12/01/02

As stark as a morgue photo, the picture from FirstEnergy Corp.'s files captures a reactor in distress.

Something is hemorrhaging atop the massive steel lid that covers the radioactive core of the Davis-Besse nuclear plant near Toledo.

The vivid color print, taken in April 2000, shows rust trails the hue of dried blood spilling from inspection ports on the reactor's sloping dome.

The corrosion stains end in piles of white-brown debris at the lid's edge. The loose clumps of dried acid are trapped there like fallen leaves against a fence by the ring of huge bolts that locks the 80-ton cap in place.

Anyone who saw the image that has come to be known as the "red photo" would have to question whether the lid - a vital safety barrier - was damaged.

"I would have concluded that a serious corrosion problem probably existed" on the lid, said Digby Macdonald, an international corrosion expert who directs Pennsylvania State University's Center for Advanced Materials.

But federal regulators never got that chance.

FirstEnergy's nuclear division didn't share the 2-year-old photo with senior Nuclear Regulatory Commission staffers last fall. It wasn't in the batch of images the company provided the NRC in November 2001 as part of FirstEnergy's successful campaign to convince the agency the lid was OK, and to justify postponing a costly shutdown to inspect it.

The photo didn't surface until April, on page 93 of a thick FirstEnergy report. The document attempts to explain in hindsight how the company had allowed boric acid sludge left behind by leaking reactor coolant to chew a pineapple-size hole all the way through the 6.5-inch-thick lid.

The unprecedented hole, found a month earlier, jeopardized the plant's safety, rocked the nuclear industry and is expected to cost the company nearly $400 million in repairs and replacement power purchases.

The omitted photo is just one example of what regulatory officials say are FirstEnergy's multiple failures over almost a decade to accurately document and communicate what the company knew to be the worsening condition of Davis-Besse's reactor lid.

Those misrepresentations - especially during the crucial NRC review last autumn - are the subject of an agency criminal probe. They also are the subject of a new allegation by a watchdog group, the Nuclear Information and Resource Service, which is calling for Davis-Besse's license to be revoked. The NRC is reviewing that claim and may address it in its ongoing investigation.

The NRC already has determined that FirstEnergy's nuclear division violated agency rules requiring that information be accurate and complete. The company insists, without further explanation, that it did nothing criminal. But if the inquiry under way by the NRC's criminal unit, the Office of Investigations, verifies intentional wrongdoing, plant personnel and FirstEnergy managers could find themselves answering to a federal grand jury, or facing hefty civil fines.

The findings also could affect the NRC's decision on whether or when to allow Davis-Besse to resume making electricity.

Neither the government nor FirstEnergy has been willing to say much publicly about the records issue because of the investigations. In the last few weeks, however, a picture of the evidence in the case and the company's defense has emerged from newly released NRC reports as well as a Plain Dealer review of thousands of pages of inspection documents, meeting transcripts and briefing materials.

While not contesting that their records were inaccurate and incomplete, FirstEnergy officials have sought to portray the documentation problems as benign miscommunications or misinterpretations rather than deliberate attempts to deceive. They have said that evidence such as the "red photo" that gave a more detailed indication of the lid's condition was available if the NRC had looked hard enough.

"It was there for the asking," said company spokesman Todd Schneider. "Being our regulator, the NRC has full access to the plant, to our documents, to just about every part of our operation."

But that rationale sidesteps the key legal issue of why material in FirstEnergy's files sharply differs from the rosy picture the company painted for the NRC late last year to justify the reactor's continued operation.

"I think that's a little bit disingenuous," Brian Sheron, the agency's associate director for project licensing and technical analysis, said. "We were asking them to provide us with all the information to support their argument to operate beyond Dec. 31. Apparently, we did not get everything."

The NRC itself is under fire from critics, including some members of Congress, for allowing the plant to delay its lid inspection last fall. Angry and embarrassed agency staffers say they made the right call based on the information they had.

"If we knew they had three or four inches of [acid] caked on top of the head . . . that would have started the chain" of more intense questioning, Sheron said.

Had FirstEnergy disclosed that its inspections dating at least to 1998 had consistently found red, rusty lumps of acid on the lid - increasingly large deposits that weren't fully cleaned off so the surface underneath could be checked - "we would have challenged the licensee then and there to explain what we were seeing," Sheron said. "If we didn't get a reasonable explanation, we probably would have taken action to try to force them to shut down."

String of inaccuracies

The string of inaccurate and incomplete Davis-Besse records that the NRC has identified began in 1993.

At that time, managers at the plant and at FirstEnergy's predecessor, Toledo Edison Co., were debating whether to modify a platform that sits atop the reactor lid. The structure helps support the dozens of control rods that pass in and out of the reactor's core through sleeves, or nozzles, in the lid to regulate the nuclear reaction. It also holds insulation to contain the reactor's fierce heat.

The problem with the service structure, though, was its close fit. At the top of the lid, there was only a 2-inch gap between the lid's metal surface and the insulation, making inspection of that area extremely difficult.

To check the lid's condition every two years during the plant's refueling shutdown, inspectors attached a video camera to a pole and poked it through one of the 16 small "mouse holes" that ring the service structure's base. But it was hard to get the camera all the way to the top of the lid.

Davis-Besse's sister plants had begun cutting larger ports in the structure to allow for better inspection and cleaning. In March 1990, a Davis-Besse engineer recommended that the plant do the same after finding boric acid residue from leaking coolant in several places on the lid. He reminded his bosses of the acid's potential for harm.

Managers finally decided in September 1993 that the modification wasn't needed. The reason, according to the cancellation notice signed by four high-level managers, was because "cleaning of the reactor vessel head [lid] during last three outages [in 1990, '91 and '93] was completed successfully without requiring access ports."

That statement wasn't accurate, the NRC has determined. Agency inspectors who reviewed Davis-Besse records from the 1991 and '93 refueling shutdowns found that workers had allowed acid deposits to remain on the lid each time the reactor was restarted.

FirstEnergy's own review this year notes that there are no records indicating the lid was inspected at all in 1990. The FirstEnergy report doesn't say what, if anything, the 1991 records show, but acknowledges the company can't verify the effectiveness of the lid-cleaning done in 1993.

The record-keeping flaws at Davis-Besse continued in 1998. Plant documents from that year stated that workers had cleaned acid buildup from the lid, even though the company noted as an aside that its reactor's manufacturer, Babcock & Wilcox, considered such deposits harmless. Plant records also said inspections had shown the lid surface was free of "any" corrosion damage.

All three statements were incorrect, the NRC has found.

A videotape of the 1998 inspection showed fist-sized clumps of red, rusty acid on parts of the reactor lid, and Davis-Besse workers again allowed some of them to remain, especially on the hard-to-reach top of the dome. That precluded plant personnel from knowing, as they claimed to, that the underlying metal was OK. In fact, the hole in the lid had started its rapid growth that year, FirstEnergy surmises, in the very area workers had left uncleaned.

Also, none of the nine Babcock & Wilcox reports the NRC examined contained the reassuring statement FirstEnergy had quoted: that acid residue left on the lid wouldn't cause corrosion.

There were multiple inaccuracies in Davis-Besse documents from 2000, the NRC has found, most having to do with claims that the reactor lid was rigorously cleaned and that inspection showed it to be unblemished.

"Work performed without deviation," noted an April 25, 2000, order signed by the reactor coolant system engineer detailing the lid-cleaning activities. "Engineering displayed noteworthy persistence in ensuring boric acid accumulation from the reactor head was thoroughly cleaned," trumpeted a July 7, 2000, report by the plant's quality assurance unit.

None of it was true. As FirstEnergy acknowledged in reports to the NRC this year, Davis-Besse personnel were under intense pressure to stay on the tight work schedule during the refueling outage so the plant could resume making electricity - and money - as soon as possible.

Workers examining the lid at the start of the 2000 outage found rock-hard, "lava-like" piles of acid that clogged some of the mouse holes and hindered the video camera's path. They did some cleaning, but with time running out, managers decided to stop, leaving some acid clumps in place and part of the lid unchecked. Contrary to policy, they didn't do a written evaluation to justify their actions.

Eighteen months later, when FirstEnergy officials were pressing the NRC to postpone the mandatory lid inspections that most other plants were doing to look for possible nozzle cracks, they assured the agency their lid was in good shape. But as the NRC would later discover, the evidence the company provided was selective and misleading.

In letters and in-person briefings to the NRC staff at the agency's Rockville, Md., headquarters, company officials mentioned having found "some" boric acid in past inspections. But they didn't reveal the alarmingly rusty characteristics and amount of the acid residue - by this time nearing 900 pounds, they later found out - that had been accumulating for years.

In one meeting, for example, FirstEnergy nuclear division president Robert Saunders "said he knew there was some light dusting of boron in certain spots. But he said he was not concerned that was from major leakage," recalled the NRC's Sheron.

And in an Oct. 17, 2001, letter, FirstEnergy nuclear division support services director L.W. Worley told NRC staffers that the lid was cleaned in 1996. He added that re-reviews of the videotapes from that inspection and ones in 1998 and 2000 "did not identify any leakage in the . . . nozzle-to-head areas that could be inspected."

When NRC staffers continued to push for a shutdown by Dec. 31, FirstEnergy officials volunteered to fly to Rockville with the inspection videotapes so NRC staffers could see for themselves. But the company didn't show any tapes that depicted the masses of rusty acid accumulated at the center of the lid, according to the NRC's subsequent interviews with staffers who attended the meeting.

"The NRC staff members recollected that they were shown freeze-frame video images that depicted inspectable nozzles, i.e. free of significant deposits," an NRC task force reported last month. "The nature and extent of boric acid deposits remaining on the [lid] . . . were not disclosed."

The NRC has always been heavily dependent on the candor of the utilities it regulates. There are 103 commercial nuclear reactors in the United States, each a highly complex machine with dozens of operational issues per day that require attention and generate thousands of pages of paperwork.

Even in the best of times, the agency has only a dozen or so people monitoring the day-to-day operations of an individual plant - two or three on-site inspectors and the rest at regional offices or at headquarters.

Because of its staffing level, "the NRC doesn't count every thread on every bolt; we focus on things that are safety-significant," the agency's Sheron said. "We poke, we probe, we ask questions. But for the most part, we rely on the licensee. Our whole regulatory process is based on trust."

The agency's oversight of Davis-Besse was particularly vulnerable at the exact time the hole in the reactor's lid was forming and growing, in the late 1990s.

The resident inspector's post at the plant was vacant for a year; the job of senior project engineer for Davis-Besse at the NRC's Midwest regional office was left empty for 20 months. And there were serious problems at other area reactors that required attention, so the amount of time the agency spent on inspections at Davis-Besse plummeted to an eight-year low.

However, the NRC's own shortcomings don't explain FirstEnergy's repeated failures to disclose what it knew.

Legal review

After news of the NRC's probe of possible criminal wrongdoing leaked this spring, FirstEnergy asked one of its law firms to review staff activities at the plant during the past decade.

The company won't discuss or release the findings, but an executive told stock analysts in September that while Davis-Besse managers had made poor decisions in operating the reactor and dealing with federal regulators, the law firm found no behavior "which would rise to criminal."

Instead, in numerous filings and meetings with the NRC to explain itself, FirstEnergy has depicted former Davis-Besse managers as production-obsessed and out of touch with the plant, and workers as being naive about the potential for boric acid deposits to harm the reactor lid if left in place. Only wet boric acid posed a corrosion threat, and plant personnel wrongly thought that, once the steel lid's searing heat instantly dried the leaking coolant, the acid deposits left behind couldn't get wet again.

But if Davis-Besse personnel truly believed the acid buildup was harmless, why not acknowledge its presence? Why say or imply that it had been fully cleaned away when it hadn't?

"That is one of the standards problems we're trying to correct at the plant - that cleaning the head back then meant cleaning as much as you could, not the entire head," said FirstEnergy's Schneider. "That's about all I can say on that. Those issues will come out in the investigation."

FirstEnergy has fired, transferred or reprimanded some senior employees in connection with the hole in the lid. But it is mum on whether those moves, which included the departure of nuclear division Vice Presidents Guy Campbell and Howard Bergendahl and engineering director John Wood, were because of the record-keeping inaccuracies.

The NRC's Sheron said FirstEnergy nuclear division President Robert Saunders told him the disciplinary steps were a consequence of its law firm's review. "I don't think a company would fire somebody if it concluded they hadn't done anything wrong," Sheron said.

FirstEnergy's Schneider has said the company expects to be fined for its overall lapses, but that the management changes and renewed focus on safety should be enough to regain the trust of the NRC and the public.

A national nuclear watchdog group disagrees. Concerned that the NRC will accept superficial changes at Davis-Besse and not push for fundamental reforms, the Nuclear Information and Resource Service recently filed a complaint with the NRC. It alleges that FirstEnergy records contained false and inaccurate statements about the Davis-Besse reactor lid, and that the company's analyses of the event have failed to explain why.

The NRC has assigned the allegation to a review board, and, if it's deemed serious enough, it could be incorporated into the Office of Investigation's ongoing work.

Davis-Besse "should have its operating license revoked," said Paul Gunter, director of NIRS' reactor watchdog project and the author of the complaint. "Our concern remains that the NRC is going to go along with this plan to just replace managers at Davis-Besse as the solution to underlying problems with the management culture that places production over safety."

With Davis-Besse aiming to finish its repair work by late January or February, it's possible the NRC may have to decide whether to let the plant resume operating before the agency's criminal inquiry is complete.

Although it may not have the final report in hand before restart, the special NRC panel overseeing Davis-Besse's rehabilitation will have a good idea of what the findings will be, said its chairman, Jack Grobe. The decision will hinge on whether FirstEnergy has corrected whatever deficiencies the probe finds.

"We'll have to have confidence in the plant personnel" before letting Davis-Besse power up again, Grobe said. "This is a critical element."

For full coverage of Davis-Besse, go to www.cleveland.com/davisbesse/

To reach these Plain Dealer reporters:

jmangels@plaind.com, 216-999-4842

jfunk@plaind.com, 216-999-4842