Nuclear Accident And The Elimination Of Risk Thirty-nine years ago, one of the most senior nuclear safety officials in the United States penned a stark memorandum to colleagues. It warned that a key bulwark against a catastrophic accident in the kind of reactors now in meltdown at Fukushima was so flawed it should be banned. He got a reaction within a week. The idea of a ban was “attractive”, wrote an even more senior official, who went on to head the main US nuclear safety agency. But implementing it “could well be the end of nuclear power”. This extraordinary exchange is only one of a series of warnings over recent decades about the safety of the General Electric “Mark I” Boiling Water Reactors, which comprise five of the six installed at the stricken complex in Japan – and all those so far hit by the world’s first multi-reactor nuclear disaster. As workers battle to contain the ravages of the many-headed monster unleashed at the complex by last week's earthquake and tsunami, it is of course too early to work out what precisely has caused the apparent partial meltdowns at three of the reactors, and the exposure of highly radioactive spent fuel in a storage pool at a fourth. But the story of the warnings sheds some light on attitudes in the young nuclear industry which may, only now, be taking their severest toll. The memo was written on September 20, 1972 by Dr Stephen Hanauer, then working at the US Atomic Energy Commission. He was concerned that weaknesses in the design of the container vessel around the reactor core – the main shield against the escape of radioactivity – could cause it to fail in an accident, leading to dangerous releases. He pointed out that the vessel was smaller than was "conventional" and lent itself to the build-up of hydrogen – which seems to have played a part in the events at Fukushima – becoming "a more serious problem". He recommended that the commission "adopt a policy of discouraging further use" of that kind of containment and "that such designs not be accepted for construction permits filed after a date to be decided (say two years after the policy is adopted)". On September 25, Dr Joseph Hendrie, the deputy director for technical review at the commission's directorate of licensing, weighed in with his reaction. The idea of such a ban, he wrote in a memo of his own, "is an attractive one in some ways", adding that alternative containments had "the notable advantage of brute simplicity". But he added that the concept that worried Dr Hanauer had been so widely accepted "by all elements of the nuclear field", including the regulators, that it had become "firmly embedded in the conventional wisdom". He went on: "Reversal of this hallowed policy, particularly at this time, could well be the end of nuclear power." It would, he concluded, "throw into question the continued operation of licensed plants", could make new reactors being reviewed "unlicensable" and "would generally create more turmoil than I can stand thinking about". Five years later, he became the US government's top watchdog on the industry as chairman of its Nuclear Regulatory Commission (NRC). But that was not the end of the warnings. Harold Denton, the NRC's own director of nuclear reactor regulation – who had been one of the heroes of the successful struggle to prevent the Three Mile Island accident becoming a catastrophe – is reported to have voiced his concern in 1986 about the containment being "smaller" than conventional ones, adding that a giant official safety study had shown "something like a 90 per cent probability of that containment failing". Nor was this the only issue to arise over the Mark I's design. In 1976, three General Electric engineers resigned because, in the words of one of them last week, the effects of a loss of coolant to the reactor core had not been fully taken into account. The result, Dale Bridenbaugh added: "could tear the containment apart and create an uncontrolled release". He said the company had eventually agreed to retrofit their reactors to take account of such concerns. The result: "the Mark I is still a little more susceptible to an accident that would result in a loss of containment". General Electric says: "The Mark I meets all regulatory requirements and has performed well for over 40 years." It adds that it had been told in 1980 that US regulatory staff, "including Dr Hanauer", had accepted that the containment was "safe" and that Mr Bridenbaugh's concerns are "unrelated to the current situation." But it puts up little defence of a design flaw that does appear to be at the heart of the crisis at Fukushima – the extraordinary practice of putting the pool where the highly radioactive used fuel is stored on an upper floor of the reactor building, instead of, as is normal, having it at ground level. This has made it particularly hard for workers to try to fill it with the water needed to isolate the fuel, keep it cool and stop it from melting – and is one reason why helicopters have had to be used. General Electric merely offers that the pool "is not above the reactor vessel and not designed to drain to the reactor vessel". Well, one would hope so! Yet the Mark I continues to go strong. Twenty three are still in operation in the US alone (though there are none in Britain) and one, which has been operating in Vermont since 1973, was granted another 20 years of life the day before the earthquake struck off Japan. New reactors have been designed to be much safer, with the European Pressurised Reactor (EPR) – planned to be built here over the next few years – especially so. But official documents show that the EPR will produce several times more of the radioactive iodine and caesium that would be rapidly released in an accident than do present-day reactors. The trouble is that human beings have a way of overriding safety systems. The 1957 Windscale fire occurred when a physicist threw a switch too soon when carrying out a routine operation. The Three Mile Island accident was caused by a whole series of human errors, while at Chernobyl, operators under pressure to complete a test deliberately and progressively switched off every one of its safety systems until, in the words of the chief investigator into the catastrophe, the reactor "was free to do as it wished". "Plants grow more complex," said one leading nuclear engineer after Three Mile Island. "Safety hangs increasingly on the human error factor, and we can't eliminate it. Many of our operators have seen emergencies only on a simulator. The real thing can look quite different, and they may have just 60 terrified seconds to act." Or, as one witness to the accident put it: "Bells were ringing, lights were flashing and everyone was grabbing and scratching." Try as we may, there will be more accidents. Yet, despite the reactions of some panic-stricken politicians to events in Japan, abandoning nuclear power is not an option, especially in a world needing to combat climate change. And it's better to have new reactors than old ones. But, equally, the huge disruption that a disaster can cause, as reactors are subsequently closed for safety checks and new ones delayed, and the anti-nuclear revolt that inevitably ensues, make it unwise to become too dependent on nuclear power. Instead of falling in and out of love with the atom, as we regularly do, we need a more sensible, watchful, relationship.