Magnetic Skyrmion Lattices and the Dzyaloshinskii-Moriya Interaction

The Dzyaloshinskii-Moriya interaction

The Dzyaloshinskii-Moriya interaction arises microscopically from spin-orbit coupling in magnetic systems that lack inversion symmetry, either because the crystal structure itself is chiral (as in the B20 compound MnSi) or because inversion symmetry is broken locally at an interface between a ferromagnet and a heavy metal with strong spin-orbit coupling (as in engineered Pt/Co/Ir or similar multilayer thin films). Unlike the familiar symmetric Heisenberg exchange, which is written as the dot product of two neighboring spins and simply favors parallel or antiparallel alignment, the DMI is written as a vector, the DMI vector, dotted into the cross product of two neighboring spins, and it explicitly favors spins tilting at a fixed chirality, a fixed sense of rotation, relative to one another, rather than simply aligning or anti-aligning. This is the crucial ingredient that twists what would otherwise be a uniform ferromagnetic state into a helical or spiral spin texture at zero field, and, in combination with an external magnetic field that tends to align spins uniformly, produces the delicate energetic balance from which localized skyrmions and skyrmion lattices emerge. Because the DMI has a fixed sign set by the underlying spin-orbit coupling and crystal or interface chirality, it also fixes the chirality of the resulting skyrmions, distinguishing so-called Bloch-type skyrmions, common in bulk chiral magnets, where the spins rotate in a plane tangential to circles around the skyrmion core, from Neel-type skyrmions, common in interfacial thin-film systems, where the spins rotate radially, pointing directly toward or away from the core as they wind.

Topological protection and the skyrmion number

The defining mathematical feature of a skyrmion is its topological charge, also called the skyrmion number, computed by integrating the local solid angle swept out by the magnetization direction over the two-dimensional plane, which for an ideal isolated skyrmion evaluates to exactly an integer, typically plus or minus one. This integer is a topological invariant: it can only change if the magnetization field passes through a singular, discontinuous configuration somewhere, since any smooth, continuous deformation of a texture with topological charge one necessarily still has topological charge one, no matter how much you locally distort it. Physically, unwinding a skymion back into the uniform background state therefore requires passing through such a singular point, which in a real magnetic material corresponds to a sharp, localized reversal of a single spin, an event that costs a finite and often substantial energy barrier. This is precisely what makes skyrmions robust: they resist being destroyed by thermal fluctuations, weak external perturbations, or interactions with material defects far more effectively than conventional, topologically trivial magnetic bubbles or domains, whose destruction pathway does not require crossing any such barrier. The size of this protecting energy barrier depends on the material's exchange stiffness, DMI strength, and anisotropy, and understanding and engineering it is central to skyrmion-based technology, since a barrier too low means unwanted thermal skyrmion creation and annihilation (data corruption in memory applications), while a barrier too high can make deliberately writing or erasing skyrmions on demand more difficult.

Phase diagram: helical, skyrmion lattice, and field-polarized states

The equilibrium magnetic phase of a chiral magnet depends sensitively on the ratio of DMI strength to exchange stiffness and on the strength of the applied external magnetic field, producing a rich phase diagram. At zero or very low field, the DMI-exchange competition typically stabilizes a helical spin state, a spontaneous, periodic spiral of spin orientation with a wavelength set by the ratio of exchange stiffness to DMI strength, and no applied field is needed to produce this periodic twisting. As an external field is applied and increased, the helical state is progressively distorted and, over an intermediate range of field strength, it can transition into the hexagonally close-packed skyrmion lattice phase, in which the energy cost of maintaining full helical order everywhere is outweighed by the benefit of aligning most of the sample with the field while still allowing the DMI to stabilize localized, topologically protected twists. At still higher field strength, the Zeeman energy dominates completely, the skyrmions are annihilated one by one or collectively, and the system enters the trivial field-polarized (uniformly aligned) state. In many bulk materials this skyrmion lattice phase pocket, sometimes visualized as a small closed region often nicknamed the 'A-phase' in the original MnSi studies, occupies only a narrow window of field and temperature, though interfacial and thin-film engineering, along with careful materials design, has substantially widened and in some systems even extended this stability window down to room temperature, a critical requirement for practical applications.

Skyrmion lattice formation and neutron scattering signatures

The self-assembly of individual skyrmions into a regular triangular lattice is itself a striking example of emergent order: identical, repulsively interacting particle-like textures naturally pack into the same close-packed hexagonal arrangement that minimizes energy for any set of mutually repelling objects in two dimensions, from vortices in type-II superconductors to colloidal particles. The direct experimental signature of this self-organized lattice is a beautiful six-fold symmetric diffraction pattern observed in small-angle neutron scattering, where neutrons scatter off the periodic magnetic texture and produce six sharp Bragg-like spots arranged in a hexagon in reciprocal space, directly reflecting the real-space triangular skyrmion lattice, first reported for bulk MnSi in a landmark 2009 study that established the skyrmion lattice as a genuine thermodynamic phase of matter rather than a metastable artifact. Complementary real-space imaging techniques, particularly Lorentz transmission electron microscopy and spin-polarized scanning tunneling microscopy, have since directly visualized individual skyrmions and their lattice arrangement in both bulk crystals and engineered thin films, confirming the characteristic swirling spin texture and its close-packed hexagonal ordering, and enabling detailed studies of lattice defects, dynamics, and the nucleation and annihilation of individual skyrmions in real time.

Skyrmions for spintronic memory and logic

The technological appeal of magnetic skyrmions rests on several combined advantages over conventional magnetic domain walls used in existing racetrack-memory concepts. Because skyrmions are topologically protected, they are intrinsically more stable against thermal noise and pinning at material defects than domain walls, which can be easily distorted or destroyed. Because they are typically nanometers to tens of nanometers in size, far smaller than typical domain-wall-based bit spacing, they promise much higher data storage density. Perhaps most importantly, skyrmions can be displaced by spin-polarized electric currents at current densities several orders of magnitude lower than what is required to move conventional domain walls, thanks to an efficient coupling between the conduction electron spins and the skyrmion's topological spin texture, described by an emergent electromagnetic-like force sometimes called the topological Hall effect, in which the skyrmion's winding spin texture acts on passing conduction electrons much like a real magnetic field acts on a charged particle. This combination of stability, small size, and low-current mobility underlies proposed skyrmion racetrack memory, where a train of skyrmions is shifted along a nanowire past a read/write head, as well as proposed skyrmion-based neuromorphic and unconventional computing schemes that exploit their particle-like dynamics, interactions, and even skyrmion-antiskyrmion pair creation and annihilation as computational primitives.

Frequently asked questions

Try it live

Everything above runs in your browser — open Magnetic Skyrmion Lattices and the Dzyaloshinskii-Moriya Interaction and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.