Published online 4 January 2001 | Nature | doi:10.1038/news010104-6


Earth scientists iron out their differences

Two groups of geochemists suggest elementary solutions to some mysteries of the Earth's mantle.

Most substances that make up the planet can be accounted for by standard theories of how the Earth formed, but some elements are just not where they're expected to be. In this week's Nature1, J. Wade and B. J. Wood of the University of Bristol suggest that they may have tracked down one of the errant elements.

They believe that niobium, which is not, as had been predicted, in the rocky mantle surrounding the Earth's molten core, but is instead lurking in the liquid iron at the centre. At the same time, researchers at the University of Bayreuth have unravelled the behaviour of other elements -- vanadium, chromium and manganese -- that don't always appear where they should.

Like a fine Scotch whisky, the provenance of the Earth can be discerned from the subtle blend of substances that it contains. Most of the solid Earth consists of just a handful of different elements. The hot core, about 5,600 kilometres across, is mostly iron. Surrounding this is the mantle and atop the mantle is the crust.

The Earth was created when large rocky blobs called planetesimals collided and coalesced in the early Solar System. The energy released by these impacts left most, perhaps all, of the planet molten, whereupon the rocky magma separated from the molten iron. The iron sank to the core, and the rock (mostly silicate) floated on top.

Some elements are more soluble in iron than in magma. These are called 'siderophiles', meaning iron-loving, and they are thought to have segregated to the Earth's core when the world was a young fiery ball. Because of this, the mantle is low in siderophile elements. This becomes clear if we compare the chemical composition of the mantle with those of a certain kind of meteorite called chondrites.

Chondrites are almost as old as the Earth. They are thought to be left-over lumps of the planetesimals from which the planets formed. But they did not segregate into iron-rich and silicate-rich regions, so their elemental abundances tell us what the Earth's total inventory of elements would be like if we could measure it.

Some siderophile elements, such as vanadium and chromium, are more abundant in chondrites than in the Earth's mantle, suggesting that they must indeed have flocked to the iron core. But other iron-lovers, such as calcium and scandium, are surprisingly abundant in the mantle, they are known to dissolve better in molten silicate than in iron. These odd elements are said to be 'refractory' lithophiles.

And then there is niobium. The chemical properties of this rare element mean that it should be a refractory siderophile -- there should be as much of it in the mantle as in a chondrite. But, mysteriously, there isn't.

Some geochemists have proposed that the 'missing' niobium is hidden away in a reservoir in a deep part of the mantle, like a pocket of cocoa in a poorly mixed chocolate cake. But it is hard to explain how such reservoirs could have arisen.

Deciding that most of what is known about niobium is assumption and analogy, Wood and Wade studied the element's behaviour experimentally, at the high temperatures and pressures characteristic of the deep Earth.

They report that, contrary to popular belief, niobium behaves more like the siderophiles vanadium and chromium than 'refractory' elements such as calcium. So, the researchers say, like the other siderophiles, niobium probably dissolved in the iron core.

In a similar set of experiments reported in Earth and Planetary Science Letters2, Christine Gessmann and David Rubie of the University of Bayreuth in Germany looked at the puzzling partitioning of vanadium, chromium and another siderophile, manganese.

These elements are not as under-represented in the mantle as earlier attempts to measure their solubility in iron suggest they should be. So Gessmann and Rubie took measurements at higher temperatures and pressures than before to get a better idea of how these elements might really have behaved in the young Earth.

They find that if the temperature is high enough (over 3,300 °C), the solubility of vanadium, chromium and manganese changes enough to solve the puzzle. Such temperatures could have been reached following a collision between the Earth and a large planetesimal -- the like of which is widely thought to have formed the Moon. 

  • References

    1. Wade,J. & Wood, B.J.The Earth's 'missing' niobium may be in the core. Nature 409,75 - 78 2001. | Article | PubMed | ISI | ChemPort |
    2. Gessmann,C.K.& Rubie, D.C.The origin of the depletions of V, Cr and Mn in the mantles of the Earth and Moon. Earth and Planetary Science Letters 184,95 - 107 2000. | Article | ISI | ChemPort |