May 1, 2003
By MATTHEW L. WALD
 
ADSWORTH, Tex., April 30 — Reactor experts around the country hope that there is something unique about Reactor No. 1 at the South Texas Project here. If not, the little crust of white powder that technicians found at the bottom of the reactor vessel, a discovery that has brought operations here to a halt for the indefinite future, could be the beginning of a broad problem for the nuclear power industry.
    The powder, which managers here repeatedly compare in volume to about half an aspirin tablet, is boric acid, which is used in reactor cooling water to soak up excess neutrons, and its presence under the vessel presumably means there is a leak.
    Highly corrosive when damp, boric acid has been found in the last few years on the lids of reactor vessels around the world. A plant near Toledo, Ohio, accumulated 900 pounds, some of which ate away a football-size chunk of steel in the vessel lid, leaving only a thin stainless-steel liner and bringing the plant uncomfortably close to accident.
    But until the discovery here, on April 12, nobody had ever seen a leak on the bottom. A leak in that location is far harder to repair, and would be harder to control if a significant hole developed in the vessel, although the chances of accident seem far smaller than they did in Ohio.
    "It is something different," said Gary Parkey, vice president of the South Texas Project Nuclear Operating Company, which runs the two reactors here.
    Measuring the problem and then resolving it will take new applications of technology, he said, adding with no evident pleasure, "We are at the cutting edge of this issue."
    Until the discovery, in an inspection during a routine shutdown for maintenance, the Nuclear Regulatory Commission believed that it understood the mechanism for leaks in reactor vessels. It assumed that such leaks were caused by an occurrence called stress corrosion cracking, which, after long periods of operation, develops in hard metals that are under strain and high temperature.
    Not long ago, the commission developed a formula combining temperature and years of operation, and used it to tell reactor operators around the country whether they needed to shut down promptly for inspection or could do the job at a more convenient time.
    But the South Texas Project, here amid beef cattle and wildflowers 90 miles southwest of Houston, is only 15 years old, and its reactors operate at a relatively low temperature. That raises the possibility that there may be a problem even with plants that scored well in the regulatory commission's formula, and have not been inspected.
    "If this turns out to be stress corrosion cracking, and there's nothing unique about it, then it raises questions about the validity of that equation," said Brian W. Sheron, the commission's associate director for project licensing and technical analysis.
    That would be bad news for the nation's 102 other commercial power reactors, which despite vast electricity deregulation have prospered in the last few years, by achieving new levels of reliability.
    The South Texas Project boasted last year that its Reactor No. 1 generated more electricity than any other in the nation in 2001, and ranked eighth among the 433 power reactors worldwide. A majority of the plant is owned by two commercial companies: Reliant Energy and an American Electric Power subsidiary, AEP Central Power and Light; an additional 44 percent is owned by the municipal utilities of Austin and San Antonio.
    The plant's operators underscore that they caught the problem early.
    "There was no puddle, no buildup of boric acid on the bottom," said the general manager, Ed Halpin.
    David Lochbaum, a nuclear safety engineer at the Union of Concerned Scientists, a group often sharply critical of nuclear operators, also pointed out that the leak had been discovered early, but in an inspection, he said, that was more thorough than the regulatory commission requires.
    "It does show the prudence of looking periodically in places you don't expect to have problems," he said, "rather than blindly assuming you're not going to have problems except where you're looking."
    One difficulty at the Ohio plant, Davis-Besse, was that management delayed taking the time to remove thermal insulation around the vessel lid to check for leaks. As a result, corrosion continued unnoticed for years. That corrosion has been a nightmare for the Davis-Besse owners, keeping the plant shut for 14 months so far and probably some months to come.
    At both plants, the leaks occurred in places where the builders had installed "penetrations" of the vessel. At Davis-Besse, those penetrations allowed control rods to enter the core (to shut the reactor) and to be withdrawn (to start it up).
    Here, the penetrations are smaller, for dozens of pencil-size detectors that are pushed up into the core to measure the flow of neutrons, the subatomic particles that sustain the chain reaction. The leak apparently involves two of these penetrations.
    The safety implications of the problem seem manageable. Even if the reactor were to have spit out a tube through which a neutron detector enters, the hole would have been only 1.5 inches in diameter, well within the capacity of emergency pumps to keep up with. And though water in the reactor was kept at a pressure of 2,250 pounds, it is not clear that the reactor had been anywhere near to ejecting a tube by the time the boric acid was discovered little more than two weeks ago.
    Still, while managers here do not yet know the cause of the leak, or precisely how they will repair it, experts say the job will be complicated and will involve significant radiation hazards.
    The reason is that after months or years of sitting in the core, neutron detectors are intensely radioactive. If the fuel of Reactor No. 1 is removed, as operators expect for this repair, then the detectors will most likely rest at the bottom of the vessel, since they are not designed to be pushed into the middle of the reactor unless the fuel, and its associated hardware, are present. Mr. Halpin, the general manager, said engineers were working on a way to leave the probes in the middle of the vessel even with no fuel present. But if they rest at the bottom, then a technician working to repair the leak there could, in just a few minutes, receive from them as large a dose of radiation as the industry allows a worker in a year.
    Plant managers say they presume that the leaking parts are a pair of welds, each connecting one tube to the stainless-steel liner inside the vessel. But that place is too hard to reach for repairs, so they plan a weld on the lower surface, where the tubes exit the vessel, six inches outside the liner.
    The problem is that this would permanently leave cooling water with the corrosive boric acid in contact with the vessel wall, which is not stainless but instead plain old carbon-steel, which can corrode. Not only would the corrosion itself be a problem, but particles from the resulting rust could do damage as they floated around the reactor.
    The managers, however, argue that the volume of exposed carbon-steel would be very small, and that the temperature during operations is too high to allow corrosion anyway, a point on which they will have to satisfy the Nuclear Regulatory Commission before they can proceed. Members of the commission's staff will discuss the problem with the managers at a public meeting in Washington on Thursday.
    In the meantime, representatives from two associations of reactor operators, and from five nuclear plants, have visited here and are watching carefully.