1,720,974 research outputs found
Multiphysics simulations of magnetic nanostructures
No full textMultiphysics simulations of magnetic nanostructures by Matteo Franchin.In recent years the research on magnetism has seen a new trend emerging, characterised by considerable effort in developing new nanostructures and finding new ways to control and manipulate their magnetisation, such as using spin polarised currents or light pulses. The field of magnetism is thus moving towards the multiphysics direction, since it is increasingly studied in conjunction with other types of physics, such as electric and spin transport, electromagnetic waves generation and absorption, heat generation and diffusion. Understanding these new phenomena is intriguing and may lead to major technological advances. Computer simulations are often invaluable to such research, since they offer a way to predict and understand the physics of magnetic nanostructures and help in the design and optimisation of new devices.For the preparation of this thesis the Nmag multiphysics micromagnetic simulation package has been further developed and improved by the author. The software has also been extended in order to model exchange spring systems. Using Nmag, we carried out micromagnetic simulations in order to characterise the magnetisation dynamics in exchange spring systems and derived analytical models to validate and gain further insight into the numerical results. We found that the average magnetisation moves in spiral trajectories near equilibrium and becomes particularly soft (low oscillation frequency and damping, high amplitude) when the applied field is close to a particular value, called the bending field.We studied spin transport in exchange spring systems and investigated new geometries and setups in order to maximise the interaction between spin polarised current and magnetisation. We found that by engineering a trilayer exchange spring system in the form of a cylindrical nanopillar, it is possible to obtain microwave emission with frequencies of 5-35 GHz for applied current densities between 0.5-2.0 x 1011 A/m2 and without the need for an externally applied magnetic field. We proposed a one dimensional analytical model and found a formula which relates the emission frequency to the geometrical parameters and the current density
Ocaml Performance In Computational Science: Some Data And Code; Supplement For Publication
No full textSupplement for manuscript "Nmag micromagnetic simulation tool -- software engineering lessons learned" by Hans Fangohr, Maximilian Albert and Matteo Franchin (2016).
This repository contains some test programs that help investigating the performance of code emitted by the OCaml compiler (ocamlopt) for some of simple cases which are particularly relevant to numerical code. The tests are provided in separate directories. Each directory contains the OCaml version and the C++ version of the same test. Here is a list of the available directories:
multidim-arrays: test accesses to multi dimensional arrays implemented (i) as regular OCaml arrays and (ii) using the Bigarray OCaml module.
array-sum: test addition of two large unidimensional arrays.
summation: test computation with low memory bandwidth requirements
See the README files in the individual subdirectories for more details on the investigations performed and results obtained.</span
Domain wall motion in perpendicular anisotropy nanowires with edge roughness
No full textWe study field-driven domain wall (DW) motion in nanowires with perpendicular magnetic anisotropy using finite element micromagnetic simulations. Edge roughness is introduced by deforming the finite element mesh, and we vary the correlation length and magnitude of the roughness deformation separately. We observe the Walker breakdown both with and without roughness, with steady DW motion for applied fields below the critical Walker field Hc, and oscillatory motion for larger fields. The value of Hc is not altered in the presence of roughness.The edge roughness introduces a depinning field. During the transient process of depinning, from the initial configuration to steady DW motion, the DW velocity is significantly reduced in comparison to that for a wire without roughness. The asymptotic DW velocity, on the other hand, is virtually unaffected by the roughness, even though the magnetization reacts to the edge distortions during the entire course of motion, both above and below the Walker breakdown.A moving DW can become pinned again at some later point ('dynamic pinning'). Dynamic pinning is a stochastic process and is observed both for small fields below Hc and for fields of any strength above Hc. In the latter case, where the DW shows oscillatory motion and the magnetization in the DW rotates in the film plane, pinning can only occur at positions where the DW reverses direction and the instantaneous velocity is zero, i.e., at the beginning or in the middle of a positional oscillation cycle. In our simulations pinning was only observed at the beginnings of cycles, where the magnetization is pointing along the wire.The depinning field depends linearly on the magnitude of the edge roughness. The strongest pinning fields are observed for roughness correlation lengths that match the domain wall width
Enhanced spin wave propagation in magnonic rings by bias field modulation
No full textWe simulate the spin wave (SW) dynamics in ring structures and obtain the ω k dispersion relations corresponding to the output waveguide. Different bias field con-figurations affect the transfer of SW power from one arm of the structure to the other arm. To this end, we show that circular or radial bias fields are more suitable for energy transfer across the ring than the conventional horizontal bias field Hx. The SW dispersion shows that modes excited, when the bias field is along the ring radius, are almost 10 dB higher in power when compared to the modal power in the case of Hx. This is also corroborated by the SW energy density in the receiving stub. <br/
Nmag micromagnetic simulation tool – software engineering lessons learned
No full textWe review design and development decisions and their impact for the open source code Nmag from a software engineering in computational science point of view. We summarise lessons learned and recommendations for future computational science projects. Key lessons include that encapsulating the simulation functionality in a library of a general purpose language, here Python, provides great flexibility in using the software. The choice of Python for the top-level user interface was very well received by users from the science and engineering community. The from-source installation in which required external libraries and dependencies are compiled from a tarball was remarkably robust. In places, the code is a lot more ambitious than necessary, which introduces unnecessary complexity and reduces maintainability. Tests distributed with the package are useful, although more unit tests and continuous integration would have been desirable. The detailed documentation, together with a tutorial for the usage of the system, was perceived as one of its main strengths by the community
Micromagnetic simulations of magnetoelectric materials
Full text linkSome magnetic materials show a magnetoelectric coupling between inhomogeneous magnetization patterns and electric polarization that is sufficiently strong to allow external control of magnetization structure by electric fields. Numerical simulations of “magnetoelectric” materials of this type require an extension of the standard micromagnetic model which conceptually parallels the introduction of spin-current interaction terms. We show how the micromagnetic simulator “Nmag” can be extended to support the inhomogeneous magnetoelectric interaction term and also give a simple self-contained example for simulating the micromagnetic dynamics of a magnetoelectric system in the presence of an external electric field.<br/
Parallel execution and scriptability in micromagnetic simulations
Full text linkWe demonstrate the feasibility of an ‘encapsulated parallelism’ approach towards micromagnetic simulations that combines offering a high degree of flexibility to the user with the efficient utilization of parallel computing resources.While parallelization is obviously desirable to address the high numerical effort required for realistic micromagnetic simulations through utilizing now widely available multiprocessor systems (including desktop multicore CPUs and computing clusters), conventional approaches towards parallelization impose strong restrictions on the structure of programs: numerical operations have to be executed across all processors in a synchronized fashion. This means that, from the user’s perspective, either the structure of the entire simulation is rigidly defined from the beginning and cannot be adjusted easily, or making modifications to the computation sequence requires advanced knowledge in parallel programming.We explain how this dilemma is resolved in the Nmag simulation package in such a way that the user can utilize without any additional effort on his side both the computational power of multiple CPUs and the flexibility to tailor execution sequences for specific problems: simulation scripts written for single processor machines can just as well be executed on parallel machines and behave in precisely the same way, up to increased speed. We provide a simple instructive magnetic resonance simulation example that demonstrates utilizing both custom execution sequences and parallelism at the same time. Furthermore, we show that this strategy of encapsulating parallelism even allows to benefit from speed gains through parallel execution in simulations controlled by interactive commands given at a command line interface
Proposal for a standard micromagnetic problem: spin wave dispersion in a magnonic waveguide
Full text linkIn this paper, we propose a standard micromagnetic problem, of a nanostripe of permalloy. We study the magnetization dynamics and describe methods of extracting features from simulations. Spin wave dispersion curves, relating frequency and wave vector, are obtained for wave propagation in different directions relative to the axis of the waveguide and the external applied field. Simulation results using both finite element (Nmag) and finite difference (OOMMF) methods are compared against analytic results, for different ranges of the wave vector
Role of boundaries in micromagnetic calculations of magnonic spectra of arrays of magnetic nanoelements
No full textWe have used micromagnetic simulations performed with open and periodic boundary conditions to study the influence of the presence of array boundaries on the spectra and spatial profiles of collective spin wave excitations in arrays of magnetic nanoelements. The spectra and spatial profiles of collective spin waves excited in isolated arrays of nanoelements and those forming a part of quasi-infinite arrays are qualitatively different even if the same excitation field is used in the simulations. In particular, the use of periodic boundary conditions suppresses the excitation of non-uniform collective modes by uniform excitation fields. However, the use of non-uniform excitation fields in combination with periodic boundary conditions is shown to enable investigation of the structure of magnonic dispersion curves for quasi-infinite arrays (magnonic crystals) in different directions in the reciprocal space and for different magnonic bands. The results obtained in the latter case show a perfect agreement with those obtained with the dynamical matrix method for infinite arrays of nanoelements of the same geometry and magnetic properties
- …
