Finite Element Modeling of Crystallographic Preferred Orientation (CPO) in Two- Phase Aggregates: Rrelevant for Anisotropy of the Earth’s lower Lower Mantle

Abstract:

Deformation of lower mantle minerals has been studied experimentally in some detail, but the underlying physical processes that govern deformation are still not well understood. We have implemented the orthorhombic symmetry in the viscoplastic finite element code FEpX developed at Cornell University, making this code useful for understanding plasticity of mineral aggregates like bridgmanite (with perovskite structure) and ferropericlase, the main mineral components of the lower mantle.

High pressure radial diamond anvil cell (rDAC) and large volume press (D-DIA) experiments suggest that CPO is much weaker in 2-phase aggregates compared to 1-phase aggregates. But so far this has not been modeled satisfactorily with polycrystal plasticity theory that assumes that grains deform homogeneously, such as the Taylor or self-consistent approaches. Since it is expected that local heterogeneity plays a crucial role in deformation of 2-phase aggregates, we use a finite element approach where each grain in the aggregate is composed of individual elements. The constitutive equations between stress and strain are solved for all the elements in the aggregate, taking the local environment of each element into account. The FEpX code models the mechanical behavior of a virtual polycrystal, generated by the Neper code, which deforms by dislocation slip over large strain paths. The slip systems for the orthorhombic perovskite structure have been implemented into the code and, in particular, the development of CPO due to different combinations of slip systems is examined in the 2-phase aggregate of bridgmanite and periclase.

Preliminary results show that there is an important correlation between CPO of both bridgmanite and periclase with the relative volume fractions and the microstructure of the aggregate. If the soft periclase is interconnected it absorbs deformation by grain boundary sliding instead of by dislocation glide. Model results are compared with high pressure r-DAC experiments, where x-ray diffraction patterns are analyzed with the software MAUD to quantify CPO in the sample. Experimental CPO is compared with predictions from the FEpX model. With this approach we can draw conclusions about deformation processes in complex aggregates, which have a great significance in many geological processes.