UCLA

Strain-mediated magnetoelectric composites have received considerable attention due to their multi-physics interaction. The ability to control magnetism at the nanoscale through strain induced electric field have made these composite more energy efficient relative to electric current methods. These strains can also be localized to address individual magnetostrictive nano-elements in an array. These feature have made magnetoelectric composites a potential candidate for magnetic-based devices like cell sorting and memory devices. The work presented in this dissertation provides pathways for these applications. The first part presents an alternative to capture and release of superparamagnetic (SPM) particles using FeGa microstructure on a piezoelectric PMN-PT substrate. This work is complemented with material characterization and micromagnetic-based modeling. The second part introduces a mathematical framework to model highly magnetostrictive thin-films under the influence of large residual stresses. This model was tested and validated with respect to polycrystalline Terfenol-D thin-film. The model also suggests a method to partially demagnetize a micro-scale single-domain, which can be of usefulness in cell sorting applications. The third part introduces a novel design of an epitaxial Terfenol-D memory bit on a piezoelectric substrate. This memory bit is actuated through localized electrodes which trigger an OOP deterministic clocking mechanism. This device can provide non-volatile data storage and energy efficient writing capability. Collectively, the findings in this dissertation are used to explore the influence of crystallinity in the magnetization process of highly magnetostrictive thin-films like FeGa and Terfenol-D. The mathematical framework presented in this dissertation expand on previous micromagnetic models to consider the influence of these nano-crystals. Such tools will be practical in future work involving the design and modeling of these magnetoelectric composites.