Advancing diamond exploration - novel techniques for the interpretation of diamond indicator minerals

Gregory M Yaxley1Hugh O'Neill1 and Andrew Berry2

1 Research School of Earth Sciences, Australian National University, Canberra, ACT 0200, Australia
2Department of Earth Science and Engineering, Imperial College, London, UK

 

Top left panel: Representative raw XANES spectra for synthetic garnets with Fe3+/∑Fe = 1.0, 0.46 and 0.01 (red, green, blue respectively - compositions known from stoichiometry of skiaigite - almandine - andradite series, kindly supplied by Prof. Alan Woodland, University of Frankfurt) showing Fe K-edge and near edge structure in inset. Data was collected on the Australian National Beamline Facility, KEK, Tsukuba, Japan; Bottom left panel: Expanded view of inset in A, showing pre-edge spectral features with backgrounds (dashed lines); Bottom right panel: Pre-edge peaks after background subtraction and best fit from fitting a number of Gaussian components; Top right panel: Pre-edge peak centroids as a function of Fe3+/∑Fe, showing linear correlation, the basis of the calibration.

We have developed new mineralogical tools applicable to the search for diamonds in Australia and overseas. This project was an ARC Linkage Project with industry partners BHP-Billiton, de Beers and Rio Tinto Exploration, through AMIRA International.

Techniques of high-pressure experimental petrology were used to develop mineral thermometers based on partitioning of Zn and Mn between upper mantle minerals. A new synchrotron-based technique for determining the redox state of the upper mantle is also being developed. These tools are being applied to fragments of garnet peridotite transported from the deep lithosphere to the surface by deeply-derived, occasionally diamondiferous kimberlite magmas. The resulting temperature and redox information will provide fundamental constraints on lithospheric diamond stability, and will be important in diamond exploration programs at a strategic level for targeting cratonic lithospheric domains more likely to contain high grades of diamonds, and at a more local level for targeting particular kimberlites for grade assessment by expensive bulk sampling techniques.

High-pressure experiments, focussing on Mn partitioning between garnet and olivine under upper mantle pressure-temperature conditions, have been completed. Algorithms for a Mn partitioning thermometers based on statistical fitting of experimental Mn-Mg and Fe-Mn exchange data between garnet and olivine have been determined. Experiments measuring Zn partitioning between Cr-spinel and olivine were also conducted and an olivine-spinel Zn-based thermometer was developed. These thermometers were applied to a comprehensive range of natural garnet ± spinel peridotite xenoliths samples from kimberlites erupted through the Kaapvaal and Slave Cratons. They performed extremely well in most cases, when compared with conventional thermometers (eg two-pyroxene thermometry etc).

The calibration of a synchrotron-based, X-ray Absorption Near Edge Structure Spectroscopy (XANES) method for determining Fe3+ contents of mantle garnets has proved unexpectedly challenging, despite a promising start (Figure 1), due to complexities relating to the compositional variations of natural garnets. We therefore adopted the alternative approach of obtaining the Fe3+ data by conventional Mössbauer Spectroscopy in collaboration with Prof Alan Woodland (Uni of Frankfurt) and by the newly developed electron microprobe based Flank Method in collaboration with Dr Heidi Höfer (Uni of Frankfurt). This will still allow determination of the oxygen-fugacity depth profiles through the lithospheric section represented by the garnet peridotite xenoliths supplied by the sponsor companies. These calculations are currently underway.