17 June 2005: Depleted mantle has been around since the start
Current models for the formation of the Earth's mantle are probably incorrect, report US geochemists. The researchers discovered a hitherto unsuspected difference in the abundance of two neodymium isotopes in meteorites and terrestrial rocks.
Neodymium (Nd) is a lanthanide metal with a molecular mass of 144. Two isotopes of neodymium are produced by the radioactive decay of two isotopes of samarium (Sm), another lanthanide metal. 147Sm decays to 143Nd, with a half-life of 106 billion years, while 146Sm decays to 142Nd, with a half-life of only 103 million years.
Scientists have used the decay of Sm to Nd to probe the geochemical evolution of the Earth, because both elements are present in the Earth's mantle and crust, but not in its metallic core. By measuring the levels of the different isotopes of Nd in rocks and lava, and comparing them with the levels found in chrondritic meteorites, which are thought to be the building blocks of the rocky planets, scientists have constructed models of how the Earth's interior has evolved over time.
In these models, the crust is generated by partial melting of the mantle, which gradually transfers Sm and Nd from the mantle to the crust, with Nd transferred faster than Sm. This is reflected in a mantle that is now relatively depleted in Nd and Sm, and where the ratio of 143Nd (from 147Sm) to 144Nd is higher than in chondritic meteorites. However, the ratio of 142Nd to 144Nd in the mantle and crust should remain the same as in chondritic meteorites, because all the 146Sm would have decayed to 142Nd prior to the start of the crust-forming process.
Using a thermal ionisation mass spectrometer, Maud Boyet and Richard Carlson from the Carnegie Institution in Washingon, DC, have now discovered that the ratio of 142Nd to 144Nd in chondritic meteorites is actually 20 parts per million lower than in terrestrial rocks. According to Boyet and Carlson, the most likely explanation for this finding is that the Earth's mantle split into two layers - a thin bottom layer enriched with Sm and Nd and a thick top one depleted in these elements - within just 30 million years of its formation.
'The calculations given in the paper suggest that a layer only 180km thick at the base of the mantle could be big enough, if sufficiently enriched in Sm and Nd, to explain the early depletion of the oceanic mantle,' said Boyet. Jon Evans