DI008-10
A non-linear Clapeyron slope of the post-garnet transition determined by advanced multi-anvil and in-situ X-ray observation technologies, and its multiple effects on mantle dynamics in various tectonic settings
A non-linear Clapeyron slope of the post-garnet transition determined by advanced multi-anvil and in-situ X-ray observation technologies, and its multiple effects on mantle dynamics in various tectonic settings
Wednesday, 9 December 2020: 06:06
Virtual
Abstract:
The subduction of cold slabs and upwelling of hot plumes are two major modes of mantle convection causing activities at the Earth’s surface, such as volcanism and earthquakes, and the chemical evolution of the interior. Such dynamic motions are enhanced or impeded by the density changes of major mantle minerals at phase transformations, especially by bridgmanite-forming reactions and their temperature dependencies, namely, Clapeyron slopes. Determining the Clapeyron slopes is therefore essential to understanding mantle dynamics. Conventional experimental procedures have, however, hampered accurate determinations of Clapeyron slopes due to the sluggish kinetics of phase transitions and pressure changes upon heating. In this study, we determined the phase boundary of the post-garnet transition (breakdown of garnet to bridgmanite plus corundum) in Mg3Al2Si3O12 based on the strict definition of the phase equilibrium by means of advanced multi-anvil and in-situ X-ray diffraction techniques to eliminate these problems. The results demonstrate that the phase boundary is curved in a downward-convex manner: the Clapeyron slope changes from −2 MPa/K to +3 MPa/K with increasing temperature from 1450 K to 2100 K. The post-garnet transition therefore impedes slab subduction due to its negative slope at low temperatures, whereas it enhances hot plume upwelling due to its positive slope at high temperatures. The effects of the phase transition on mantle convection are therefore numerous. This is the first report of a Clapeyron slope varying with temperature in mantle phase transitions. All major mantle phase boundaries therefore need to be re-determined employing experimental procedures that are strictly based on the definition of the phase equilibrium.