THERMAL CONDUCTIVITY OF LOWER MANTLE MINERALS

Please join us for a graduate student seminar this Friday March 16 at 3:30 pm in rm 155 Geology presented by David Mensah, MSc candidate:

THERMAL CONDUCTIVITY OF LOWER MANTLE MINERALS

Thermal conductivity calculations of the Earth’s lower mantle have proven to be a difficult task to perform owing to the extreme temperature and pressure conditions, which also make experimental measurements almost impossible. Therefore, thermal conductivity of the Earth’s lower mantle is usually inferred by extrapolating limited data obtained at low pressure-temperature conditions. Measurements at low temperatures and pressures are extrapolated to lower mantle conditions, but such extrapolations are dependent on various assumptions and physical models. These tend to produce a myriad of discrepancies in recent evaluations for the lattice thermal conductivity of the major lower mantle minerals such as iron (Fe)-bearing bridgmanite (MgSiO3) and periclase (MgO). These discrepancies in turn lead to uncertain estimations and even speculations when calculating the thermal conductivity of the lower mantle.

There is therefore the need for further experimental or theoretical work at different P-T conditions pertaining to the lower mantle as well as using different methods. This will help support or criticize existing models for the convection style of the mantle, which is based on estimated values to foretell possible dynamic structures. In this study, the lattice thermal conductivity of MgSiO3 bridgmanite has been calculated from first principles methods at three different pressures (0 GPa, 13 GPa and 40 GPa) and at 300 K, using molecular dynamics via the heat flux autocorrelation function (HCACF). We report lattice thermal conductivities of 25-30 W/m/K at 0 GPa, 160-200 W/m/K at 40 GPa and 900-1200 W/m/K at 13 GPa. Experimental studies showed that the lattice thermal conductivity can be reduced by as much as 50% when a substantial amount of iron is incorporated in bridgmanite (Manthilake et al., 2011). Applying this to our data obtained at 0 GPa further reduces it to 12.5-15 W/m/K, which is within the range of previous estimates at 5-30 W/m/K and is also comparable to the traditionally assumed thermal conductivity of the lower mantle (i.e. 10 W/m/K). Data obtained at 13 GPa and 40 GPa are tentative at best and must be reconsidered when further calculations become available.