Computational Infrastructure for Geodynamics

Influence of initial preferred orientations on strain localisation in non-linear viscous anisotropic materials

by Tamara de Riese, Paul D Bons, Enrique Gomez-Rivas, Albert Griera, Maria-Gema Llorens and Ilka Weikusat

We present a series of numerical simulations which aim to investigate (1) the influence of an initial crystallographic preferred orientation on localisation behaviour in an anisotropic material and (2) the role of layering/passive markers on the development of deformation structures.

Cite as:

Riese, T. de ., Bons, P. D., Gomez-Rivas, E., Griera, A., Llorens, M.-G., & Weikusat, I.. (2020, July 20). CPO_StrainLocalisation.pdf (Version 1). Tectonics Community Science Workshop 2020. https://doi.org/10.6084/m9.figshare.12678296.v1

@article{deRiese2020,
author = "Tamara de Riese, Paul D Bons, Enrique Gomez-Rivas, Albert Griera, Maria-Gema Llorens and Ilka Weikusat",
title = "{CPO_StrainLocalisation.pdf (Version 1)}",
year = "2020",
month = "7",
url = "https://2020cigtectonics.figshare.com/articles/poster/CPO_StrainLocalisation_pdf/12678296",
doi = doi.org/10.6084/m9.figshare.12678296.v1}

COMMENTS

Posted by Pamela Burnley on Jul 28^th, 2020

Very interesting poster! I have a few questions. In your model set up, are there spatial domains defined for each grain, e.g. grain boundaries? It looks like you have properties assigned to points in the model, are those grain centroids? What would your model look like if you had no noise (the +-5 degrees). Also in the histograms of c-azimuth orientations, what do the green and black dots represent? Thanks!

Posted by Tamara de Reise on Jul 29^th, 2020

Dear Pamela, thanks for your questions!

each point (Fourier point) on the grid represents a crystallite/single grain with a constant internal crystal orientation. And initially each point is assigned a crystal orientation, and the anisotropy, which is defined as the ratio between the critical resolved shear stresses of basal and non-basal slip systems. Otherwise, I don't define grain boundaries or other parameters, it's a very simple setup.

I never did simulations without noise. I have to assign the three Euler Angles, and the FFT code does this for me. And it doesn't work when I use no noise. But I did some simulations with more noise, and then those deformation structures don't develop anymore. Instead, we get shear bands, which usually develop when starting with an initial random distribution.

Oh, and thanks for pointing out that I forgot describing that part. The black lines show the frequency distribution for the entire model box, and the green lines represent the c-axes orientations in the part of the model where the 10% highest von Mises strain rate are located.

Thanks!