Effect of mass disorder on the lattice thermal conductivity of MgO periclase under pressure

Abstract

Thermal conductivity of mantle materials controlling the heat balance and thermal evolution of the Earth remains poorly constrained as the available experimental and theoretical techniques are limited in probing minerals under the relevant conditions. We report measurements of thermal conductivity of MgO at high pressure up to 60 GPa and 300 K via diamond anvil cells using the time-domain thermoreflectance technique. These measurements are complemented by model calculations which take into account the effect of temperature and mass disorder of materials within the Earth. Our model calculations agree with the experimental pressure dependencies at 300 and 2000 K for MgO. Furthermore, they predict substantially smaller pressure dependence for mass disordered materials as the mechanism of scattering changes. The calculated thermal conductivity at the core-mantle boundary is smaller than the majority of previous predictions resulting in an estimated total heat flux of 10.4 TW, which is consistent with modern geomodeling estimates.

Hydration-reduced lattice thermal conductivity of olivine in Earth’s upper mantle

“…Such a large pressure-induced suppression of hydrous Fo90 lattice thermal conductivity is likely due to higherdensity ionic defects when compressed to higher pressures. Such weaker pressure dependence of thermal conductivity was also reported in MgO doped with ionic defects of iron at high pressures (26). At low-pressure regime, the replacement of cations (Si 4+ , Mg 2+ , and Fe 2+ ) in Fo90 with a small amount of H + does not substantially affect the heat capacity, phonon group velocities, and phonon scattering rates.…”

“…ref. 26), the more iron ions in the olivine, the stronger the phonon-defect scattering, leading to a lower lattice thermal conductivity. Upon compression, Λ anhy increases rapidly with increasing pressure to ∼11.6 (±2.3) W·m −1 ·K −1 at 13.1 GPa, a pressure near the 410-km discontinuity.…”

The curious case of cuprous chloride: Giant thermal resistance and anharmonic quasiparticle spectra driven by dispersion nesting

“…Calculated κ(P) of CuCl at RT is shown in Figure 5 with measured data 12 . For P>0.74 GPa, both calculated and measured κ(P) decrease with increasing P, an unusual behavior as κ typically increases 57,58 due to increasing acoustic velocities and reduced acousticoptic coupling as optic frequencies shift higher. Previously, decreasing κ(P) was attributed to increasingly negative TA Grüneisen parameters 12 , however, TA contributions to the calculated κ are minimal (inset to Fig.…”

Thermal conductance of strongly bonded metal-oxide interfaces

“…High pressure measurements ensure stiff interfaces with strong atomic bonds. 21 Additionally, the reduction in lattice constant and stiffening of elastic constants with increasing pressure 22 allow us to systematically study how G compares to max G across a range Our standard thermal model for interpreting TDTR data assumes that the laser energy is deposited at the metal film surface and that the intensity fluctuations of the probe beam are proportional to the metal film's surface temperature. 26 Both of these assumptions are invalid for TDTR measurements that use the thin SrRuO 3 films as the optical transducer because the optical penetration depth of SrRuO 3 is 50 nm at the pump/probe wavelength of 785 nm, 23 which is larger than the film thickness.…”