arXiv Analytics

Sign in

arXiv:1905.09259 [cond-mat.soft]AbstractReferencesReviewsResources

Networks and Hierarchies: How Amorphous Materials Learn to Remember

Muhittin Mungan, Srikanth Sastry, Karin Dahmen, Ido Regev

Published 2019-05-22Version 1

We show how amorphous solids such as colloidal glasses and granular materials can remember complex shear deformation histories. The slow deformation of these systems is described through a sequence of discrete plastic rearrangements which we map onto directed graphs. The mapping reveals near-perfect hierarchies of hysteresis loops and hence near-perfect return point memory (RPM). For small to moderate deformation amplitudes, the plastic transitions can be traced back to localized and reversible rearrangements (soft-spots) that interact via Eshelby type deformation fields. We find that while the interactions between soft-spots determine the network topology, this happens in a way that RPM is retained to a large extent. Observing high quality RPM in spite of a violation of the no-passing property is surprising, because no-passing is usually seen as a condition for RPM. Since severe RPM violations are rare, memory can be stored in these systems and be read out with high fidelity.