RheoMan: a five-year, ERC-funded (Advanced Grant), project to model the rheology of the Earth's mantle

Aug 17, 2016 Modelling plasticity in bridgmanite Results

In an article which just appeared in Earth and Planetary Science Letters, we present our latest results on dislocation mobility in MgSiO3 bridgmanite, and their implications for the deformation of bridgmanite at high pressure and high temperature.

The mobility of [100] and [010] screw dislocations is investigated at 30 and 60 GPa of pressure using numerical modelling. Two different features of these dislocations in MgSiO3 bridgmanite are calculated : the Peierls Potential, and the properties of kink pairs.      
The Peierls potentials of [100] screw dislocation in (010) and [010] screw dislocation in (100) are calculated with Nudged Elastic Band (NEB) calculations (fig. 1 (a)). A digest of the NEB method is available in section “More About”.   
We also model [100] and [010] screw dislocations dipoles presenting a kink pair (fig 1 (b)), respectively in (010) and (100) planes. Using these calculations, we are able to extract the energy of a single kink.

The results, obtained by numerical modelling, are combined with a continuous model, in order to characterize the thermally activated dislocation glide, controlled by the nucleation of kink pairs.



Figure 1 : (a) Peierls potential for [100] screw dislocation in (010) and [010] screw dislocation in (100) at 60 GPa of pressure. (b) Modeling of two kink-pairs in molecular statics calculation. 



Theses results allow to determine the stress required for the activation of dislocations in Bridgmanite, with respect to the temperature, and the characteristic of the deformation. The model can equally address two distinct cases : (i) a deformation in laboratory conditions characterized by a high strain-rate  (ii) a deformation at mantle conditions characterized by a very low strain-rate (about 10-16 s-1).        
The stress predicted by the model, to activate the [100](010) and [010](100) glide systems at laboratory conditions, are in excellent agreement with the stresses measured experimentally (figure 2). This article also presents the stress predicted at mantle conditions. These results allow to discuss the role of dislocation in the deformation of bridgmanite at mantle conditions.



Figure 2 : Stress required to activate [100](010) and [010](100) glide systems in laboratory conditions (curves), compared to stress measured in experiments.



To learn more:

    A. Kraych, Ph. Carrez & P. Cordier (2016) On dislocation glide in MgSiO3 bridgmanite, Earth and Planetary Science Letters, 452, 60-68, doi: 10.1016/j.epsl.2016.07.035 (open access)