talk
Diamond Geology and Exploration

Tracing the Formation and Abundance of Superdeep Diamonds

Tuesday, November 20, 2018 - 15:40 to 15:59 Theatre 1

Author(s)

M.E. Regier (Presenting)
Canadian Centre for Isotopic Microanalysis, University of Alberta

D.G. Pearson
Canadian Centre for Isotopic Microanalysis, University of Alberta

T. Stachel
Canadian Centre for Isotopic Microanalysis, University of Alberta

R.A. Stern
Canadian Centre for Isotopic Microanalysis, University of Alberta

J. Harris
Canadian Centre for Isotopic Microanalysis, University of Alberta

Super-deep diamonds from the transition zone and lower mantle are valuable targets for mining, as they are often large, gem-quality1 or ultra-valuable type IIb stones2. Hence, in mine prospects, it may become important to determine the various populations of sub-lithospheric diamonds. Unambiguously identifying a diamond’s depth of formation is difficult as some minerals can be indicative of various depth regimes (e.g. ferropericlase, Ca-walstromite, enstatite, clinopyroxene, coesite). Here, we use the oxygen isotope compositions of inclusions in Kankan diamonds from Guinea to distinguish between the various diamond-forming processes that happen at lithospheric, asthenospheric to transition zone, and lower mantle depths. In this way, we hope to establish a process by which isotope geochemistry can better constrain the populations of superdeep diamonds in kimberlites, and can assist in estimating a pipe’s propensity for large, valuable stones.

Oxygen isotopic analysis by secondary ion mass spectrometry (SIMS) is a high-precision technique that can track hydrothermal alteration that occurred at or close below the ocean floor. Our analyses of inclusions from Kankan diamonds demonstrate that garnets with 3-3.03 Si cations (pfu) have d18O that are well-constrained within the normal values expected for peridotitic and eclogitic inclusions, but that garnets with =3.04 Si cations (pfu) have consistently high d18O (median: 10‰) that slightly decreases with increasing Cr2O3. We interpret this signal as the reaction between a melted carbonate-rich oceanic slab and normal convecting asthenosphere3. In contrast, retrogressed, or former, bridgmanite has d18O values similar to primitive mantle, suggesting little involvement of slab melts.

In contrast to the worldwide suite of lithospheric inclusions of eclogitic paragenesis (median d18O of 7.03‰)4,5, diamonds derived from ~250 to 500 km have inclusions with consistent, extremely high oxygen isotopes (median: 9.32‰)6,7, due to the melting of extremely enriched carbonated oceanic crust. Diamonds from the lower mantle, however, have inclusions with primitive mantle oxygen isotopes, suggesting a different formation process. The clear distinction in inclusion d18O between lithospheric, asthenospheric to transition zone, and lower mantle diamond populations is useful in informing the depth regime of a suite of stones, especially those with inclusions of ambiguous depths (e.g. clinopyroxene, coesite, Ca-walstromite, enstatite, ferropericlase, etc.). For instance, we are currently searching for exotic oxygen isotopes in ferropericlase that indicate asthenospheric diamond growth, rather than the primitive mantle values expected for lower mantle ferropericlase. In conclusion, oxygen isotopic analyses of diamond inclusions can identify various sublithsopheric diamond populations, and may benefit the assessment of a mine’s potential for large gem-quality, or type IIb diamonds.

1. Smith, E. M. et al. Science 354, 1403–1405 (2016). 2. Smith, E. M. et al. Nature 560, 84–87 (2018). 3. Thomson, A. R. et al. Nature 529, 76–79 (2016). 4. Ickert, R.B. et al. EPSL, 364, 85-97 (2014). 5. Zedgenizov, D. et al. Chemical Geology, 422, 46-59 (2016). 6. Ickert, R.B. et al. GPL, 1, 65-74 (2015). 7. Burnham, A.D. et al. EPSL, 432, 374-380 (2015).