Physicists Scoff at Risk of Destroying Planet

by Hazel Muir

The probability that dangerous Earth-devouring particles will be born at a new particle accelerator may be very, very, very tiny, but scientists have played down the devastating potential costs in their risk assessments, a physicist now says.

Adrian Kent of Cambridge University accepts that the chances of catastrophe are minuscule. But he claims physicists are not accounting for the scale of the potential devastation -- the destruction of the entire planet in their risk analysis. "Small catastrophe risks are more costly than we've generally considered," says Kent.

Last year, sensational newspaper reports suggested that a new particle accelerator, the Relativistic Heavy Ion Collider (RHIC) at Brookhaven National Laboratory on Long Island, could put the Earth in peril. The accelerator might create blobs of matter called strangelets containing "strange" quarks, as well as the usual "up" and "down" types in ordinary matter. If a strangelet were stable and negatively charged, it might begin eating the nuclei of ordinary matter, converting them into strange matter. Eventually this could consume the entire Earth.

Physicists reassured the public by pointing out that such a chain of events is nigh-on impossible. The main reason for their confidence is that it's very unlikely that strangelets would be stable or negatively charged.

Astronomical evidence is also reassuring. Heavy ion collisions occur naturally, for instance when cosmic rays smash into heavy nuclei on the Moon. Yet the Moon has existed for five billion years without being devoured by a ravenous strangelet.

Arnon Dar, Alvaro De Roejula and Ulrich Heinz of the CERN particle physics laboratory near Geneva used the fact that stars are not being changed into strange matter at a significant rate to calculate the maximum probability of the Brookhaven collider creating a dangerous strangelet during the accelerator's lifetime. The result was less than 20 chances in a billion, which the team called "a safe and stringent upper bound." Thus reassured, the Brookhaven lab set RHIC running in June.

But Kent now argues that this probability is far from acceptable when compared with risk assessments of other possible hazards, which factor in the scale of the devastation if things go wrong. He points out that radiological protection policy in Britain deems it unacceptable for solid nuclear waste to pose more than a one-in-a-million chance of killing around five people a year on average. To achieve the same level of risk, global annihilation by a strangelet would have to have a probability of 1 in 10-15 or less, because it has the potential to wipe out all 6 billion people on the planet.

John Marburger, director of the Brookhaven lab, does not accept this line of argument. He says that the risk limit calculated by the CERN researchers was not a safety assessment in the sense used in nuclear energy safety analysis. "In my opinion, no such risk assessment is possible," says Marburger. "We do not know how a 'dangerous strangelet' would be created, and we do not know the properties of one if it were."

De Roejula agrees. "We've been standing on our heads to try to convince people it's not true and to come up with limits on it, and this number comes from very, very pessimistic assumptions." He says it is "absurd" to take such a minuscule maximum probability and multiply it by the number of people who would die if a strangelet swallowed the world. "It's not going to happen," he says.