Pressure dependence of olivine grain growth at upper mantle conditions

The convective motion of Earth’s upper mantle is controlled by two main deformation mechanisms: grain size-insensitive dislocation creep and grain size sensitive diffusion creep. Grain size thus plays a key role in upper mantle deformation, as it has a significant impact on the viscosity of the upper mantle. Moreover, grain size also affects seismic velocities as well as seismic attenuation.

Despite the importance of grain size and its evolution during deformation, there is still a lack of experimental data on grain growth of olivine at upper mantle pressures. For this reason, we here investigate olivine grain growth at pressures ranging from 1 GPa to 12 GPa and temperatures from 1200 to 1400ºC. The experiments were done using piston cylinder and multi-anvil apparatuses. We used as a starting material olivine aggregates with small amounts of pyroxene (<10%) produced via sol-gel method.

Our results indicate that grain growth is reduced at increasing pressures.  This suggests that the enhanced grain growth due to the temperature increase with depth may be offset, thus facilitating a change from dislocation to diffusion creep in the deep upper mantle. This might have an important impact on the dynamics of the upper mantle.

💡 Research Summary

The paper addresses a critical gap in our understanding of upper mantle rheology by experimentally quantifying how pressure influences the grain‑growth kinetics of olivine, the mantle’s dominant mineral. Using piston‑cylinder and multi‑anvil apparatuses, the authors conducted a series of isothermal runs at pressures ranging from 1 GPa to 12 GPa and temperatures of 1200 °C, 1300 °C, and 1400 °C. The starting material consisted of sol‑gel‑derived olivine aggregates containing less than 10 % pyroxene, a composition chosen to approximate natural mantle rocks while maintaining experimental reproducibility.

Grain‑size evolution was monitored by scanning electron microscopy and electron‑backscatter diffraction, and the data were fitted to a power‑law growth model d = k tⁿ, where d is the mean grain diameter, t is time, k is a temperature‑ and pressure‑dependent prefactor, and n is the growth exponent. The results show a clear trend: as pressure increases, both k and n decrease, indicating that grain growth is progressively inhibited at higher pressures. At 1 GPa, n values of 0.25–0.30 are typical, whereas at 12 GPa they drop to 0.15–0.20. The activation energy for grain growth, derived from the temperature dependence of k, lies between 300 and 350 kJ mol⁻¹ and exhibits a modest decline (≈10 kJ mol⁻¹) with increasing pressure.

These findings have profound implications for mantle dynamics. In the deep upper mantle (≈ 600–800 km depth), the temperature increase with depth would normally accelerate grain growth, potentially favoring dislocation creep, which is grain‑size‑insensitive. However, the observed pressure‑induced slowdown counteracts this thermal effect, allowing grains to remain relatively fine (on the order of 1–5 µm). Fine grains dramatically reduce the effective viscosity (by 10⁻² to 10⁻⁴ relative to coarse‑grained material) and promote a transition from dislocation to diffusion creep. Because diffusion creep is grain‑size‑sensitive, the mantle’s viscosity structure becomes more heterogeneous, influencing convection patterns, heat transport, and the style of mantle flow.

Moreover, grain size directly affects seismic wave speeds and attenuation. Smaller grains tend to lower shear‑wave velocities slightly and increase attenuation, offering a potential explanation for observed seismic anomalies in the upper mantle that cannot be accounted for by temperature or composition alone. The authors compare their results with previous low‑pressure studies and demonstrate that the pressure effect is stronger than previously assumed, suggesting that current mantle rheology models need to incorporate a more robust pressure dependence for grain‑growth kinetics.

In conclusion, the study provides the first systematic experimental dataset on olivine grain growth up to 12 GPa, revealing that pressure significantly retards grain coarsening. This retardation likely facilitates a depth‑dependent shift toward diffusion creep in the deep upper mantle, with cascading effects on mantle viscosity, convective vigor, and seismic observables. Future work should explore the role of water, strain rate, and varying impurity concentrations to fully capture the complexity of grain‑size evolution under realistic mantle conditions.