The Dissolution of Orthopyroxene in Potential Proto-kimberlitic Melts between 2.5 and 6.0 GPa

English Abstract:

The composition of proto-kimberlitic melts has been the subject of much debate because of their heterogeneous nature, ubiquitous modification by contamination by mantle and crustal materials, high susceptibility to alteration during and after emplacement, and the fact that they do not quench to a glass. Recently Russell et al. (2012) proposed that proto-kimberlitic melts are originally carbonatitic. During ascent, the melt assimilates orthopyroxene, which causes the melt to evolve to more kimberlitic compositions and triggers massive exsolution of CO₂, explaining the rapid ascent of the kimberlite magma. However, the experiments by Russell et al. (2012) to support this model were performed at atmospheric pressures and used Na₂CO₃ as the model carbonatitic melt. How well these experiments simulate the conditions of a rising kimberlite melt in the mantle is an open question. Therefore we investigated the assimilation of orthopyroxene in a variety of carbonatitic melts in the system CaO-MgO-Al₂O₃-SiO₂-CO₂ ± H₂O at pressures between 2.5 and 6.0 GPa. These melts had been determined in previous studies to be in equilibrium with lherzolite assemblages (Opx + Cpx + Ol ± Grt) at pressures of 6, 6.9, and 10 GPa (Dalton and Presnall, 1998; Girnis et al., 2011; Gudfinnsson and Presnall, 2005; Keshav and Gudfinnsson, 2014), and with garnet and wadsleyite at 16.5 GPa (Ghosh et al., 2014). At 2.5 GPa the carbonatitic melt and orthopyroxene reacted to form olivine and diopside, exsolving CO₂; however, at pressures above the En + Dol = Fo + Di + CO₂ reaction boundary, orthopyroxene remained in equilibrium with the carbonatitic melt with no signs of assimilation into the melt, nor were there any signs of CO₂ exsolution. These results have major implications for the Russell et al. (2012) model for kimberlite formation, as well as for our understanding of the (in)stability of orthopyroxene in proto-kimberlitic melts.