Revised Hard-Magnetic Material Theory

• Revised hard-magnetic material theory is a multiscale framework that interprets anisotropy, coercivity, and field response as interconnected phenomena rather than outcomes of a single mechanism.
• The revision refines coercivity models through defect-mediated nucleation, higher-order crystal-field effects, and optimized interfacial exchange in hard–soft composites.
• It integrates microscopic energy-scale engineering, interfacial coupling, and continuum formulations to guide the design of advanced permanent magnets and magnetoactive materials.

Revised hard-magnetic material theory denotes a contemporary reworking of hard-magnet physics in which intrinsic anisotropy, coercivity, and field-driven response are treated as multiscale phenomena rather than as consequences of a single dominant mechanism. In recent literature, the revision is visible in several directions: higher-order crystal-field terms replace second-order-only single-ion models in rare-earth intermetallics; defect-mediated nucleation and activation-volume scaling refine coercivity theory in permanent magnets; hard–soft composites are reinterpreted through optimized interfacial exchange and dipolar coupling rather than perfect exchange-spring ideals; and hard-magnetic soft materials are described by rod, plate, and shell theories with explicit magnetic body forces and body couples rather than by purely classical elasticity (Passos et al., 21 Jul 2025, Bance et al., 2015, Erokhin et al., 2016, Sano et al., 2021).

1. Conceptual scope of the revision

The revision does not consist of a single universal model. It is a family of corrections to earlier simplifying assumptions that were material-specific, geometry-specific, or scale-specific.

Subdomain	Earlier working assumption	Revised formulation
RE–TM permanent magnets	$A_2^0\langle r^2\rangle$ and an effective exchange field are sufficient	Large $A_6^6\langle r^6\rangle$ can be physically relevant and drive $J$-multiplet mixing
Hard/soft composites	Perfect exchange is optimal; remanence is additive without exchange	Weakened intergrain exchange can maximize $(BH)_{\max}$, and dipolar coupling can raise remanence in exchange-decoupled systems
Coercivity in hard magnets	Coherent rotation or classical wall pinning dominates	Reversal proceeds through formation and expansion of a domain-wall-like nucleus at defects
Hard-magnetic soft structures	Classical beam/plate/shell mechanics is sufficient	Magnetic body forces, body couples, rotation-based magnetization mapping, and micropolar stresses are required

In rare-earth intermetallics, the revision is microscopic: higher-order crystal-field terms, especially the sixth-order $A_6^6\langle r^6\rangle$ parameter in hexagonal $6/mmm$ symmetry, are now treated as potentially large and physically relevant rather than negligible (Passos et al., 21 Jul 2025). In nanocomposite permanent magnets, the revision is interfacial: the maximal energy product need not occur at perfect exchange, and even exchange-decoupled hard–soft mixtures can show remanence enhancement through magnetostatic coupling (Erokhin et al., 2016, Guzmán-Mínguez et al., 2023). In coercivity theory, the revision is mechanistic: cross-plots of activation volume and coercive field support a universal defect-controlled nucleation-and-expansion picture rather than a clean coherent-rotation/pinning dichotomy (Bance et al., 2015). In hard-magnetic elastomeric structures, the revision is continuum-theoretic: magnetic torques make the Cauchy stress asymmetric, and dimensional reduction from 3D magnetoelasticity yields explicitly magnetic rod, beam, plate, and shell equations (Dadgar-Rad et al., 2022, Yan et al., 2022).

Taken together, these developments shift hard-magnetic theory from a predominantly phenomenological description of hysteresis loops toward a hierarchy of models that resolve microscopic energy scales, interfacial couplings, and geometry-dependent field interactions.

2. Rare-earth intermetallics and the revision of the RECo$_5$ ground-state picture

A central microscopic revision concerns the RECo$_5$ family, especially SmCo$_5$, NdCo$_5$, and YCo$A_6^6\langle r^6\rangle$0. All three compounds crystallize in the CaCu$A_6^6\langle r^6\rangle$1-type hexagonal structure, space group $A_6^6\langle r^6\rangle$2, over $A_6^6\langle r^6\rangle$3–$A_6^6\langle r^6\rangle$4, with the rare earth at $A_6^6\langle r^6\rangle$5 and Co split between $A_6^6\langle r^6\rangle$6 (Co1) and $A_6^6\langle r^6\rangle$7 (Co2). Neutron powder diffraction was carried out at the PEARL diffractometer with $A_6^6\langle r^6\rangle$8 over $A_6^6\langle r^6\rangle$9–$J$0, and legacy inelastic neutron scattering spectra for SmCo$J$1 were reanalyzed using HET data with incident energies $J$2 and $J$3 (Passos et al., 21 Jul 2025).

The single-ion Hamiltonian for the RE site is written as

$J$4

with

$J$5

and crystal field

$J$6

For Sm$J$7 in site symmetry $J$8, the allowed Wybourne–Stevens form is

$J$9

The revision lies in the status of $(BH)_{\max}$0: earlier interpretations treated $(BH)_{\max}$1 and $(BH)_{\max}$2 as sufficient, whereas state-of-the-art DFT+DMFT and neutron analysis show that large sixth-order terms can mix $(BH)_{\max}$3 multiplets and alter low-energy anisotropy (Passos et al., 21 Jul 2025).

The YCo$(BH)_{\max}$4 case serves as the $(BH)_{\max}$5-only benchmark. Its magnetic structure is a strong collinear ferromagnet with moments aligned along the $(BH)_{\max}$6 axis, and the anisotropy predominantly arises from the Co1 ($(BH)_{\max}$7) site’s orbital contribution. At $(BH)_{\max}$8, refined site moments are approximately Co($(BH)_{\max}$9) $A_6^6\langle r^6\rangle$0 and Co($A_6^6\langle r^6\rangle$1) $A_6^6\langle r^6\rangle$2, both decreasing monotonically with temperature up to $A_6^6\langle r^6\rangle$3. The unit-cell volume follows

$A_6^6\langle r^6\rangle$4

with no magnetoelastic anomalies in $A_6^6\langle r^6\rangle$5, $A_6^6\langle r^6\rangle$6, or $A_6^6\langle r^6\rangle$7 (Passos et al., 21 Jul 2025). This affirms axial ferromagnetism while refining it through site resolution.

NdCo$A_6^6\langle r^6\rangle$8 displays the more intricate $A_6^6\langle r^6\rangle$9–$6/mmm$0 competition that motivated the revision. Co carries strong axial anisotropy, whereas the Nd crystal field drives an easy-plane tendency. Neutron powder diffraction directly resolves two spin reorientation transitions: below $6/mmm$1 the refined $6/mmm$2 components are negligible within uncertainty, indicating a basal-plane state; between $6/mmm$3 and $6/mmm$4 the system is conical; above $6/mmm$5 the Co axial anisotropy dominates and moments align preferentially along the $6/mmm$6 axis. The unit-cell volume follows

$6/mmm$7

but $6/mmm$8 shows two anomalies at the spin reorientation temperatures, directly linking anisotropy competition to anisotropic thermal expansion (Passos et al., 21 Jul 2025). This moves the subject from bulk-magnetization inference to temperature-resolved vector magnetic structures.

SmCo$6/mmm$9 is the hardest magnet in the series and the most direct test of the revised single-ion theory. Its prominent INS peaks remain near $_5$0 and $_5$1, but a weaker feature near $_5$2 is sensitive to $_5$3. The unconstrained fit gives $_5$4, $_5$5, $_5$6, with $_5$7. A constrained fit with $_5$8 from ab initio gives $_5$9, $_5$0, with $_5$1. The main features remain controlled by $_5$2 and $_5$3, but the sixth-order term cannot be dismissed a priori (Passos et al., 21 Jul 2025). A plausible implication is that hard-magnet design in RE–TM systems must treat higher-order CEF engineering and magnetoelastic coupling as coequal with exchange-field optimization.

3. Intrinsic hardness without rare earths: electron count, strain, and anisotropy engineering

A second major revision concerns intrinsic materials screening for rare-earth-free or rare-earth-reduced hard magnets. In this literature, the dominant intrinsic figure is the magnetic hardness parameter

$_5$4

or, in the monoboride study,

$_5$5

with the classification soft, semi-hard, and hard tied to whether $_5$6 is below $_5$7, between $_5$8 and $_5$9, or above $_5$0 (Snarski-Adamski et al., 2024, Snarski-Adamski et al., 2024). This reformulates material selection around the balance between anisotropy energy density and magnetostatic energy density rather than around magnetization alone.

First-principles work on carbides related to Fe$_5$1C and Co$_5$2C shows that orthorhombic Co$_5$3C is a particularly important case. In the $_5$4 phase, Co$_5$5C has $_5$6, robust uniaxial anisotropy with easy axis [100], and $_5$7; the paper also cites experimental $_5$8 and $_5$9 for nanoparticles (Snarski-Adamski et al., 2024). Co-rich orthorhombic and hexagonal $_5$0 alloys approach or exceed the hard threshold, and a two-dimensional compositional map over $_5$1-site and $_5$2-site substitutions reveals diagonal stripes of nearly constant MAE, described as the near isoelectronic nature of MAE. In this framework, boron is significant because it raises the Curie temperature and improves stability, while nitrogen can raise MAE numerically in Co-rich compositions. The design rule is not merely “maximize Co” or “maximize SOC,” but tune total electron count and stability simultaneously (Snarski-Adamski et al., 2024).

A parallel monoboride study extends the same intrinsic logic to MnB, FeB, and their alloys. MnB and FeB are orthorhombic, with easy axis [010] in the pure compounds. FeB itself is semi-hard in the calculations, but FeB alloys with Sc, Ti, V, Zr, Nb, Mo, Hf, Ta, or W are classified as magnetically hard with $_5$3 (Snarski-Adamski et al., 2024). The same study finds exceptionally high $_5$4 values in VCA for $_5$5, exceeding five around the Fe$_5$6Co$_5$7B composition, while also warning that these values are inflated by VCA and that actual magnetic hardnesses are nevertheless expected to remain well above unity (Snarski-Adamski et al., 2024). Here the revision is methodological as well as physical: orthorhombic anisotropy is not reduced to a single “easy-axis constant,” but resolved through the ordered energies $_5$8, the quantity $_5$9, and the complementary splitting $A_6^6\langle r^6\rangle$00.

A distinct route to rare-earth-free hardness is interstitial strain engineering in Fe–Co–B. When boron occupies octahedral interstitial sites in bcc Fe–Co, the lattice spontaneously becomes body-centered tetragonal, with the $A_6^6\langle r^6\rangle$01 axis elongated and the $A_6^6\langle r^6\rangle$02 axes compressed. Tetrahedral occupation is disfavored by $A_6^6\langle r^6\rangle$03 per B atom in bcc Fe$A_6^6\langle r^6\rangle$04B and $A_6^6\langle r^6\rangle$05 per B atom in Fe$A_6^6\langle r^6\rangle$06Co$A_6^6\langle r^6\rangle$07, and uniaxial octahedral alignment is favored over orthogonal or random orientations by energy differences such as $A_6^6\langle r^6\rangle$08 and $A_6^6\langle r^6\rangle$09 in the cited supercells (Reichel et al., 2015). In epitaxial films, spontaneous strain up to $A_6^6\langle r^6\rangle$10 lattice distortion is obtained for B contents up to $A_6^6\langle r^6\rangle$11, leading to uniaxial anisotropy constants exceeding $A_6^6\langle r^6\rangle$12, whereas further B addition causes partial amorphization and degrades both anisotropy and magnetization (Reichel et al., 2015). This revises hard-magnet theory by showing that hard behavior can be induced not only by chemistry and SOC, but also by symmetry-breaking spontaneous strain from interstitial occupancy.

These studies collectively replace a chemistry-by-analogy search strategy with a physically sharper one: tune electron count, phase stability, and local symmetry so that MAE, $A_6^6\langle r^6\rangle$13, and $A_6^6\langle r^6\rangle$14 remain jointly favorable.

4. Coercivity, exchange, and dipolar coupling in permanent-magnet microstructures

In coercivity theory, the revision is explicitly anti-reductionist. Hard-magnet coercivity is no longer treated as a direct proxy for the anisotropy field alone; it is instead decomposed into intrinsic, defect, demagnetizing, and activation-volume contributions. For RFeB magnets, a Kronmüller-type form is written as

$A_6^6\langle r^6\rangle$15

while the global activation model gives

$A_6^6\langle r^6\rangle$16

and

$A_6^6\langle r^6\rangle$17

with $A_6^6\langle r^6\rangle$18 across a wide range of RFeB samples (Bance et al., 2015). The common interpretation is that reversal proceeds through the formation and expansion of a domain-wall-like nucleus at microstructural defects, with the activation volume inheriting near-main-phase properties up to room temperature and then deviating at higher temperature (Bance et al., 2015). The angular dependence is closer to Kondorsky’s $A_6^6\langle r^6\rangle$19 behavior than to a Stoner–Wohlfarth astroid, again favoring a domain-wall-mediated process.

Hard–soft nanocomposite theory is revised in parallel. Micromagnetic simulations of Sr-ferrite/Fe and Sr-ferrite/Ni composites show that the maximal energy product is achieved not at perfect intergrain exchange, but when the exchange across grain boundaries is substantially weakened (Erokhin et al., 2016). The exchange weakening coefficient $A_6^6\langle r^6\rangle$20 was explored from $A_6^6\langle r^6\rangle$21 to $A_6^6\langle r^6\rangle$22, and for SrFe$A_6^6\langle r^6\rangle$23O$A_6^6\langle r^6\rangle$24/Fe with spherical hard grains the optimum in $A_6^6\langle r^6\rangle$25 and $A_6^6\langle r^6\rangle$26 occurs near $A_6^6\langle r^6\rangle$27, whereas overly strong coupling makes reversal too cooperative (Erokhin et al., 2016). Shape effects are likewise nontrivial: oblate hard grains can boost remanence through magnetizing dipolar fields in a high-$A_6^6\langle r^6\rangle$28 soft matrix, but spheres minimize interfacial area and can maximize coercivity. This overturns the common assumption that stronger exchange and larger elongation are always beneficial.

An even sharper revision is provided by exchange-decoupled SrFe$A_6^6\langle r^6\rangle$29O$A_6^6\langle r^6\rangle$30/Fe composites. In anisotropic injection-molded magnets with $A_6^6\langle r^6\rangle$31 Fe, the remanent polarization rises from $A_6^6\langle r^6\rangle$32 in pure SFO to $A_6^6\langle r^6\rangle$33 for both $A_6^6\langle r^6\rangle$34 and $A_6^6\langle r^6\rangle$35 Fe, even though a linear mixture model would predict an approximately $A_6^6\langle r^6\rangle$36 decrease in remanence at $A_6^6\langle r^6\rangle$37 Fe (Guzmán-Mínguez et al., 2023). The mechanism is not exchange-spring coupling: oxide and Fe$A_6^6\langle r^6\rangle$38Si shells, voids, and polymer separation suppress exchange, and micromagnetic simulations attribute the surplus remanence to dipolar alignment of a small fraction of soft spins, inferred from powder data to be roughly $A_6^6\langle r^6\rangle$39 (Guzmán-Mínguez et al., 2023). The trade-off is explicit in the same data: remanence gains are accompanied by reduced coercivity and loop squareness.

At the multiscale level, computational design now couples ab initio intrinsic parameters to micromagnetic simulations of realistic polycrystals and machine-learning-based microstructural optimization. For several candidate phases, the reported pairs of coercive field and energy density product are Fe$A_6^6\langle r^6\rangle$40Sn$A_6^6\langle r^6\rangle$41Sb$A_6^6\langle r^6\rangle$42: $A_6^6\langle r^6\rangle$43, L1$A_6^6\langle r^6\rangle$44 FeNi: $A_6^6\langle r^6\rangle$45, CoFe$A_6^6\langle r^6\rangle$46Ta: $A_6^6\langle r^6\rangle$47, and MnAl: $A_6^6\langle r^6\rangle$48, in units of tesla and $A_6^6\langle r^6\rangle$49 (Kovacs et al., 2019). The same framework states that Fe-rich rare-earth-free phases with $A_6^6\langle r^6\rangle$50–$A_6^6\langle r^6\rangle$51 and $A_6^6\langle r^6\rangle$52–$A_6^6\langle r^6\rangle$53 rarely exceed $A_6^6\langle r^6\rangle$54 even in optimized nanostructures (Kovacs et al., 2019). The revised picture is therefore one of bounded performance: intrinsic hardness is necessary, but grain-boundary thickness, grain-boundary magnetization, aspect ratio, exchange decoupling, and thermal activation determine how much of it survives in the hysteresis loop.

5. Continuum reformulations for hard-magnetic soft materials

In hard magneto-rheological elastomers and related hard-magnetic soft materials, revised theory begins from the assumption that programmed magnetization is permanent after saturation and rotates with the matrix under subsequent actuation fields. The classical stress-only framework is then insufficient, because magnetic torque generates distributed body couples. This has led to a hierarchy of reduced theories derived from 3D magnetoelastic energy rather than from ad hoc beam or plate analogies (Sano et al., 2021, Yan et al., 2021).

For beam-like structures under nonuniform fields, the 3D continuum formulation introduces the magnetic body force density

$A_6^6\langle r^6\rangle$55

the magnetic Cauchy stress

$A_6^6\langle r^6\rangle$56

and the magnetic torque density

$A_6^6\langle r^6\rangle$57

Dimensional reduction then yields a geometrically nonlinear inextensible beam equation in which torque and the tail resultant of gradient-induced forces both enter the bending balance (Yan et al., 2021). The same paper identifies the dimensionless parameters

$A_6^6\langle r^6\rangle$58

for uniform-field actuation and

$A_6^6\langle r^6\rangle$59

for constant-gradient actuation, and validates the resulting theory against experiments on hard-MRE cantilevers (Yan et al., 2021).

The rod formulation generalizes this to full 3D Cosserat kinematics. Starting from the per-length magnetization

$A_6^6\langle r^6\rangle$60

the reduced theory defines magnetic body force and couple densities

$A_6^6\langle r^6\rangle$61

and inserts them into the Kirchhoff-type balance laws

$A_6^6\langle r^6\rangle$62

(Sano et al., 2021). Uniform fields produce torques alone, whereas gradient fields produce both torques and net forces. The theory reproduces prior 2D elastica results and predicts twist, bend, and twist–bend coupled instabilities in naturally straight and helical rods (Sano et al., 2021).

Plate theory required a different revision. Earlier hard-MRE continuum models mapped magnetization by the full deformation gradient, $A_6^6\langle r^6\rangle$63, which couples magnetization to both rotation and stretch. The revised theory uses the rotation tensor alone,

$A_6^6\langle r^6\rangle$64

so that magnetic response is independent of stretch and depends only on rotation of rigid hard-magnetic inclusions (Yan et al., 2022). The corresponding magnetic potential becomes

$A_6^6\langle r^6\rangle$65

and its thin-plate reduction gives

$A_6^6\langle r^6\rangle$66

Experiments on clamped plates show that this change is crucial under aligned fields and in-plane stretching: the rotation-based 3D and 2D models agree with measured deflections, while the earlier $A_6^6\langle r^6\rangle$67-based 3D model can overpredict stiffening by roughly a factor of two (Yan et al., 2022).

For thin shells, the revision reaches the constitutive level. A 10-parameter micropolar shell model introduces midsurface translations, director translations, microrotations, and thickness stretch. The magnetic body couple in referential form is

$A_6^6\langle r^6\rangle$68

and angular momentum balance becomes

$A_6^6\langle r^6\rangle$69

so $A_6^6\langle r^6\rangle$70 is generally asymmetric (Dadgar-Rad et al., 2022). The shell formulation uses a micropolar neo-Hookean constitutive law and enhanced assumed strain modes to avoid locking at large distortions. This is not merely a numerical refinement: it encodes the claim that hard-magnetic soft structures are intrinsically couple-stress media when externally actuated by magnetic induction.

6. Nonclassical extensions: thermomagnetic cycles and interfacial anisotropy

A further revision broadens the theory of hard magnets beyond permanent-magnet statics. In thermomagnetic power generation, hard ferromagnets are reconsidered as working materials because their wide hysteresis loops access two quadrants of the $A_6^6\langle r^6\rangle$71–$A_6^6\langle r^6\rangle$72 plane. The work per cycle is written as

$A_6^6\langle r^6\rangle$73

and cycle efficiency is expressed through a four-step thermodynamic balance (Tantillo et al., 2023). Experiments on commercial hard ferrites gave average clockwise work outputs of $A_6^6\langle r^6\rangle$74 for Y30BH and $A_6^6\langle r^6\rangle$75 for HF8/22, while simulations of soft magnets biased by hard NdFeB predicted gains of $A_6^6\langle r^6\rangle$76 for Gd and $A_6^6\langle r^6\rangle$77 for LCFS-H when total system mass is counted, and approximately $A_6^6\langle r^6\rangle$78 when normalized to the soft-magnet mass only (Tantillo et al., 2023). The same perspective advocates artificial spin reorientation materials, in which anisotropy-engineered bilayers create thermally driven reorientation transitions. In this context, revised hard-magnetic theory becomes loop engineering: coercivity, remanence, bias fields, and anisotropy-temperature dependence are treated as energy-conversion variables.

An interfacial revision appears in Co/C$A_6^6\langle r^6\rangle$79 bilayers, where molecule–metal coupling produces a form of anisotropy not captured by conventional interface-SOC models. The paper introduces a spin-dependent polarization

$A_6^6\langle r^6\rangle$80

arising from asymmetric magneto-electric coupling at a $A_6^6\langle r^6\rangle$81–$A_6^6\langle r^6\rangle$82 hybrid interface (Moorsom et al., 2019). In field-cooled Co/C$A_6^6\langle r^6\rangle$83 bilayers, the first sweep at $A_6^6\langle r^6\rangle$84 shows coercivity up to $A_6^6\langle r^6\rangle$85 and exchange-bias-like offsets up to $A_6^6\langle r^6\rangle$86, with a maximum energy product $A_6^6\langle r^6\rangle$87 for a $A_6^6\langle r^6\rangle$88 Co film capped by $A_6^6\langle r^6\rangle$89 of C$A_6^6\langle r^6\rangle$90 (Moorsom et al., 2019). DFT finds preferential adsorption of C$A_6^6\langle r^6\rangle$91 on Co(111) at the hexagon–pentagon site with adsorption energy $A_6^6\langle r^6\rangle$92 and interfacial dipole density $A_6^6\langle r^6\rangle$93; the estimated electrostatic barrier to in-plane spin rotation is $A_6^6\langle r^6\rangle$94–$A_6^6\langle r^6\rangle$95 (Moorsom et al., 2019). The authors call this $A_6^6\langle r^6\rangle$96-anisotropy. A plausible implication is that hard-magnetic theory now includes interfacial magneto-electric anisotropy as a distinct route to coercivity enhancement, separate from bulk single-ion or exchange mechanisms.

7. Limitations, unresolved parameters, and prospective synthesis

The revised theory remains incomplete in each of its branches. In SmCo$A_6^6\langle r^6\rangle$97, the magnitude of $A_6^6\langle r^6\rangle$98 is still uncertain because legacy INS resolution and phonon background issues leave the $A_6^6\langle r^6\rangle$99 shoulder unresolved; the same work does not extract explicit $J$00, $J$01, or $J$02 values (Passos et al., 21 Jul 2025). In hard-magnetic rod theory, long-range dipole–dipole interactions are neglected in the reduction but can produce hysteresis, self-attraction, and self-contact in helical structures under gradient fields (Sano et al., 2021). In carbide and monoboride screening, MAE and hardness are largely $J$03 quantities, VCA can overestimate anisotropy, and microstructural coercivity is not modeled directly (Snarski-Adamski et al., 2024, Snarski-Adamski et al., 2024). In thermomagnetic cycles, the bias-field gains are computational rather than experimental, and key transport parameters such as thermal conductivity and quantitative $J$04 remain open (Tantillo et al., 2023). In Co/C$J$05 interfaces, the effect is presently limited by molecular rotational freedom near $J$06–$J$07, and room-temperature realization depends on orientation locking through surface functionalization or molecular redesign (Moorsom et al., 2019).

What is already established, however, is that “hardness” is no longer a unitary concept. It can be the robustness of an RE single-ion easy axis against $J$08 mixing, the defect-limited resistance to domain-wall-like nucleation, the survival of coercivity under intentionally weakened intergrain exchange, the dipolar alignment of soft spins in an exchange-decoupled composite, the emergence of large MAE from spontaneous tetragonal strain or near-isoelectronic band filling, or the resistance of a programmed soft structure to reorientation under a distributed magnetic body couple. Revised hard-magnetic material theory is therefore best understood as a multiscale synthesis: microscopic energy-scale engineering, interfacial and microstructural control, and geometry-aware continuum modeling are treated as mutually constraining parts of one field rather than as separate specialties.