Modeling of Nonlinear Magnetics with A Multiscale, Unconditionally Stable Time-Domain Solver Unifying Electrodynamics, Elastodynamics and Spin Dynamics
Abstract
Thin-film magnetic materials, such as yttrium iron garnet (YIG), are naturally nonlinear and dispersive in microwave regime, which can be utilized for RF signal-processing components such as frequency-selective limiters (FSL) or signal-to-noise enhancers (SNE). The state of the art in modeling RF magnetic components is significantly lagging components consisting only electric material with most commercial software being limited to modeling simple, linear material behaviors which are not representative of material response in actual hardware. Moreover, the intrinsic material properties are critically important such as the excitation of the internal spin waves through the exchange of the magnetic domains. These are important mechanisms contributing to power-dependent insertion loss and frequency-selective limitation observed in FSL and SNE. Modeling of such effects in a comprehensive way requires unifying of governing physics from material level in nanometer scales to component level in centimeter scales, where rigorous modeling of such physics spanning six orders of magnitude is impossible even with the fastest supercomputer in the world. In this project, UCLA and UMN researchers team up to attack this grand challenge through development of a multi-physics, multi-scale time-domain solver that can model the seven orders of magnitudes scale difference from nanometers to centimeters, including the physics from electromagnetic waves, acoustic waves to the nonlinear generation of dipole-exchange spin waves into one unified framework. The ultimate goal of the proposed effort is to provide a comprehensive and precise modeling solution for nonlinear magnetics. Furthermore, the program focuses on providing a design kit that is physically solid, computationally efficient, and conveniently implementable to commercial simulation software such as ADS. This later is achieved through innovative interpretations of the motions of magnetic spins and their mutual coupling, as well as comprehensive and accurate mathematical representations of the underlying physics. A compact model will be delivered as the outcome of this project that is capable of predicting the performances of magnetic components using equivalent circuits. This model for the first time enables designers to develop circuit architectures that will optimize FSL and SNE performance, and enables system designers to determine the impact of these devices would have on the overall system. Moreover, the model translates the complex magnetic behaviors into equivalent electrical behaviors, which facilitates the understanding of micro-magnetics and allows invention of novel magnetic devices.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Oct 16, 2018
- Source ID
- W911NF1710100
Entities
People
- Yuanxun Wang
Organizations
- Army Contracting Command
- Defense Advanced Research Projects Agency
- University of California, Los Angeles