[Paper Review] Head-to-head domain walls in one-dimensional nanostructures: an extended phase diagram ranging from strips to cylindrical wires

S. Jamet, N. Rougemaille+2 more|arXiv (Cornell University)|Dec 1, 2014

Magnetic properties of thin films5 citations

TL;DR

This paper presents an extended phase diagram for head-to-head domain walls in one-dimensional magnetic nanostructures, unifying flat strips and cylindrical wires through symmetry and micromagnetic simulations. It classifies domain walls into two topological types—transverse-vortex and Bloch-point walls—showing that curling structures minimize dipolar energy, with analytical scaling laws confirming energy reduction trends across geometries.

ABSTRACT

So far magnetic domain walls in one-dimensional structures have been described theoretically only in the cases of flat strips, or cylindrical structures with a compact cross-section, either square or disk. Here we describe an extended phase diagram unifying the two pictures, extensively covering the (width,thickness) space. It is derived on the basis of symmetry and phase-transition arguments, and micromagnetic simulations. A simple classification of all domain walls in two varieties is proposed on the basis of their topology: either with a combined transverse/vortex character, or of the Bloch-point type. The exact arrangement of magnetization within each variety results mostly from the need to decrease dipolar energy, giving rise to asymmetric and curling structures. Numerical evaluators are introduced to quantify curling, and scaling laws are derived analytically for some of the iso-energy lines of the phase diagram.

Motivation & Objective

• To unify the description of head-to-head domain walls across one-dimensional magnetic nanostructures, from flat strips to cylindrical wires with compact cross-sections.
• To resolve naming inconsistencies and conceptual confusion between domain walls in strips and wires, particularly regarding vortex and Bloch-point walls.
• To establish a comprehensive phase diagram based on symmetry, phase transitions, and energy minimization, using micromagnetic simulations and analytical scaling.
• To identify the key physical mechanisms—especially dipolar energy reduction through curling and asymmetry—that govern domain wall structure across geometries.
• To demonstrate that Bloch-point walls, previously thought relevant only for wires, can also be ground states in thick strips, broadening their relevance for top-down fabricated devices.

Proposed method

• Employing symmetry and phase-transition arguments to classify domain walls into two topological varieties: transverse-vortex and Bloch-point walls.
• Conducting micromagnetic simulations across a wide (width, thickness) parameter space to map the phase diagram and identify first- and second-order transitions.
• Introducing numerical evaluators to quantify the degree of magnetization curling in domain walls, enabling comparison across geometries.
• Deriving analytical scaling laws for domain wall length and energy in flat strips and cylindrical wires, particularly for transverse-vortex and asymmetric transverse walls.
• Using the dipolar exchange length $ \sqrt{2A/\mu_0 M_s^2} $ as a critical length scale to determine when flux closure and curling become energetically favorable.
• Validating simulation results against analytical predictions, especially the $ R^2 $ scaling law for energy in cylindrical wires.

Experimental results

Research questions

• RQ1 How do the topological and energetic properties of head-to-head domain walls evolve continuously from flat strips to cylindrical wires with compact cross-sections?
• RQ2 What determines the transition between transverse-vortex and Bloch-point wall configurations in terms of geometry and energy minimization?
• RQ3 Why do some domain walls exhibit curling while others show asymmetry, and which configuration minimizes dipolar energy more effectively?
• RQ4 Under what geometric conditions do Bloch-point walls become the ground state, and how do they relate to structures fabricated via top-down methods?
• RQ5 Can analytical scaling laws accurately describe the energy and length of domain walls across different 1D nanostructures, particularly in cylindrical wires?

Key findings

• All head-to-head domain walls in 1D nanostructures fall into two topological classes: transverse-vortex walls and Bloch-point walls, based on symmetry and magnetization texture.
• Curling of the magnetization, involving both transverse and longitudinal components, is the dominant mechanism for reducing dipolar energy, especially in wires with non-circular or non-square cross-sections.
• The phase diagram reveals second-order transitions associated with the development of flux-closing textures, which occur when at least one transverse dimension exceeds seven times the dipolar exchange length.
• For cylindrical wires, the energy scales as $ R^2 $, and this scaling law is confirmed by micromagnetic simulations, indicating a robust energy reduction with increasing radius.
• Bloch-point walls are not exclusive to idealized wires; they can be the ground state in thick strips, suggesting broader relevance for top-down fabricated devices.
• Asymmetric transverse walls are energetically less favorable than curling structures except in specific cases of square-cross-section wires, where edge effects reduce the efficiency of curling.