A link between enhanced sampling of Ti–rich mantle garnets and superdeep diamonds: Insights from the DO-27 kimberlite, Northern Canada

Abstract

Despite the economic significance and scientific value of sublithospheric (superdeep) diamonds, useful indicators of their presence in a kimberlite body via traditional indicator mineral approaches remain elusive. The DO-27 kimberlite on the Slave craton hosts abundant (~ 25%) superdeep diamonds, providing an excellent opportunity to examine this issue. Elemental compositions plus thermobarometry of mantle garnets from DO-27 reveal a high fraction (~ 39%) of high-Ti garnets (classes G1 and G11) predominantly derived from the lowermost lithospheric root (~ 170 to 220 km). High-Ti garnets are significantly more abundant at DO-27 than in other kimberlites from the Slave craton where superdeep diamonds are scarce or absent. New density calculations show G11 garnet bearing peridotites are as buoyant as other parts of the lithosphere. Our model invokes the transport of superdeep diamonds to the base of the lithosphere via buoyant, deeply subducted harzburgites rising from transition zone depths. Once accreted to the lithosphere, these protoliths become re-fertilised by Ti–rich asthenosphere-derived melts during periods of melt-rock interaction preceding phases of kimberlite activity. Enhanced sampling of these accreted lithologies in the lowermost lithospheric mantle by kimberlite transports anomalously high proportions of high-Ti garnets along with superdeep diamonds, although there is no direct genetic link between the two. The association between a high proportion of high-Ti garnets and superdeep diamonds at DO-27 could potentially serve as a mineral indicator approach for the presence of superdeep diamonds in other kimberlites.