NIRS - In early March 2002, First Energy (FE), true to its name, revealed a policy to drive electricity production ahead of federal safety requirements. This official policy of mismanagement pushed its Davis-Besse nuclear plant to the brink of disaster. Moreover, senior engineers at the United States Nuclear Regulatory Commission (NRC) realized a "high likelihood" that the reactor was so damaged that the risk of a nuclear accident grew greater with continued operation. Yet, federal officials were unable to issue an order to immediately shut the reactor for the necessary inspection and repair. Instead, the agency chose to ignore safety regulations and gamble with disaster to accommodate the financial interest of yet another corporate delinquent.
An NRC bulletin issued in August 2001 called for utilities operating pressurized water reactors (PWR) to inspect for dangerous cracks found in nozzles that penetrate the top of the reactor and house control rod drive mechanisms (CRDM). The NRC bulletin followed the discovery of cracks in Duke Power's Oconee reactor (see WISE News Communique 553.5309, "US: NRC ignores widespread safety flaw for decade"). Operators were instructed to look for "popcorn-like" traces of boron crystals as reactor coolant escaping from nozzle cracks. The bulletin warned that unchecked cracking in nozzles could grow to component failure, a loss-of-coolant-accident and reactor core damage. NRC required all operators to report inspection results to the agency by 31 December 2001.
However, Davis-Besse operators were eager to complete its two-year operating cycle scheduled for a refueling in April 2002. Company management sought a waiver from the inspection and an extension to the reporting deadline to meet their outage date. NRC engineers reviewing the Davis-Besse response were concerned that six of the seven other Babcock & Wilcox's (B&W) reactors like Davis-Besse had already found cracking. Only Davis-Besse had not fully inspected. A rare NRC "Order" to shut Davis-Besse for inspection by December 31, 2001 was finalized in November but never issued. In its place, a senior NRC officer reached a private compromise with the company to extend reactor operation to February 16, 2002.
When FE did inspect, workers found not only cracks but one of the 69 nozzles had come loose from the vessel head as a result of extensive corrosion around its base (see WISE/NIRS Nuclear Monitor 565.5385, "Millimeters from disaster").
Revelations into the near-miss accident at Davis-Besse go far beyond the reactor site perched on the shores of the Great Lakes near Toledo, Ohio. They warn of a dangerous gambit being played by atomic corporations in an increasingly competitive electricity market where public safety is sacrificed to ambitious production schedules. These revelations show that the NRC is willing to turn a blind eye on safety regulations to accommodate these same moneyed-interests.
Utility Admits Pushing Energy First and Placing Safety Last
For several years nozzles in the Davis-Besse vessel head were allowed to leak corrosive coolant through cracks growing in the welds. Throughout 2001, mystified operators and NRC inspectors ignored rising radiation levels and the growing presence of iron oxide particulate inside the containment. Meanwhile, leaking borated coolant flashed to steam on the carbon steel vessel head leaving a growing mound of caustic crystals.
The company was aware of the corrosion problem occurring inside the containment. Company documents from 1999 account for degraded containment radiation monitor filters due to the persistent accumulation of rust particulate.
"While the exact source of the rust is not known, the high particulate problem developed about the same time as Plant Startup (10 May 1999) after the mid-cycle outage... Subsequent filter changes were required every 24-48 hours,"
states a condition report. The report says,
"The Iron Oxide particles had a granule appearance indicating the source is from corrosion."
Rather than find on the source of corrosion, First Energy focused on keeping the reactor running and brought in HEPA filters to control the dust in the containment atmosphere.
The company was aware of a "large boron accumulation" on the vessel head. "Boric acid corrosion may be a concern," states a work order during the 2000 refueling outage. The work description is clear. Workers were to
"Clean boron accumulation from top of reactor head and top of insulation"
using high-pressure steam. The safety significance for issuing the work order was clear:
"The program is required due to degradation of the CRDM nozzle caused by Primary Water Stress Corrosion. In order to perform required inspections the nozzles as well as the penetrations must be free of boron deposits. Once the head is free from boron, new boron deposits may be easily noted and remedial action taken."
The order was signed off on 25 April 2000 as "Work completed without deviation." The reactor restarted for its two year run before the next outage.
In fact, the vessel head was never cleaned or inspected. Over 900 pounds of the acid crystal had accumulated on the reactor head. The boron deposit was left to eat a hole roughly seven inches in diameter and over six inches deep through the top of the carbon steel vessel head. A thin stainless steel inner liner, approximately 3/16th inch (less than 5mm) thick, was all that remained at the bottom of the cavity. With the load bearing iron oxidized into dust, the liner bulged out into the cavity from the reactor's internal pressure of over one ton per square inch. The liner, never intended as a pressure bearing material, cracked from the internal stress. Davis-Besse was unpredictably close to shearing open the vessel in a loss-of-coolant accident much worse than the 1979 core-melt accident at Three Mile Island-2.
In August 2002, FE offered federal regulators its analysis of the company's failure to identify and prevent the extensive corrosion of a principle safety component. The company's new and now "humbled" executive officer told NRC that "There was less than an adequate nuclear safety focus. There was a focus on production, established by management, combined with taking minimum action to meet regulatory requirements that resulted in the acceptance of degraded conditions." Indeed, we almost lost Toledo.
NRC Safety Concerns Eclipsed by Utility Financial Concerns
In theory, regulatory safeguards exist against the reckless operation of aging reactors. The NRC is mandated to "reasonably" assure the safety of the 103 U.S. reactors engaged in the dangerous business of splitting atoms to generate electricity. In practice, the Davis-Besse example warns that safety nets are rolled back to meet production schedules and maximize corporate net profits.
After NRC staff found dangerous cracking at South Carolina's Oconee Nuclear Station early in 2001, concerns over the prevalence of cracks and potential safety implications in all PWRs were raised. Thirteen of the nation's sixty-nine PWR ranked with the Oconee units as "highly susceptible." Davis-Besse was among that number. Staff focused concern on those units to shutdown and inspect vessel pressure boundaries and report back. This concern prompted the issuance of NRC Bulletin 2001-01 "Circumferential Crack of Reactor Pressure Vessel Head Penetration Nozzles" on 3 August 2001. On 4 September 2001, FE requested the waiver.
NRC engineers challenged the company request by preparing an "Order" to require a shutdown by 31 December 2001 for the safety inspection. Nearly two thousand pages of documents obtained by NIRS through the Freedom of Information Act reveal an array of staff concerns as they quietly began drafting the Order. As one NRC staffer wrote following FE's amendment request,
"The level of degradation that has been found in other plants, if left undetected and uncorrected, would result in a gross failure of the reactor coolant boundary. On that basis, staff does not have reasonable assurance that adequate protection is achieved by plants that do not perform inspections that are sufficient to detect this type of degradation."
Davis-Besse's managers were not to be dissuaded by NRC concerns. On 11 October 2001, FE officials appeared before the Commissioners Technical Assistants to defend their waiver request. Senior managers testified before the Commission that all of Davis-Besse's vessel head penetrations were
"free from characteristic boron deposits using video recordings from the previous 2 refueling outages"
made both before and after cleaning the head and
"Davis-Besse has done more and better quality inspections than other plants."
In so doing, company managers provided patently false statements to the Commission.
Internal NRC documents show that FE Vice President Guy Campbell pressured NRC officials to allow the reactor to operate until the regularly scheduled April outage.
"I told him that based on the operating experience there is a high likelihood that they (Davis-Besse) have leaks- he agreed,"
states the NRC-sourced email. The communication concluded,
"I told him that the reactor coolant pressure boundary leakage would not satisfy the regs [regulations - Ed.] or tech specs, that it would fail to maintain defense in depth and margins, and that there could be a crack large enough to cause them problems before April."
In fact, Davis-Besse's licensed condition required them to shut down within six hours of recognition of any reactor coolant leakage. The NRC was technically justified and legally authorized to immediately shut Davis-Besse down. But as a NRC senior official reported
"we are exercising discretion noting it would be clearly punitive to immediately shut a plant down and they sit there for a month waiting to obtain the correct inspection equipment etc."
By NRC's own admission, the agency compromised safety regulations by deliberately allowing Davis-Besse to operate in violation of its operating license in providing the arbitrary 31 December 2001 deadline. This was a technically indefensible concession to the utility's financial interest. Yet even that was not enough for First Energy.
By letter dated 21 November 2001 a detailed Order was finalized by staff and signed by Nuclear Reactor Regulation Director Sam Collins. It was forwarded to the Commissioners through the agency's Executive Director of Operations. The Order was to be issued to the utility no less than 5 days upon Commission receipt. The Order states
"...a potentially hazardous condition exists and warrants the issuance of an Order"
to shut down Davis-Besse by 31 December 2001 and perform inspections on the vessel head.
An internal NRC memorandum dated the same day and exchanged between two technical assistants to Commissioners Nils Diaz and Greta Dicus reveals what happen next:
"Sam Collins stopped by around 3:00 pm. RE: Davis-Besse status... Licensee does not want an order: perception; replacement fuel; financial markets... Licensee is trying to come with contingency plans if stopped at end of January or end of February. If February, would have fuel and contracts in place. Minimum impact. If January, significant."
The Order was suddenly withdrawn following a meeting with FE on 28 November 2001. Instead, Mr. Collins allowed Davis-Besse to operate until 16 February 2002.
Federal Investigations Mount As Davis-Besse Lumbers Towards Restart
The Davis-Besse damage and the negligence of the utility and the NRC has launched three federal investigations including a criminal proceeding against FE for providing false information and an internal investigation into how the NRC deferred its regulatory responsibility. At the request of Ohio members of Congress the U.S. General Accounting Office is also looking into the near-miss accident. A Republican-led Congressional inquiry has already quietly closed without any findings but the call for a Congressional investigation continues to be heard.
After going through an extensive "junkyard" of cancelled reactor construction projects, FE managed to find an unused B&W vessel head of similar vintage at Michigan's unfinished Midland reactor. The total cost to replace the corroded reactor lid, fix other equipment and buy replacement power now stands at US$400 million, mostly to be paid for by company stockholders. The utility sought restarting the reactor by early December 2002 but company projections are being delayed by an unraveling story of mismanaging an inherently dangerous industry.
Once again, NRC bowed to corporate concerns that idling a reactor with an early shutdown would be perceived poorly on the financial market. NRC will likely continue its business-as-usual approach to defer safety standards to protect the coffers of an economically struggling atomic power industry. The public interest community must remain focused on the fact that our safety is increasingly at risk from utilities running aging reactors harder while seeking to cut costs under diminishing regulatory oversight. Left unchecked, this kind of corporate advocacy over public safety becomes an increasingly dangerous bargain converging on nuclear disaster.
Source and contact: Paul Gunter, Reactor Watchdog Project, NIRS; World Information Service on Energy
