Research: High Pressure/Temperature Mineral Science – Mineral Properties to Planetary Processes

My research involves using large volume high-pressure/temperature apparatus to investigate mineral/melt properties under conditions of Earth’s deep interior. This information can then be used to understand large-scale planetary processes. I am also interested in developing novel experimental techniques as well as understanding on a fundamental level what effects pressure has on crystalline materials.

Current research interests:

Incorporation of water in 'dry' mantle minerals

Most of the so-called nominally anhydrous minerals (NAMs) that constitute Earth’s mantle actually contain trace amounts of water in the form of structurally-incorporated hydrogen. Even though the amount of ‘water’ stored in mantle minerals is only of the order of 100s to 1000s ppm H₂O by weight this constitutes a major terrestrial reservoir equivalent to several times the volume of water present in all of Earth’s oceans. The presence of small quantities of H in minerals also has a major effect on mineral and mantle properties such as rheology, melt relations and electrical conductivity. My interests lie in characterizing mechanisms of H incorporation in various minerals, understanding the influence that this has on mineral (and bulk mantle) properties, and in determining the role of H incorporation in NAMs in recycling ‘water’ back into the mantle during subduction of oceanic lithosphere. Most of the work I conduct involves synthesizing large, high-quality H-bearing crystals or annealing gem-quality samples under mantle conditions (high P-T), and studying H incorporation using polarized Fourier Transform Infrared spectroscopy or Ion microprobe (SIMS).

Figure 1. FTIR spectra from H-saturated stishovite synthesized at 15 GPa, 1500°C. Absorption is due to O-H stretching; the main feature is clearly resolved into two separate IR active bands, assigned to interstitial H coupled to an Al_Si^/ defect on the adjacent octahedral site and H uncoupled from other defects. H incorporation in stishovite provides a mechanism for transporting water deep into the lower mantle in subducting oceanic crust. Work that we have conducted in the low-pressure analogue TiO₂ system suggests that subsequent transformation of silica to the α-PbO₂ structure at the base of the lower mantle will result in complete H loss. Work such as this highlights the importance of understanding how hydrogen in minerals interacts with other defects. I have been developing novel techniques for trying to directly determine how hydrogen becomes 'coupled' to certain types of defect, and the effect that this has on hydrogen mobility. For example, in stishovite a proportion of the hydrogen is coupled to Al subsitutional defects (previously it had been assumed that all hydrogen would associate with Al defects).

Role of subduction in mantle-wide geochemical recycling
Subduction of oceanic lithosphere provides the main mechanism for recycling material back into Earth’s deep interior. I am interested in determining what role deep subduction of material (beyond the depths of sub-arc magmatism) plays in subsequent mantle processes such as large-scale melting and in volatile recycling. Geochemical evidence from ocean island basalts is consistent with the presence of ancient, subduction-modified oceanic crust in mantle source regions of intraplate magmatism. However, we still do not know what these 'signatures' actually represent. Is ancient subducted material trapped at seismic discontinuities (such as the core-mantle boundary) and entrained in upwelling plumes, or is material present as small scale heterogeneities present in a vigorously convecting mantle? I am interested in constraining the role that ancient crust plays in mantle melt processes, mainly by determining what effect the presence of ancient crust has on the major and trace element chemistry of subsequent mantle melts. A major aim of this work is to provide constraints on the role of subducted material during mantle melting; i.e. data that can be used to reassess geochemical 'signatures' in oceanic island basalts.

[figure 2]

Figure 2. Above left, BSE image (scale bar = 5µm) of sectioned capsule showing phase relations in a model subduction-modified crust at 8 GPa, 1650ºC. During low degree partial melts (up to at least 15 wt%) rutile or its high pressure polymorph TiO₂(II) is present in solid residues along with garnet and clinopyroxene. The presence of rutile during lower pressure melting of subducting crust results in characteristic Nb and Ta depletion, although this is not observed in ocean island basalts (OIB). The absence of Nb and Ta depletion is only consistent with direct melting of ancient crust as a source for OIB if degree of partial melting is sufficient to remove TiO₂ phases from solid residues, or if depth of partial melting is >300 km. Right, this is because pressure has a marked effect on the ability of TiO₂ phases to partition Nb (red) and Ta (blue). This can be ascribed to structural changes in silicate melts at elevated pressures, most notably an increase in proportion of ^[5]Si and ^[6]Si at the expense of ^[4]Si. The dominant influence that melt structure has on trace element partitioning suggests that relative enrichment of melts in Nb and Ta (e.g. decrease in Zr/Nb) is expected even when TiO₂ phases are not present. Understanding the important influence that pressure has on trace element behaviour might provide a means for constraining depth of melting in the mantle from melt geochemistry.

Deformation processes at high-pressure/temperature
The dynamics of Earth’s interior are strongly dependent upon the rheological properties of mantle minerals. I have developed a novel apparatus (with Simon Redfern from the University of Cambridge, and researchers at the Université P&MCurie), based on the Paris-Edinburgh cell, for studying the anelastic behaviour of materials under simultaneous high pressure and temperature conditions. Shear stress is applied to samples held under high-pressures (up to 12 GPa) and high temperatures (up to 2000°C) in this opposing anvil apparatus by rotating one the anvils with respect to the other. The advantage of this technique is that pressurization and deformation of samples are not coupled, meaning that large shear stresses and varying strain rates can be applied and accurately controlled. The apparatus is designed to be portable so that it can easily be transported to synchrotron and neutron beamlines for performing in-situ investigations. Initial testing of the apparatus involved performing a series of experiments deforming polycrystalline Zr at pressures up to 5 GPa. On-going work inolves performing in-situ experiments at ESRF.

[ropec3]

Figure 3.Cross section of rotating Paris-Edinburgh cell (roPEC) apparatus, based on a modified V7 450 tonne press frame. (A) hydraulic ram used to pressurise sample, (B) plate preventing rotation of upper (stationary) anvil. This plate is bolted to two of the tie rods, (C) housing for the lower (rotating) anvil, (D) central piston, (E) 3:1 steel-reinforced timing belt, (F) gear-head on externally-mounted servo, (G) externally mounted servo with Harmonic Drive© 100:1 reduction gearing on output, and lined to digital encoder, (H) breech with artillery thread, (I) spherical roller thrust bearings, (J) conical spacer, (K) input gear for reduction gearing, (L) support plate for servo and housing to reduction gearing, M) 161:1 Harmonic Drive© reduction gearing, (N) lower (rotating) carbide anvil, (O) upper (stationary) carbide anvil.

[graphite]

Right. Tomographic X-ray image of deformed sample, showing topology of the internal Au foil strain marker. This marker was originally a flat piece of foil, but has been 'twisted' during deformation of the sample. Sample volume (light grey area) is approximately 3.5 mm diameter, 2 mm in length. The dark grey area is the gasket material.