Phase stabilities of MgCO3 and MgCO3-II studied by Raman spectroscopy, x-ray diffraction, and density functional theory calculations

Carbonates are the major hosts of carbon on Earth's surface and their fate during subduction needs to be known to understand the deep carbon cycle. Magnesite (${\mathrm{MgCO}}_{3}$) is thought to be an important phase participating in deep Earth processes, but its phase stability is still a matter of debate for the conditions prevalent in the lowest part of the mantle and at the core mantle boundary. Here, we have studied the phase relations and stabilities of ${\mathrm{MgCO}}_{3}$ at these $P,T$ conditions, using Raman spectroscopy at high pressures ($\ensuremath{\sim}148\phantom{\rule{0.16em}{0ex}}\mathrm{GPa}$) and after heating to high temperatures ($\ensuremath{\sim}3600\phantom{\rule{0.16em}{0ex}}\mathrm{K}$) in laser-heated diamond anvil cell experiments. The experimental Raman experiments were supplemented by x-ray powder diffraction data, obtained at a pressure of 110 GPa. Density-functional-theory-based model calculations were used to compute Raman spectra for several ${\mathrm{MgCO}}_{3}$ high-pressure polymorphs, thus allowing an unambiguous assignment of Raman modes. By combining the experimental observations with the density-functional-theory results, we constrain the phase stability field of ${\mathrm{MgCO}}_{3}$ with respect to the high-pressure polymorph, ${\mathrm{MgCO}}_{3}$-II. We further confirm that Fe-free ${\mathrm{MgCO}}_{3}$-II is a tetracarbonate with monoclinic symmetry (space group $C2/m$), which is stable over the entire $P,T$ range of the Earth's lowermost mantle geotherm.