The head of the Nuclear Regulatory Commission told public officials that chances for an explosion at Three Mile Island were "pretty close to zero," even as he ordered his staff into a frantic effort to calculate the risks, according to a chronology of the March event released yesterday.
It turned out that Chairman Joseph M. Hendrie was right and the danger of a blast was negligible, but neither he nor most of his staff was certain of that until two days later. Meanwhile they continued to reassure the public for 36 hours until news reporters revealed their worry: that a hydrogen bubble might blow the reactor apart and spread radiation over much of Pennsylvania.
The chronology, compiled by the NRC staff at the request of the president's commission on the accident at Three Mile Island, showed that Hendrie was the first to begin wondering about the possibility of a blast.
"I speculated . . . on whether free oxygen was being generated and going into the bubble," Hendrie recalled for the chronology. This was on the morning of Friday, March 30, two days into the accident. Oxygen would have to be present in order for the hydrogen to explode. "I wanted an estimate of the oxygen evolution rate . . . I also asked for estimates of the flammability limit and the detonation limit."
During the rest of Friday, NRC staff members made feverish calculations of the amount of oxygen that would have to be present for the mixture to go off. They also worried that the bubble would grow to the point of uncovering the reactor core, possibly causing overheating and a meltdown of the fuel.
Friday evening, however, Hendrie personally reassured Pennsylvania Gov. Richard Thornburgh and Rep. Mike McCormack (D-Wash.) that there was little danger.
"What are the potentials for an explosion that would rupture the core?" Thornburgh asked over the telephone.
"There isn't any oxygen in there to combine with that hydrogen, so the answer as far as I know is pretty close to zero," Hendrie replied.
Later, he told McCormack that "there isn't anything in the core it [the hydrogen] can react with. You know? No. There's either none or very little [oxygen]," he said.
An hour later, however, Hendrie told his fellow commissioner, Peter A. Bradford, that although the bubble was "probably pure hydrogen," he wasn't sure and wanted more numbers. "Over some period of time . . . you're going to begin to get enough oxygen up in there to worry about the thing. And if there's anything I don't particularly think I need at the moment, it's flammable -- you know, for the bubble to be in a flammable configuration."
Several NRC staff members and their outside consultants said immediately that the high pressure inside the reactor core would prevent any free oxygen from escaping into the bubble, but they had trouble getting anyone to listen.
Through Saturday evening, recalled Warren S. Hazelton of the NRC's engineering branch, he tried to convince his superiors there was no danger, "But apparently I did not succeed in changing anyone's mind . . . I was not pleased."
Edson Case, deputy director of reactor regulation at the NRC, said yesterday he would do it the same way again. "When there's a lack of complete knowledge, it's better to approach things from a conservative direction," he said.