The detected isotopes of barium, strontium and lanthanum would be created in the core of a nuclear reactor, which produces energy by splitting uranium atoms in a chain reaction. These isotopes would have been released if a core exploded, says Claire Corkhill, a nuclear scientist at the University of Sheffield, UK.

Any damage an explosion might have caused to the reactor core would probably have led to the release of radioactive iodine and caesium, says Marco Kaltofen, a nuclear scientist at the Worcester Polytechnic Institute and the environment investigation firm Boston Chemical Data Corp, both in Massachusetts. An uncorroborated report in The Moscow Times on 16 August said that local doctors had traces of caesium-137 in their muscle tissue. And a Norwegian nuclear authority detected an unexplained spike in radioactive iodine-131 almost 700 kilometres away in Svanhovd after the blast. But this could be from another source: iodine-131 can be released in small quantities during the production of radionuclides for medical purposes, says Corkhill.

Boris Zhuikov, head of the Laboratory of Radioisotope Complex at the Institute for Nuclear Research of the Russian Academy of Sciences in Moscow, has an alternative explanation. His calculations show that if an explosion damaged the housing of a nuclear reactor, rather than the core, and caused a leak of radioactive noble gases — which are a product of fission — then by the time the nuclei reached the detector in Severodvinsk they would have decayed to leave precisely the isotopes observed.