"The Characterization of Large Deformation Magnetorheological Elastomers" 

Paul Mitalski, MSE PhD Candidate 

Advisor: Professor Chelsea Davis

Zoom Link

ABSTRACT

Magnetorheological elastomers (MREs) are a class of smart materials consisting of magnetically responsive particles embedded in an elastomeric matrix. When subjected to an external magnetic field, the mechanical properties of MREs change, making them well-suited for applications including actuators, sensors, biomedical devices, and soft robotics. In recent years, research on MREs has expanded significantly, particularly in the characterization of their behavior under magnetomechanical loading and finite deformation. For this dissertation, the model material system of carbonyl iron particles in commercial silicone is used. This work first focuses on experimental methods, material characterization, and constitutive modeling to define the relationship between the mechanical response and externally applied magnetic fields. The hyperelastic behavior of MREs is investigated across varying magnetic field strengths and particle concentrations to capture and predict their nonlinear stress response. Novel experimental setups are developed to improve and advance MRE characterization. Moreover, predicted stress responses under magnetomechanical loading are evaluated using a calibrated multicomponent constitutive model across a range of particle concentrations. Recognizing that traditional hyperelastic constitutive models are limited to near equilibrium deformation conditions, the study then transitions to modeling rate-dependent behavior. A Deep Operator Network (DeepONet) is implemented to capture the finite strain rate dependency of the elastomer matrix. The neural network is trained over a broad range of strains and strain rates to predict both attributes of the testing system and the stress response of the material. Together, these contributions represent a step toward the general modeling and simulation of multiphysics, rate-dependent materials.