Propagation-based classification of linear magnetoelectric response in dielectrics
Eduardo Bittencourt 1,2, Elliton O.S.R. Brandão 1, Érico Goulart 3, Danilo H. Spadoti 1 · 2026
Read Propagation-based classification of linear magnetoelectric response in dielectrics by Eduardo Bittencourt 1,2, Elliton O.S.R. Brandão 1, Érico Goulart 3, Danilo H. Spadoti 1 (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-19235.txt.
Introduction
Bittencourt and three colleagues sort out how electromagnetic waves travel through uniform dielectrics that respond magnetoelectrically, meaning an electric field can stir magnetization and vice versa. Working in the geometric-optics limit, they derive a compact fourth-degree relation for phase speed and split the response matrix into trace, symmetric, and antisymmetric parts. The trace part stays silent, the antisymmetric part gives two closed-form branches, and the symmetric part drives the richest directional effects.
Catalog record
| CREATOR | Eduardo Bittencourt 1,2, Elliton O.S.R. Brandão 1, Érico Goulart 3, Danilo H. Spadoti 1 |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42086 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.19235
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Propagation-based classification of linear magnetoelectric response in dielectrics Eduardo Bittencourt 1,2, Elliton O.S.R. Brandão 1, Érico Goulart 3, Danilo H. Spadoti 1 Abstract We study electromagnetic wave propagation in homogeneous dielectrics endowed with a linear magnetoelectric (ME) response in the geometric-optics regime. Assuming isotropic permittivity and permeability while keeping a generic 3 × 3 3\times 3 ME matrix α i j \alpha_{ij} , we derive the eikonal (Fresnel) eigenvalue problem for the polarization vector and obtain a compact quartic dispersion relation for the normalized phase speed r = v / v d r=v/v_{d} , where v d = ( μ ε ) − 1 / 2 v_{d}=(\mu\varepsilon)^{-1/2} is the phase speed of the underlying dielectric. We then classify the propagation effects of α i j \alpha_{ij} by decomposing it into trace, symmetric-traceless, and antisymmetric sectors. We show that (i) the pure-trace sector is propagation-silent at leading geometric-optics order; (ii) the antisymmetric sector yields a factorized quartic and produces two branches with closed-form phase speeds, including regimes where | v | > v d |v|>v_{d} ; and (iii) the symmetric-traceless sector encodes the richest directional dependence through algebraic invariants that control the Fresnel wave surface and polarization mixing. Finally, we discuss how the predicted phase-speed shifts can be accessed by phase-sensitive transmission and resonant techniques, and we outline numerical workflows to validate the analytic dispersion and map polarization signatures in bulk and finite geometries. I Introduction Linear magnetoelectricity belongs to the broader class of bianisotropic media, where the electromagnetic excitations (D_{i},\,H_{i}) depend linearly on both field strengths (E_{i},\,B_{i}) . Such cross-couplings are central both to condensed-matter realizations (e.g., ME and multiferroic compounds) and to engineered metamaterials designed for polarization control and nonreciprocal functionalities. For a modern review emphasizing phenomenology and material platforms, see, e.g., 8 . Recent progress in optical Tellegen metamaterials further illustrates how large effective ME couplings can be achieved at visible frequencies ( 30 ) . Historically, the linear ME effect was anticipated on symmetry grounds in antiferromagnets by Dzyaloshinskii 7 , and subsequently observed in Cr 2 O 3 by Astrov 1 and shortly thereafter by Folen, Rado, and Stalder 9 . Beyond foundational interest, magnetoelectrics have been repeatedly proposed as functional materials for sensors, transducers, and electrically controlled magnetic devices; for a survey of applications and device-oriented perspectives, see Scott 33 and Fuentes-Cobas 10 . These developments motivate a careful analysis of how ME couplings influence wave propagation, including how they modify phase speed and polarization in a way that can be probed experimentally. From a covariant perspective, linear media can be described by a constitutive tensor \chi^{\mu\nu\rho\sigma} mapping the field strength to the excitation; in premetric electrodynamics one obtains a quartic Fresnel equation for the wave covector k_{\mu} whose coefficients are built from \chi^{\mu\nu\rho\sigma} ( 16 ; 14 ; 15 ; 19 ; 18 ; 21 ) . A key structural result is the irreducible decomposition of the constitutive tensor into a principal part (20 components), an axion part (1 component), and a skewon part (15 components). In particular, the axion contribution is known to drop out of the local Fresnel equation (it is “propagation-silent” in geometric optics), whereas the principal part controls birefringence and the skewon sector is tied to dissipation/nonreciprocity ( 16 ; 14 ; 15 ; 19 ; 18 ; 21 ) . A subtle but important point in linear magnetoelectricity is the distinction between reciprocal (lossless, energy-conserving) and nonreciprocal (Tellegen-type) couplings. In covariant language, energy conservation and reciprocity are naturally expressed as symmetries of \chi^{\mu\nu\rho\sigma} . In particular, the absence of skewon contributions is tied to a symmetric constitutive mapping and typically excludes nonreciprocal (dissipative) behavior 16 ; 19 ; 25 . In 1+3 language, analogous restrictions appear as symmetry relations among the 3\times 3 blocks entering the constitutive law. A complementary geometric viewpoint is provided by effective-geometry methods: in nonlinear electrodynamics (NLED) and related theories, the characteristics of the field equations define effective optical metrics or, more generally, quartic wave surfaces that may factorize into two metric light cones (birefringence) or remain genuinely quartic ( 5 ; 26 ; 11 ; 27 ; 31 ) . While our focus is linear ME response in media (rather than NLED in vacuum), the underlying mathematical structure is analogous: geometric optics reduces Maxwell’s system to an algebraic eigenproblem whose characteristic equation generically defines a quartic wave surface. This makes it natural to interpret ME-induced phase-speed splitting and direction-dependent propagation in the language of wave-surface geometry. In this broader context, it is useful to recall that NLED models also display nontrivial constraints related to causality and to electric–magnetic duality. A classic analysis of duality rotations in nonlinear electrodynamics, which underlies several later developments in Born–Infeld-type effective geometries, is given by Gibbons and Rasheed 12 . Although our present analysis concerns linear ME media, the same conceptual warning applies: phase-speed enhancement does not by itself imply superluminal signaling, and a fully causal interpretation requires accounting for dispersion and loss (group and signal velocities), which we leave for future work. In this work we isolate the role of magnetoelectricity by considering homogeneous isotropic permittivity and permeability, while keeping a general linear ME matrix \alpha_{ij} . The resulting dispersion relation can be written as a quartic polynomial for the normalized phase speed r=v/v_{\mathrm{d}} and depends on a small set of algebraic invariants of the ME contribution. The central technical point is that, after a 1+3 split and in the geometric-optics limit, the canonical decomposition of \alpha_{ij} into its trace, symmetric-traceless, and antisymmetric parts provides a practical propagation-based classification : each sector has a sharply distinct impact on phase speed and polarization eigenmodes. In particular, we show that the pure-trace sector is silent at leading order (mirroring axion-like behavior in the premetric decomposition), while the antisymmetric and symmetric-traceless sectors produce measurable signatures, including regimes where the phase speed exceeds the dielectric value v_{\mathrm{d}} . Related wave-propagation analyses based on Fresnel surfaces and constitutive classifications have also been explored in lower-dimensional settings, for instance, in recent studies of (2+1)-dimensional electrodynamics with general linear 13 and nonlinear constitutive laws 3 . Finally, it is worth emphasizing that more elaborate propagation phenomena may arise once one goes beyond the present assumptions, for instance, by incorporating nonlinear response or higher-order ME coefficients (see, for instance, Refs. 4 ; 23 ; 6 and references therein). The paper is summarized as follows. In Sec. II , we introduce the linear, local constitutive relations and fix our conventions for the dielectric tensors and the reciprocal cross-coupling matrix \alpha_{ij} . In Sec. III we take the geometric-optics limit and reduce Maxwell’s equations to a Fresnel-matrix eigenvalue problem for the polarization amplitude. This yields a compact quartic dispersion relation for the normalized phase speed r=v/v_{d} , written solely in terms of algebraic invariants of the rescaled magnetoelectric tensor \tilde{\alpha}_{ij} . In Sec. IV we classify the propagation effects of \alpha_{ij} by decomposing it into trace, antisymmetric, and symmetric-traceless sectors, and we show how each sector affects (or does not affect) the dispersion relation. In Sec. V we analyze the polarization eigenmodes associated with each physical branch, including the appearance (in some configurations) of quasi-longitudinal electric components constrained by Gauss’ law, and we provide explicit expressions