Accurately dating the formation of Earth's core is key to understanding the birth and evolution of Earth and other planets in the Solar System. Until now, scientists have used a radioisotope dating technique based on the decay of uranium into lead to decipher the timeline of this event. But this method relies on assumptions that a team of petrologists and geochemists has now overturned.

Before Earth's metallic core formed, molten metals coexisted with less dense silicate-based materials in a fiery furnace in which pressures and temperatures were high. In a process thought to be similar to that by which rocks rich in iron are smelted to separate out impurities and produce metallic iron, these molten metals are thought to have migrated away to the centre of the Earth in the first 30 million to 40 million years after the formation of the Solar System.

Scientists previously believed that lead has an affinity for iron metals, and that as a result most of Earth's lead would have been whisked away to the planet's metallic core some 2,900 kilometres below the surface. This explained why a substantial amount of lead that was present during the early formation of the Solar System is now 'missing' from Earth's mantle.

Uranium, meanwhile, stayed behind and dissolved into the silicate portion of the evolving planet. The uranium includes two isotopes that decay to two different lead isotopes at specific, known speeds. As a result, or so the established dogma held, most of the lead now at the planet's surface should be a product of decayed uranium. Researchers therefore thought that they could determine the rough time at which Earth formed by calculating when the uranium first began to decay.

“Most people thought that core formation could explain the lead isotope signature of the Earth,” says Markus Lagos, a geochemist affiliated with the University of Bonn. But he and his co-workers now show that lead does not have a preference for metallic iron, but instead partitions with lighter silicate materials such as those seen in Earth's crust and mantle (see page 89).

The theory that the missing lead is in Earth's core had been around for some 40 years but had never been rigorously tested, says Lagos. So the team performed partitioning experiments with a mixture containing metals, sulphides and silicate material at temperature and pressure conditions that simulated those of early Earth. “We found that lead does not go into the metal phase,” says Lagos. “So whatever the lead isotopes tell us, it is not the timing of core formation.”

Although the team has resolved one key problem, a question still remains — where did the missing lead go? One possibility is that Earth's building materials could already have been depleted in lead. Another theory is that lead, which is quite volatile at high temperatures, may have been lost by degassing to the early atmosphere. And if Earth then collided with other planetary bodies early in the Solar System's evolution, the lead fingerprint may have been stripped from the atmosphere by the impact. Because it is commonly believed that the Moon formed through such a collision, the lead isotope signatures may instead date the formation of the Moon.