Previous page     Contents     Next page

In this section an overview of the most important on-going projects is given. Information concerning recently-completed projects can be obtained from the publication lists of sections 4.1 and 4.2. Please note that the following contributions should not be cited.

3.1 Phase Transformations and Deformation of Mantle Minerals

Whereas the Earth's upper mantle (to a depth of 410 km) consists primarily of olivine, pyroxenes and garnet, the transition zone (410-660 km) is made up of wadsleyite or ringwoodite (the high pressure polymorphs of (Mg,Fe)₂SiO₄) and garnet, and the lower mantle consists mainly of (Mg,Fe)SiO₃ perovskite and magnesiowüstite. This mineralogically-layered structure is a result of high-pressure phase transformations which cause sharp increases in density, in particular at the depths of 410 and 660 km.

The structure and composition of the mantle is constrained by comparing seismic velocity measurements with elastic properties of candidate mineral phases. Elastic properties depend strongly on crystal structure and symmetry. Therefore, it is critical to understand how these properties vary with depth in the Earth. The first report of this section shows that garnets in the transition zone are likely to have cubic symmetry even though quenched high-pressure garnets can be tetragonal. This latter effect also depends on composition.

High-pressure phase transformations have an important influence on the dynamics of solid-state convection in the mantle because of the large density changes involved. If the kinetics of such transformations are sluggish due to low temperatures, as in subduction zones, the low-density phases may survive metastably to far greater depths than would be predicted from phase equilibria data. Olivine, for example, may survive to depths >600 km in subduction zones even though it should transform to wadsleyite at a depth of ~ 400 km under equilibrium conditions. The metastable persistence of low density phases in the deep mantle has consequences for the buoyancy forces which drive convection, for the state of stress in subducting lithosphere and is believed to be related to the origin of deep-focus earthquakes. Two reports in this section present new results on the mechanisms of high-pressure (Mg,Fe)₂SiO₄ phase transformations which will have important consequences for predicting transformation kinetics in the mantle. Previous kinetic models were based on the assumptions that the nucleation of new phases occurs only on grain boundaries and that the growth rate of the product phase is constant at a fixed pressure and temperature. The studies reported below show that both these assumptions can be invalid.

The rheology and mechanical properties of the mantle are poorly understood, even though they are of great importance for models of mantle convection. The rheology of olivine, the most abundant mineral in the upper mantle, has been extensively studied in the past and further new results are presented here. The way the presence of other minerals (pyroxenes and garnet) can affect upper mantle rheology is not well understood and is the topic of one of the reports in this section. To experimentally investigate the rheology of minerals which are stable in the transition zone and lower mantle is challenging and one solution is to study analogue materials (which are stable at low pressure) as described in the report of perovskite rheology included below. However, considerable progress is being made to develop experimental techniques for performing rheological studies at transition zone pressures (see 3.8d).