Reactor Design Edges Toward Approval, but Not Without Complaints

The Nuclear Regulatory Commission is moving toward approval of the Westinghouse AP1000 reactor and its strikingly different containment design, which has far fewer pumps and valves plus a safety system that relies mostly on foolproof forces, like water flowing downhill or heat rising. But complaints over the design persist. In a notice on Feb. 24 in the Federal Register, the commission invited public comment on its intention to approve the design; the comment period runs until May 10.

The notice quickly drew comment from Representative Edward J. Markey of Massachusetts. In a letter sent on Tuesday to the chairman of the regulatory commission, Gregory Jaczko, Mr. Markey asked that the commission hold off until it has resolved a dispute with one of its staff members, John S. Ma, a senior structural engineer, over whether a shield building in the new design could withstand an earthquake or the impact of an airplane crash. “Taxpayer dollars should not be spent on reactors that could be at risk of suffering a catastrophic core meltdown in the event of an aircraft strike or a major earthquake,’’ Mr. Markey wrote.

The shield building does not exist in the classic Westinghouse design that is used in the United States and around the world. Current Westinghouse reactors have a thick concrete dome, with a steel liner, that forms the containment.

The new design has a relatively thin steel shell with air space around it. In case of an accident, the classic containment structure is a kind of thermos bottle holding in unwanted heat. But in the new design, the circulation of air would carry off excess heat.

The shell is surrounded by the shield building, which protects the shell from earthquakes, tornadoes and the hypothetical hijacked airliner. It also supports a huge tank of water that would be used for cooling in an emergency but which would not need pumps because it is on the roof.

The N.R.C. rejected the company’s first design for a shield building, largely because of arguments by Mr. Ma that it might not survive an earthquake. So Westinghouse redesigned it.

Now, Mr. Ma has complained that some of the computer codes used to analyze the new design were not precise enough and that some of the materials used were too brittle. “It has not been demonstrated that the building can absorb and dissipate energy imparted on the structure by an impact or a seismic event,’’ he wrote in a “non-concurrence.” And the company did not use a design code issued by the American Concrete Institute, he said.

The degree of scrutiny of a design element should be in proportion to its importance, he said, and this part is very important.

But Thomas A. Bergman, the director of the division of engineering in the commission’s Office of New Reactors, said that the use of the code was only suggested by the agency, not required. Changes recommended by Mr. Ma “would result in a more robust structure,’’ he wrote. “However, these changes are not necessary to meet regulatory requirements.’’

He acknowledged that the structure relied on a steel and concrete composite “to an extent never before reviewed by the N.R.C.”
After some discussion, he said, “the non-concurrence remains in place.”

In an interview on Tuesday, Mr Bergman said what while Mr. Ma’s changes would have made the shield wall capable of surviving a more powerful earthquake, the rest of the structure would not be stronger, so the change would not be useful.

The chairman of the Nuclear Regulatory Commission, Mr. Jaczko, told reporters on Tuesday morning that the agency had had substantial discussions about Mr. Ma’s non-concurrence and that this was “something the public should see as a confidence booster.”

He said two panels had examined the issues and that he had met with Mr. Ma. “The design meets our standards,” he said. “His changes would lead to enhancements in safety, but without them, it is acceptable and fully safe.”

A panel of commission experts previously evaluated and rejected the idea that another component of the containment, a metal shell within the shield building, could quietly rust away and create holes through which radioactive materials inside would be drawn out by the chimney effect of the surrounding building.