MR23B-2665
Modeling plasticity of MgO at the mesoscale using 2.5D Dislocation Dynamics.

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Riccardo Reali, Patrick Cordier, Philippe Carrez and Karine Gouriet, University of Lille 1, Villeneuve d'Ascq, France
Abstract:
In the lower mantle, viscosity results from the rheological behavior of the two main constituent

minerals, namely (Mg,Fe,Al)SiO3bridgmanite and (Mg,Fe)O ferropericlase. Understanding how

these phases deform is thus of primary importance in geophysics. This is also a very challenging

task, since the extreme conditions to which the lower mantle aggregate is subjected are not

reachable in laboratory experiments.

In this study, the contribution of dislocations to the deformation of periclase at the mesoscale is

investigated by Dislocation Dynamics (DD) simulations, a modeling tool which considers the

collective motion and interaction of dislocations. Dislocations are expected to be one of the most

efficient strain producing mechanisms. To model their behavior a so-called 2.5D DD approach is

employed. Within this method, dislocations are considered as straight segments perpendicular to a

2D reference plane and local rules are added to mimic 3D behavior [1]. Furthermore, both the glide

and climb mechanisms can be taken into account [2].

Before simulating the deformation of MgO under P, T and strain rate conditions of the lower

mantle, it is necessary to benchmark the model at ambient pressure, in order to compare the

simulated behavior with experiments performed in the same conditions.

At high temperatures (1500-1900 K) the strain-controlling mechanism results from the interactions

between dislocations. In this regime the influence of climb may be important: to investigate the

competition between glide and climb mechanisms, creep simulations in pure glide conditions were

performed in a wide range of temperatures and applied stresses and compared to simulations where

climb is explicitly included. Power law creep parameters are evaluated and compared with

experimental data.

[1] D. Ǵomez-Garćıa, B. Devincre, and L. P. Kubin, Phys. Rev. Lett. 96, 125503 (2006).

[2] F. Boioli, P. Carrez, P. Cordier, B. Devincre, and , M. Marquille, accepted Phys. Rev. B (2015).