Training Deflected States in 2D Magnetoelastic Sheets*

The concept of training is largely associated to living beings preparing themselves for a task in the near future. However, it is also possible to train inanimate matter. Trainable materials can be taught to which degree to alter their properties in response to environmental changes. To be able to "learn", materials must have reconfigurable internal degrees of freedom, and they must be able to retain these newly reconfigured states. Here, we use Molecular Dynamics simulations to study the mechanics and training of magnetoelastic sheets, composed of hydrocarbon-coated superparamagnetic nanoparticles and manipulated with magnetic fields. Experiments show that repeated deflection and heating cause the sheet to retain more and more deflection each cycle, resulting in a deflected trained state. Our MD simulations show how this training can result from microscopic configurational changes, suggesting a rearrangement of the hydrocarbons in the deflected state when heated. Due to the combined insights of simulation and experiments, magnetic nanoparticle sheets are an ideal platform for the exploration of general principles of mechanical training of materials.