Encoding Fast and Fault-Tolerant Memories in Bulk and Nanoscale Amorphous Solids

We investigate the memory effects in cyclically deformed amorphous solids through computer simulations. Applying oscillatory shear deformations in all orthogonal directions during encoding creates fault-tolerant memories that are agnostic to the reading direction. Our extensive system size analysis shows that memory encoding is faster in small systems and becomes exceedingly challenging as systems approach the thermodynamic limit. To capitalize on the quickness of memory encoding in small system sizes, it is important to demonstrate that memory can be encoded in nanoscale objects, where open surfaces play a crucial role. We achieve this by going from 3D bulk to pseudo-1D nanorods. Using tension-compression cycles on these nanorods, we show that memory encoding and reading are also possible in the presence of free surfaces.