected.
He also acknowledges that finding a foreign supplier could be difficult. Russia presents obvious political problems. Canada produces an abundance of tritium, but Canadian law prohibits selling tritium for military purposes. France and Britain might be willing to sell tritium, although their estimated production capacity falls short of U.S. START I needs. But they could probably supply U.S. requirements under START II, depending on the size of the “hedge” stockpile and the needs of the British and French arsenals.
As the political fallout of Energy's decision continues, the 1999 Defense Authorization Act restricts spending any money on tritium production. This one-year hiatus provides policy-makers and other parties with time for reflection and serious study of tritium requirements under different scenarios, including START I, II, and III, as well as unilateral reciprocal reductions of nuclear arms.
Advocates of continuing reductions in nuclear arms have long argued that if the United States reduces the size of its arsenal at a steady rate, the actual production of new tritium could be postponed for a very long time, even indefinitely.
In his press conference, Richardson made an oblique reference to that possibility: “If our goal of reaching further arms reduction agreements is reached, we may not need to exercise this [tritium production] option for many years.”