Volatility and Fractionation of Nominally Refractory Elements | INSTITUT DE PHYSIQUE DU GLOBE DE PARIS


Aller au compte twitter

Research Departments

Cosmochemistry, Astrophysics and Experimental Geophysics

  Volatility and Fractionation of Nominally Refractory Elements

Sujet proposé pour 2015
Encadrant (et co-encadrant) : 

The major-element composition of the Earth is a result of the fundamental processes that
occurred very early in its history, roughly in the first 100 Ma. Accretion, giant impacts, core
formation and magma ocean crystallisation play major roles in isolating, volatilising, and
fractionating elements. The elemental and isotopic abundances of Si, Mg, Fe, Al, and Ca are known
accurately for the different meteorite groups (in particular the undifferentiated meteorites, called
chondrites) and to certain extent for the upper mantle by doing some assumptions, but they are only
postulated for the lower mantle and core. The lower-mantle and core are unsampled reservoirs, and
their compositions are usually modelled using a standard geochemical mass balance considerations
(BE= Bulk Earth, CHUR= chondrites, UM= upper mantle, LM=lower mantle, C=core):
As mentioned above, the composition of the reservoirs CHUR and UM have a known and it
is possible to measure their compositions, whereas the composition of LM and C are not known.
This is seemingly and open-ended problem as we have more unknowns than constraints. However,
another constraint can be brought by two other equations to close the problem, and they are based
on experimental determination of partition coefficients between liquid and solid silicates on the one
hand, and between metal and silicates on the other:
CUM/CLM=Dliq/sol CC/(CUM+CLM)=Dmet/sil
The experimental study of metal-silicate partitioning during core formation, as well as
liquid-solid partitioning during magma ocean crystallisation is therefore an essential tool to study
the global composition of the Earth.
However, one can always take the problem one step back; there is no fundamental reason
why the BE must be identical to the CHUR for all elements. In other words, there are reasons why
the Earth should not be chondritic (for all elements), the most important being the volatility. We
know that Earth is non chondritic in volatile elements, because these are volatilised during
accretion. We know that the Earth is chondritic for the very refractory elements like the REEs, Al,
Ca, Mg, as these are not fractionated in the upper mantle. But we do not know whether Si1 and Fe
are present in chondritic abundance. Could these be depleted in the BE by very high temperature
volatilisation during accretion?
We can reproduce and study high-temperature volatilisation occurring during accretion, in
the lab, by performing laser-levitation high-temperature experiments. The aim of this thesis is to
produce a series of devolatilised chondritic compositions (from the three different chemical groups:
carbonaceous, ordinary, enstatite) as well as a range of peridotitic/pyrolitic compositions. They will
be levitated at very high temperature and quenched to glasses. These will be analysed for the bulk
elemental and isotopic compositions. These will be studied as a function of temperature, time, and
bulk chemistry. The aim is to see variations in Si/Mg, Fe/Mg, as well as associated isotopic
fractionation in these elements, while maintaining Al/Mg and Ca/Mg contents in the starting
material; these could provide a paradigm shift in our understanding of the accreting Earth and its
major-element isotopic composition, especially for two of its most important elements: Silicon and
The work will follow a unique multidisciplinary approach (experimental petrology + isotope
geochemistry) and will tackle one of the most important problems in the planetary sciences.