The importance of criticality was outlined this week in a report by Britain's nuclear inspectors into the failings at the Dounreay nuclear facility in Scotland: "The hazard of criticality is not being afforded the respect it deserves," they said.
The key to building a nuclear reactor and an atomic bomb is making sure that criticality (or "critical mass") is not reached until you want it to. In the third case of a reprocessing facility for nuclear waste, criticality is something to be avoided at all costs.
A chain reaction begins when an atom of a radioactive material like uranium- 235 splits and throws out two neutrons. If there is another U-235 atom near, one of those neutrons is sufficient to split that atom too, releasing another two neutrons and the chain reaction has begun.
There is also energy released, and it is this combination of a chain reaction and energy release that is so destructive in an atom bomb, and convenient (as long as it's kept controlled) in a fission reactor.
The important thing for controlling criticality is making sure that those neutron sources are kept apart, and that the amount of neutrons being released do not reach levels which could accidentally start a chain reaction. About 10 kilograms of 90 per cent pure U-235 suffices to make an atomic bomb. But if you get the same sort of mass in a larger space while there are lots of radioactive sources around - which can happen during fuel reprocessing - then you could conceivably reach criticality while not intending to. That is what the nuclear inspectors at Dounreay were worried about.
What are the consequences of accidental criticality? One such event happened on June 17 last year, at the Russian nuclear facility Arzamas-16. Dr Aleksandr Zakharov, a 42-year-old senior research worker with years of experience, was assembling the pieces of an experiment to reach criticality using enriched U-235. He was behind thick glass, using remote control systems to put together a sphere of material; the idea was to approach criticality very carefully, then get out of the room before putting the final piece in place using a robot. The result would not be a mushroom cloud, but a self-sustaining reaction, generating heat and neutrons.
But the penultimate piece fell wrongly. It set off the chain reaction ahead of time, before Dr Zakharov could get out of the control room.
Dr Zakharov saw a blue flash as the mass went critical. The room and the control room were bathed in neutron radiation. The system didn't explode - it was too small for that - but Dr Zakharov had already received a fatal radiation dose. After unsuccessfully trying to stop the reaction he exited the facility and closed the hatch, reported the incident to management and lost consciousness. He died in hospital 36 hours later.
It took six days to get a robot to break apart the assembly and stop the reaction: the neutron radiation levels fell by a factor of 10 million.
In that context, one can see why nuclear inspectors emphasise the respect one should pay to criticality.
CHARLES ARTHURReuse content