Ferroic Magnetic Octupole Order

• Ferroic magnetic octupole order is a rank-3 magnetic ordering defined by uniform octupole polarization with nearly zero net dipole magnetization, leading to hidden yet robust macroscopic effects.
• It is characterized experimentally via symmetry-sensitive probes such as MOKE, SHG, circular dichroism, and neutron scattering that reveal anomalous Hall, Nernst, and optical signals.
• Material realizations in systems like Mn3Sn, FeF2, and 5d double perovskites illustrate its role in ultrafast dynamics, Berry-curvature induced transport, and nonlinear magnetoelectric responses.

Searching arXiv for recent and foundational papers on ferroic magnetic octupole order. Searching for foundational work on Mn3Sn, FeF2 surfaces, altermagnets, and 5d double perovskites. Ferroic magnetic octupole order is a form of long-range magnetic ordering in which the primary order parameter is not a dipole magnetization but a magnetic multipole of rank three. In crystalline solids this order can be formulated either as a cluster multipole built from a non-collinear spin texture or as a bulk, gauge-invariant tensor quantity of the magnetization density. It is therefore the natural language for time-reversal-symmetry-breaking states with vanishing or nearly vanishing net magnetization, yet with robust anomalous transport, magneto-optical activity, nonlinear responses, and domain structure. Across antiferromagnetic Weyl semimetals, altermagnets, pyrochlore systems, and spin-orbit-coupled double perovskites, ferroic magnetic octupole order has emerged as the primary descriptor of magnetic phases that are “hidden” to conventional magnetometry but not to symmetry-sensitive probes and Berry-curvature-driven observables (Miwa et al., 2021, Sato et al., 30 Apr 2025).

1. Definition and formal description

A standard continuum definition writes the magnetic octupole tensor as

$$\mathcal{O}_{ijk} = \int d^3r\, \mu_i(\mathbf{r})\, r_j r_k,$$

where $\mu_i(\mathbf{r})$ is the $i$-component of the magnetization density. In this form the octupole is a rank-3 tensor, even under inversion because $r_j r_k$ is even, and odd under time reversal because $\boldsymbol{\mu}(\mathbf{r})$ changes sign. In periodic crystals an equivalent thermodynamic formulation defines the spin magnetic octupole as the conjugate variable to the second spatial derivative of the magnetic field,

$$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$

which avoids the gauge ambiguity of the position operator in the Bloch basis and yields a gauge-invariant bulk quantity (Bhowal et al., 2024, Sato et al., 30 Apr 2025).

The bulk octupole tensor can be decomposed into irreducible components. In the quantum theory for periodic crystals, the eighteen components of $M_{ijk}$ separate into a totally symmetric rank-3 octupole, a magnetic toroidal quadrupole, an ordinary magnetic dipole, and an anisotropic magnetic dipole. The anisotropic magnetic dipole,

$$M'_i \coloneqq \frac{3 M_{jji} - M_{ijj}}{\sqrt{10}},$$

has the same symmetry as conventional spin and orbital magnetic dipoles but carries no net magnetization. This result is central for antiferromagnets that display anomalous Hall conductivity without ferromagnetism: the octupole tensor contains dipolar-like irreducible content while remaining fundamentally a higher-rank order parameter (Sato et al., 30 Apr 2025).

In many materials the most practical description is not an atomic tensor but a cluster multipole. The archetypal example is Mn$_3$Sn, where the inverse triangular order on kagome triangles is treated as a cluster magnetic octupole. In that setting the long-range order is “ferroic” because equivalent clusters carry the same octupole polarization throughout the crystal, that is, a $Q=0$ order of magnetic octupoles (Miwa et al., 2021).

2. Symmetry, ferroicity, and relation to antiferromagnetism

The defining symmetry content of ferroic magnetic octupole order is the coexistence of broken time-reversal symmetry with vanishing or negligible net dipole magnetization. This is why octupolar phases are frequently described as hidden magnetic order. In centrosymmetric antiferromagnets, the zeroth-order coupling to a magnetic field vanishes because the net magnetization is zero, and the first-order coupling can also vanish because inversion symmetry forbids a bulk magnetic quadrupole. The magnetic octupole then becomes the lowest-rank allowed magnetic multipole (Ōiké et al., 2024, Sato et al., 30 Apr 2025).

“Ferroic” does not imply ferromagnetism. It means that the octupole polarization is uniform. In Mn$\mu_i(\mathbf{r})$0Sn, for example, the ferroic alignment of cluster octupoles breaks time-reversal symmetry macroscopically even though the spontaneous magnetization is only $\mu_i(\mathbf{r})$1 per formula unit and the spin magnetization from canting is negligible for the large Kerr and Hall signals. The octupole carries the same irreducible representation $\mu_i(\mathbf{r})$2 as spin magnetization and can therefore induce MOKE, anomalous Hall effect, and anomalous Nernst effect without conventional ferromagnetic order (Miwa et al., 2021, Higo et al., 2018).

A useful classification is given by magnetic point groups. Type-I groups allow magnetic dipoles, Type-II groups forbid dipoles but allow magnetic quadrupoles, and Type-III groups forbid both dipole and quadrupole while allowing magnetic octupoles. It is in these Type-III groups that octupole order is the lowest-rank magnetic order parameter and where responses such as the anomalous Hall effect are often symmetry-forbidden, while the nonlinear magnetoelectric effect and piezomagnetism remain allowed (Ōiké et al., 2024).

This symmetry logic also underlies altermagnetism. In $\mu_i(\mathbf{r})$3-wave altermagnets, the combination of antiferroic electric quadrupole order and antiferromagnetic spin order induces a ferroic magnetic octupole $\mu_i(\mathbf{r})$4, even in the absence of spin-orbit coupling. The resulting order breaks time reversal, keeps zero net magnetization, and produces nonrelativistic spin splitting with $\mu_i(\mathbf{r})$5-wave symmetry (Ko et al., 1 Aug 2025). A common misconception is therefore that antiferromagnets without net moment must be magneto-optically silent; the octupole framework shows that this is not generally true.

3. Material realizations and structural motifs

The current literature spans several distinct realizations of ferroic magnetic octupole order, including cluster-octupolar non-collinear antiferromagnets, collinear compensated antiferromagnets, altermagnets, and spin-orbit-coupled double perovskites. The unifying feature is that a rank-3 magnetic multipole, rather than a dipole, governs the macroscopic phase and its responses (Miwa et al., 2021, Bhowal et al., 2024, Pourovskii et al., 2021, Usachev et al., 16 Jun 2026, Han et al., 6 May 2025).

System	Octupolar description	Reported consequence
Mn$\mu_i(\mathbf{r})$6Sn	Ferroic order of cluster magnetic octupoles on kagome bilayers	Large AHE, ANE, MOKE; ultrafast octupole dynamics
FeF$\mu_i(\mathbf{r})$7	Bulk ferroically ordered magnetic octupoles in a compensated antiferromagnet	Emergent surface multiferroicity and surface linear magnetoelectric effect
Ba$\mu_i(\mathbf{r})$8OsO$\mu_i(\mathbf{r})$9 ($i$0 Ca, Mg, Zn)	Ferro-ordered octupoles within a low-energy $i$1 doublet	Single second-order transition and gapped magnetic excitations
CoF$i$2	Ferrotype magnetic octupole in a collinear altermagnet	Electric- and magnetic-dipole-forbidden SHG below $i$3 K
Eu$i$4Ir$i$5O$i$6	Ferroic octupole order of AIAO/AOAI domains in a Weyl semimetal	Zero-field CD $i$7 and Kerr effect $i$8 radians

Mn$i$9Sn remains the clearest metallic example. Its inverse triangular order is essentially pure octupolar order: cluster multipole analysis finds that the octupole contributes to the expansion by more than $r_j r_k$0. The non-collinear spin texture breaks time-reversal symmetry, stabilizes Weyl points, and induces large Berry curvature near the Fermi energy (Miwa et al., 2021).

FeF$r_j r_k$1 provides a complementary collinear and centrosymmetric setting. There the bulk is non-polar, non-magnetized, and has no bulk linear magnetoelectric effect, yet it hosts ferroically ordered magnetic octupoles. The surface inherits the bulk octupole and, because inversion is broken at the boundary, develops a linear magnetoelectric effect, a net surface magnetization, and a net surface electric dipole moment. The surface is therefore multiferroic even though the bulk is not (Bhowal et al., 2024).

In 5d$r_j r_k$2 double perovskites, advanced many-body first-principles calculations for Ba$r_j r_k$3OsO$r_j r_k$4 show that the ground state is formed by ferro-ordered octupoles coupled by superexchange interactions within the ground-state $r_j r_k$5 doublet. This result was advanced precisely to resolve the debate between Jahn–Teller-distorted quadrupolar order and octupolar order in these materials (Pourovskii et al., 2021). By contrast, Sr$r_j r_k$6MgReO$r_j r_k$7 realizes antiferro octupolar order rather than ferroic order, but its spectroscopic signatures clarify how octupolar phases differ from conventional dipolar magnets (Mosca et al., 31 Mar 2025).

Pyrochlore and altermagnetic examples broaden the structural scope. In Eu$r_j r_k$8Ir$r_j r_k$9O$\boldsymbol{\mu}(\mathbf{r})$0, the all-in–all-out and all-out–all-in antiferromagnetic states are ferroic octupole domains of a magnetic Weyl semimetal (Han et al., 6 May 2025). In CoF$\boldsymbol{\mu}(\mathbf{r})$1, the antiferromagnetic phase below $\boldsymbol{\mu}(\mathbf{r})$2 K is described by a ferrotype magnetic octupole $\boldsymbol{\mu}(\mathbf{r})$3, which acts as the order parameter of a collinear altermagnet (Usachev et al., 16 Jun 2026).

4. Experimental signatures and domain imaging

The experimental challenge is that the primary order parameter is high-rank. Conventional magnetization measurements are often weak or ambiguous, whereas probes that couple to symmetry, Berry curvature, or multipolar form factors become decisive. The most direct examples are magneto-optical Kerr effect, circular dichroism microscopy, second-harmonic generation, and neutron scattering at large momentum transfer (Higo et al., 2018, Han et al., 6 May 2025, Usachev et al., 16 Jun 2026, Mosca et al., 31 Mar 2025).

Mn$\boldsymbol{\mu}(\mathbf{r})$4Sn established the metallic benchmark. Despite a vanishingly small magnetization of $\boldsymbol{\mu}(\mathbf{r})$5, it exhibits a large zero-field MOKE with a polar Kerr rotation angle of $\boldsymbol{\mu}(\mathbf{r})$6 milli-degrees at room temperature, comparable to ferromagnetic metals. First-principles calculations for the fully compensated antiferromagnetic state reproduce a large Kerr spectrum, and varying the small net magnetization produces almost no change in the calculated Kerr angle. This large MOKE made it possible to image magnetic octupole domains and their field-induced reversal directly (Higo et al., 2018).

Eu$\boldsymbol{\mu}(\mathbf{r})$7Ir$\boldsymbol{\mu}(\mathbf{r})$8O$\boldsymbol{\mu}(\mathbf{r})$9 extends domain imaging into a topological antiferromagnet with vanishing net moment. Optical circular dichroism microscopy resolves AIAO and AOAI octupole domains below the Néel temperature. The reported zero-field signals, $$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$0 in circular dichroism and $$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$1 radians in Kerr effect, were attributed to Berry-curvature effects from Weyl nodes rather than to a detectable magnetic moment (Han et al., 6 May 2025).

In CoF$$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$2, ferroic-octupolar order is probed by electric- and magnetic-dipole-forbidden SHG. The observed nonlinear polarization is

$$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$3

and the temperature dependence of SHG reveals the transition at $$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$4 K. This result is important because it provides a direct nonlinear-optical probe of a collinear altermagnetic octupole order parameter in a centrosymmetric lattice (Usachev et al., 16 Jun 2026).

Neutron spectroscopy gives a different window. In antiferro octupolar Sr$$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$5MgReO$$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$6, the ordered state yields quasi-gapless magnetic excitation spectra and superstructural neutron diffraction reflexes peaking at large scattering momenta, rather than the low-$$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$7 behavior characteristic of dipolar order (Mosca et al., 31 Mar 2025). This does not by itself establish ferroic octupole order, but it clarifies the generic spectroscopic signature of rank-3 magnetism: octupolar form factors can shift spectral weight toward larger $$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$8.

5. Transport, dynamics, and nonlinear responses

One of the strongest motivations for the octupole framework is that it unifies equilibrium order with nonequilibrium response. In Mn$$M_{ijk} \coloneqq -\left( \frac{\partial F}{\partial[\partial_{r_j}\partial_{r_k} H_i]} \right)_{T,\mu,\mathbf{H},\partial_{\mathbf{r}}\mathbf{H}},$$9Sn, the anomalous Hall effect, anomalous Nernst effect, and MOKE are explicitly stated not to be induced by the spin-magnetization due to canting but by the cluster magnetic octupole. The AF spin texture with ferroic octupole order induces large Berry curvature due to Weyl points in momentum space, leading to large transverse response (Miwa et al., 2021).

This viewpoint generalizes to altermagnets. In $M_{ijk}$0-wave altermagnets the magnetic octupole Hall effect is a transverse flow of octupole moments induced by an electric field. Some octupole Hall components track the spin-splitter effect, while other components remain symmetry-allowed even when the spin-splitter current is forbidden. The octupole current is therefore not reducible to ordinary spin transport (Ko et al., 1 Aug 2025).

A second route is the nonlinear magnetoelectric effect,

$M_{ijk}$1

whose response tensor has the same symmetry as magnetic octupoles. The intrinsic contribution involves the quantum metric and Berry-connection polarizability, and in a magnetic Weyl semimetal phase of the pyrochlore model the intrinsic NMEE becomes large because the relevant tensor is enhanced near Weyl points. The same work argues that the NMEE has a sizeable value that can be detected by the magneto-optical Kerr effect (Ōiké et al., 2024).

The dynamical consequences are especially striking in Mn$M_{ijk}$2Sn. Time-resolved MOKE directly observed octupole oscillations with a high-frequency optical mode at $M_{ijk}$3 THz and a lower-frequency collective mode at $M_{ijk}$4 GHz at $M_{ijk}$5 T, with effective damping constants $M_{ijk}$6 and $M_{ijk}$7. The inferred switching time is $M_{ijk}$8 ps, a hundred times faster than the case of spin-magnetization in a ferromagnet, and the maximum Néel-type domain-wall velocity is predicted to exceed $M_{ijk}$9 km/s for the low-temperature spin length. The work therefore describes a regime of giant effective damping of the order parameter, not a large microscopic Gilbert damping (Miwa et al., 2021).

6. Broader significance, misconceptions, and current directions

The major conceptual correction introduced by this literature is that zero net magnetization does not imply absence of magnetic order, Berry curvature, or optical activity. Ferroic magnetic octupole order breaks time-reversal symmetry, can transform in the same irreducible representation as magnetization, and can therefore generate Hall, Nernst, Kerr, and SHG signals while remaining nearly silent in bulk magnetization. This is why octupolar phases are simultaneously hidden and observable: hidden to rank-1 probes, observable to symmetry-sensitive and topological probes (Higo et al., 2018, Miwa et al., 2021).

A second recurrent issue is the relation between bulk and boundary. In FeF$$M'_i \coloneqq \frac{3 M_{jji} - M_{ijj}}{\sqrt{10}},$$0, bulk ferroic octupoles determine surface magnetoelectric multipoles, surface magnetization, and surface polarization. This was formulated explicitly as a bulk-boundary correspondence of unconventional antiferromagnets (Bhowal et al., 2024). A plausible implication is that interfaces and thin films may be the most efficient platforms for reading out and controlling octupole order, even when the bulk is inversion-symmetric and magnetization-free.

The current frontier lies in direct probes and active control. SHG has now been demonstrated as a direct optical probe of a ferroic-octupolar order parameter in CoF$$M'_i \coloneqq \frac{3 M_{jji} - M_{ijj}}{\sqrt{10}},$$1 (Usachev et al., 16 Jun 2026). Circular-dichroism microscopy has directly imaged octupole domains in Eu$$M'_i \coloneqq \frac{3 M_{jji} - M_{ijj}}{\sqrt{10}},$$2Ir$$M'_i \coloneqq \frac{3 M_{jji} - M_{ijj}}{\sqrt{10}},$$3O$$M'_i \coloneqq \frac{3 M_{jji} - M_{ijj}}{\sqrt{10}},$$4 (Han et al., 6 May 2025). Multipole Hall transport and nonlinear magnetoelectricity have been proposed as robust signatures in altermagnets and Weyl systems (Ko et al., 1 Aug 2025, Ōiké et al., 2024). At the same time, 5d double perovskites continue to define the microscopic many-body problem of multipolar order, including the distinction between ferro-octupolar and antiferro-octupolar ground states and the role of crystal-field splitting, exchange anisotropy, and excited multiplets (Pourovskii et al., 2021, Kozlowski et al., 2024).

Taken together, these results place ferroic magnetic octupole order alongside ferromagnetism and ferroelectricity as a genuine ferroic principle in correlated solids. Its distinctive feature is that the ordered quantity is rank-3, often cluster-based or gauge-invariant only at the bulk-tensor level, yet it controls measurable macroscopic phenomena with unusual speed, topology, and boundary sensitivity.