TVA tritium production could increase risk of terrorist attacks

Editor's note: This is the second story in a three-part series.

By Kathy Helms-Hughes

   Last Tuesday in his State of the Union Address, President George W. Bush warned of ongoing terrorist plans to attack nuclear reactors in the United States, citing diagrams found in Afghanistan of American nuclear power plants.
   It wasn't the first time since Sept. 11 that Americans had been warned of such an attack. Late last year, Tennessee Valley Authority, among others, were put on heightened alert status following similar warnings.
   On Friday, Sen. Harry Reid, D-Nev., chairman of a nuclear regulation subcommittee, said he plans to convene a classified Senate hearing on nuclear plant safety following numerous reports that they might become targets.
   Is the threat real? Some congressional members believe so and are urging passage of the Nuclear Security Act of 2001, which would give the Nuclear Regulatory Commission 180 days to prepare an assessment of the design, operation, and physical protection of sensitive nuclear facilities. Implementation of security plans would follow within 18 months.
   Kenneth Bergeron, a physicist who managed research on nuclear reactor safety and tritium production for nearly 25 years at Sandia National Laboratories, spent many years studying reactor accidents, primarily for the NRC. He has several reservations, especially in light of the terrorist threat, about a Department of Energy plan to use TVA's Watts Bar and Sequoyah nuclear plants for the production of tritium.
   Tritium, a radioactive gas used to boost the firepower of nuclear weapons, is needed to maintain the U.S. nuclear weapons stockpile. The United States and Russia currently are in negotiations to reduce the stockpile to no more than 1,700 to 2,200 nuclear warheads on each side within 10 years, thereby delaying the need for tritium. It has not been produced in the United States since 1988, when the government shut down its last weapons reactor at Savannah River Site in Aiken, S.C., following several radiation leaks and procedural error.
   According to Bergeron, who filed comments in January with the NRC voicing his objections to using TVA's reactors for tritium production:
   * Watts Bar and Sequoyah would be bumped up the ladder in significance, from civilian targets to military targets for terrorists;
   * They are not particularly well protected;
   * The two plants at Sequoyah and one at Watts Bar are three of only 10 ice condenser plants in the country, all manufactured by Westinghouse. Bergeron believes the "ice condensers," which are actually gigantic baskets of ice chips suspended to absorb steam and heat in case of a nuclear reactor accident, would leave the plants exceptionally vulnerable.
   "I was manager of a group that produced computer programs that simulated reactor accidents. You can't blow up a reactor, that's not allowed," Bergeron said. So his study group conducted smaller, "pretty exciting, pretty violent" experiments.
   "You would use computer simulation codes and translate those experiments to real reactors, with all of the bells and whistles, to get some insight as to what would happen if something really bad went wrong.
   "I was one of the people that wrote probably the last study that was done for the NRC on ice condensers," Bergeron said. That study was published in April 2000, a year after he retired from Sandia.
   Bergeron said one thing that came out of research following the March 28, 1979, reactor accident at Three Mile Island which nearly resulted in a nuclear meltdown, was that among the 103 reactors across the country, "there's a huge difference in their performance under what are called core-melt accident conditions. If people are going to die, that's how it's going to happen," he said.
   The research committee focused on risk analysis and found "an enormous difference between the best performers and the worst performers," he said.
   "It's like if you did a safety analysis on a car and you only worried about a 5 mph bumper accident and never worried about things like roll-overs. You'd have to find a big difference," Bergeron said. "And this category of ice condensers (at Watts Bar and Sequoyah) is in the bottom of the barrel."
   The reason they are such poor performers, he said, has to do with hydrogen gas, which is produced as the core melts down, just as it did at Three Mile Island.
   "But Three Mile Island, that was a very strong containment," Bergeron said. "It's what's called a large-dry containment. Ice condensers are weak (in comparison), and that's their problem. They depend on controlling accidents by big baskets of ice. Ice makes things worse; it doesn't make things better. What the ice does is it sucks the steam out of this mixture of air, hydrogen and steam that might be coming out of the reactor, and it makes it more concentrated hydrogen."
   A large-dry containment such as Three Mile Island is basically a huge steel-lined concrete dome. "Kind of brute force -- nothing fancy about it. But it's the kind that performs very well," Bergeron said.
   Using ice to absorb steam allowed TVA to construct a smaller containment, he said. "At Watts Bar, you can hardly see the containment building. You go to another plant that's got a large-dry, and it's one of the biggest things on the site. But you save a lot of money by having a smaller containment building. The walls aren't as thick."
   Rather than ice making up for lack of strength in the building, "just the opposite happens. The ice makes the hydrogen worse," Bergeron said. "I'm not arguing that the ice condensers should be shut down. I'm arguing that it's the last place you want to try to experiment to produce something other than electricity."
   TVA's unfinished Bellefonte Nuclear Plant, on the other hand, which first was offered by TVA for tritium production, has the same type containment as Three Mile Island.
   In the late 1980s, it was debated whether to shut down the ice condenser plants and require the utilities to make significant safety changes, Bergeron said. "Everybody kind of knew that these ice condensers were among the worst, but the NRC kind of said to itself: 'Is it really worth it to have a war on our hands ...?' So they kind of said, 'Well, let's just keep on keeping on ... and try not to have a core-melt accident.'
   "You have to look at accidents like Three Mile Island not in terms of the number of people that got hurt, but in how close you got to a really big event -- and they got very close," Bergeron said.
   "The big problem in my mind (with ice condensers) is that they are very vulnerable to accident. If there is a core-melt accident, then there's a very high probability that instead of containing it inside the containment building, the radioactivity gets out."
   Since his retirement from Sandia, Bergeron has just been a member of the public. But, he said, "I was so offended by this business to use ice condensers that I thought it was time to use my expertise to do something that is in the public interest."
   He fired off comments to the NRC berating the Watts Bar/Sequoyah decision. "I've been kind of part of the establishment for all these years, but I just think this is a terrible decision. It says to me that people who are making decisions like this are not being responsible.
   "I'm not anti-nuke," Bergeron said. "I'm just anti-this decision. They try to paint everybody that's against them with the same brush. They don't have to have the argument if they kind of dismiss you as a critic."
   In July 1994, members of Earth First!, considered a radical environmental group, demonstrated at Watts Bar, resulting in more than $480,000 in sabotage and down time at that facility. But it's not the environmental terrorists that concern Bergeron; it's terrorists such as al-Qaida, who are able to kill thousands of people in one well-calculated move.
   "Everybody's worried, and nobody knows quite what to do about it. When it comes to Watts Bar and Sequoyah today, here's three reactors out of 103. If terrorists decide to attack a nuclear power plant, they've got about a 3 chance out of 100 that it's going to be them.
   "But when you turn those plants into very important parts of the United States' nuclear weapons program, if you were a terrorist who hated our country and resented our position on nuclear weapons ... wouldn't that bump up in your mind the symbolic value of those plants to attack? I think it stops being just 3 plants out of 103 and starts becoming the most vulnerable facilities in the nuclear weapons complex," he said.
   Another issue is the facilities are not particularly well-protected, according to Bergeron. "It's nothing like the weapons complex. They've got specialists that spend all of their time thinking about what kind of a threat, how many people might be involved in an attack, and what would happen after this/after that.
   "Compared to safeguards at a real weapons complex facility, Watts Bar and Sequoyah are very low. They're industrial level. This is one of my arguments (to the NRC). I frankly think they should build a new facility inside one of the weapons complex facilities. And they've got the time to do it. There's no rush," he said.
   Ann Harris, a safety advocate who formerly worked for TVA, said, "The security of Watts Bar and Sequoyah are jokes. They cannot make them secure without shutting down the river traffic because they're on the main channel. There's no protection. There are anywhere from eight to 14 barges being pushed up and down the river all the time. You don't know what's on them."
   Lisa Cutler, of the National Nuclear Security Administration, said DOE and its contractors "have taken a fresh look at our security to try and make sure things are as secure as possible. ... But we don't generally talk about our security arrangements, publicly anyway. We'd have to do them all over again."
   Bergeron also contends production of tritium at TVA reactors would further erode the "safety margin" at those plants. Safety margin, translated, means "here's where you have to be in order to be safe -- and then we go this much farther," he said.
   "Any changes that make a plant a little bit less safe is called 'erosion of safety margin,' and when you're trying to manage safety, you look very carefully at how much you're eroding safety margin.
   "It's like looking at the tread on your tire. At a certain point you've got plenty of rubber. After awhile it gets less and less and you start thinking about changing the tire."
   One way the safety margin would be eroded in tritium production deals with the spent fuel pool, he said. "If you're going to have a hotter fuel pool and it's going to be closer to boiling, if someone stops the cooling process, it's going to take less time before it starts boiling. Then if it continues boiling, it's going to take less time for the water level to drop, and as soon as the water level starts exposing the spent fuel, then you're starting to get a radioactive release."
   Another erosion would be a higher rate of release of tritium into the river. "Every plant releases a certain amount of tritium. But tritium is really, really tricky stuff. Water made out of tritium behaves almost exactly like water. Trying to keep it from escaping is technically very difficult to do. There's a certain amount that's going into the river right now that's monitored very carefully by everybody," he said.
   In the early 1990s, large tritium releases at DOE's Savannah River Site forced people living downstream to have to switch water supplies, Bergeron said. The toxicity of tritiated water is many times greater than exposure to tritium gas, according to a radiological assessment prepared for the Centers for Disease Control.
   "The question is, are you changing the safety picture for people who live around that reactor? My answer is, yes, you are," Bergeron said.
   Another bone of contention is the need for tritium in light of the Strategic Arms Reduction Treaty (START), which mandated the U.S. maintain around 6,000 warheads. Under a subsequent treaty, START II, that number was reduced to 3,500, further delaying the need to create a new tritium supply.
   "We don't need tritium. We won't need it for at least 15 years," Bergeron said. "DOE's own policy, up until about 1995-96, was that we wouldn't need tritium until 2011, and someone just decided to change that."
   It is not projected that a new supply will be needed by 2005. But, Bergeron said, "Bush is talking about reducing the nuclear arsenal by even more -- down to 1,700 from 2,200. That would extend the time period by another 12 years.
   "When you need tritium depends on exactly when your arms control cuts are going to occur, because you use the tritium that comes out of your surplus weapons to feed the needs of the ones that aren't surplus," he said. "I think they are misleading the public about the real need for tritium."