Can primitive kimberlite melts be alkali‐carbonate liquids: Composition of the melt snapshots preserved in deepest mantle xenoliths

Abstract

AbstractThe study of kimberlite rocks is important as they provide critical information regarding the composition and dynamics of the continental mantle and are the principal source of diamonds. Despite many decades of research, the original compositions of kimberlite melts, which are thought to be derived from depths > 150 km, remain highly debatable due to processes that can significantly modify their composition during ascent and emplacement. Snapshots of the kimberlite‐related melts were entrapped as secondary melt inclusions hosted in olivine from sheared peridotite xenoliths from the Udachnaya‐East pipe (Siberian craton). These xenoliths originated from 180‐ to 220‐km depth and are among the deepest derived samples of mantle rocks exposed at the surface. The crystallised melt inclusions contain diverse daughter mineral assemblages (>30 mineral species), which are dominated by alkali‐rich carbonates, sulfates, and chlorides. The presence of aragonite as a daughter mineral suggests a high‐pressure origin for these inclusions. Raman‐mapping studies of unexposed inclusions show that they are dominated by carbonates (>65 vol.%), whereas silicates are subordinate (<13 vol.%). This indicates that the parental melt for the inclusions was carbonatitic. The key chemical features of this melt are very high contents of alkalis, carbon dioxide, chlorine, and sulfur and extremely low silica and water. Alkali‐carbonate melts entrapped in xenolith minerals likely represent snapshots of the primitive kimberlite melt. This composition is in contrast with the generally accepted notion that kimberlites originated as ultramafic silicate water‐rich melts. Experimental studies revealed that alkali‐carbonate melts are a very suitable diamond‐forming media. Therefore, our findings support the idea that some diamonds and kimberlite magmatism may be genetically related.