MODELING MAGNETO-ACTIVE ELASTOMERS USING THE FINITE ELEMENT METHOD

Open Access
Author:
Sheridan, Robert Michael
Graduate Program:
Mechanical Engineering
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
July 16, 2014
Committee Members:
  • Paris R Vonlockette, Thesis Advisor
Keywords:
  • Magnetoactive elastomer
  • finite element method
  • action origami
Abstract:
Magneto-active elastomers (MAE) are a new type of smart materials that consist of hard-magnetic particles such as barium ferrite in an elastomer matrix. Under the application of a uniform magnetic field, the MAE material undergoes large deformation as the material bends due to magnetic torques acting on the distribution of hard-magnetic particles. This behavior demonstrates the potential of MAEs to act as remotely-powered actuators. MAEs are different from magnetorheological elastomers (MRE) which use soft-magnetic iron particles in place of the hard-magnetic particles. Traditional MREs, formed with carbonyl iron, are primarily driven by magnetic interactions between particles that cause magnetostriction and associated phenomena. In this work, MAEs were fabricated using 30% v/v 325 mesh M-type barium ferrite (BaM) particles in Dow Corning HS II silicone elastomers. Prior to curing, the samples were placed in a uniform (~2 Tesla) magnetic field to align the magnetic particles and produce a magnetization oriented in the direction of the applied magnetic field. For this work, the magnetic particles were physically oriented either in-plane or through-plane. The geometries studied in this work consisted of a cantilever beam MAE specimen (where experimental data is collected in [1]), a symmetric two-segment and an asymmetric two-segment accordion structure, and a four-segment accordion structure. In the four-segment and both two-segment geometries, MAE patches were bonded to a magnetically inert polydimethylsiloxane (PDMS) substrate. The ability of these structures to exert load over a range of prescribed displacements highlights the ability of the MAE actuators to perform useful work. The structures studied in this work deformed in either a bending or folding mode. The data collected from the experiments included tip force vs. displacement vs. magnetic field for the cantilever geometry, axial force and bend angle vs. magnetic field for the two-segment geometries, and average bend angle vs. magnetic field for the four-segment geometry. Results show increases in the measured tip force, axial force, and bend angle versus the applied field for the aforementioned geometries. The experimental results were then compared to results from a finite element analysis (FEA). The FEA methodology, performed in the commercial FEA software package Comsol, employs the Maxwell stress tensor applied as a traction boundary condition at the interface between magnetic and non-magnetic domains. Results show fair to excellent agreement for all structures studied. Discrepancies between the experimental results and the simulations may be attributed to the inability to capture the actual distribution of the magnetizations of the particles, which are assumed perfectly aligned, in the finite element model; the assumption that the Maxwell surface stress accurately models the particle-field and particle-particle interactions of the MAE; or the failutre to capture any non-uniformities in the applied magnetic field. Future work will focus on using the platform developed in this work to optimize structure design with regard to the magnitude and direction of the MAE's magnetization, size of the active material, and placement of the active material.