Axionic Electrodynamics

Definition

Axionic electrodynamics is the extension of Maxwell electrodynamics by a topological coupling of the electromagnetic field to a pseudoscalar field $\theta (x)$ (the axion or $\theta$-term):

${\mathcal{L}}_{\text{axion}}=-\frac{1}{4e^2}{F}_{\mu \nu }{F}^{\mu \nu }+\frac{\theta }{16{\pi }^2}{F}_{\mu \nu }{\tilde{F}}^{\mu \nu },\text{i.e.}{S}_{\theta }=\frac{\theta }{8{\pi }^2}\int F\wedge F,$

where ${\tilde{F}}^{\mu \nu }=\frac{1}{2}{ϵ}^{\mu \nu \rho \sigma }{F}_{\rho \sigma }$ is the dual field strength. The axionic term $F\tilde{F}\propto \mathbf{E}\cdot \mathbf{B}$ (in components) mixes electric and magnetic fields.

Normalization. $A$ couples to integer charges and $e$ sits in front of the Maxwell term, so that $\theta \cong \theta +2\pi$ on spin manifolds; see Notes. This is the convention of Witten (2026) and of the generalized-symmetries course, Week 12. It was corrected on 2026-09-21: an earlier version had $\theta /8{\pi }^2$ in front of $F\tilde{F}$, twice the $2\pi$-periodic normalization and inconsistent with the equations of motion below.

When $\theta =\theta (x)$ is a dynamical field (the axion), the equations of motion become:

$\frac{1}{e^2}{\partial }_{\mu }{F}^{\mu \nu }=J_e^{\nu }+\frac{1}{4{\pi }^2}({\partial }_{\mu }\theta ){\tilde{F}}^{\mu \nu },$ ${\partial }_{\mu }{\tilde{F}}^{\mu \nu }=J_m^{\nu }.$

The axionic coupling produces magnetoelectric mixing: an electric field generates a magnetic field and vice versa, proportional to $\nabla \theta$. Modified boundary conditions at interfaces where $\theta$ jumps produce surface Hall currents and anomalous response.

In condensed matter physics, axionic electrodynamics describes topological insulators: materials where the bulk electromagnetic response is characterized by a quantized $\theta =\pi$ (due to time-reversal symmetry), leading to the topological magnetoelectric effect:

$\mathbf{P}=\frac{\theta e^2}{2\pi hc}\mathbf{B},\mathbf{M}=\frac{\theta e^2}{2\pi hc}\mathbf{E}.$

In QCD, the $\theta$-term $\frac{\theta }{32{\pi }^2}{G}_{\mu \nu }^a{\tilde{G}}^{a\mu \nu }$ is the analog, with the strong CP problem arising from experimental bounds $\theta <10^{-10}$.

The dyonic condensate produces axionic electrodynamics naturally: when both electric and magnetic condensates coexist (dyons — particles with both electric and magnetic charges), the effective theory of the condensate includes an axionic coupling mixing the electric and dual magnetic gauge fields.

Role in Research

Axionic electrodynamics appears in the confinement-duality and condensed-matter-connections research lines as an effective description of phases with simultaneous electric and magnetic condensation.

Key connections:

• Julia-Toulouse + dyons: when the Julia-Toulouse mechanism is applied with dyonic defects (carrying both electric and magnetic charge), the effective low-energy theory after condensation contains axionic coupling terms. This connects the topological defect condensation picture to axionic electrodynamics.

• Topological insulators: the condensed matter realization of axionic electrodynamics ($\theta =\pi$ topological insulators) provides a concrete physical system where the same mathematics applies. The bridge to high-energy physics through this shared mathematical structure is part of the condensed-matter-connections program.

• BF theory: the BF topological field theory $\mathcal{L}=B\wedge F$ that emerges from the Julia-Toulouse mechanism is related to axionic electrodynamics by integrating out the auxiliary $B$ field, establishing a chain of dualities.

The axionic coupling also plays a role in the strong CP problem and axion physics (Peccei-Quinn mechanism), though this is not a primary focus of the current research.

Relations

• julia-toulouse-mechanism — dyonic condensation in the Julia-Toulouse framework generates axionic coupling terms; the effective theory after condensation is axionic electrodynamics
• dual-superconductor — the dual superconductor with dyonic condensate (both electric and magnetic monopole condensates) leads to a phase described by axionic electrodynamics
• confinement — the axionic coupling in the QCD vacuum ($\theta$-term) is related to topological properties of the vacuum and has implications for the confinement mechanism and CP violation
• condensed-matter-connections — topological insulators realize $\theta =\pi$ axionic electrodynamics; this is a concrete condensed matter realization of the same mathematics appearing in gauge theory confinement
• chern-simons-theory — the θ-term on a region with boundary is a Chern–Simons term on the boundary, at level $\Delta \theta /2\pi$
• quantum-hall-effect — a θ-wall carries Hall conductance $(\Delta \theta /2\pi )e^2/h$; the $\theta =\pi$ surface is half-quantized

Papers

See confinement-duality and condensed-matter-connections for full paper lists.

Notes

• The $F\tilde{F}$ term is a total derivative in flat space: $F_{\mu \nu }{\tilde{F}}^{\mu \nu }={\partial }_{\mu }({ϵ}^{\mu \nu \rho \sigma }{A}_{\nu }{F}_{\rho \sigma })$, so it does not affect equations of motion when $\theta$ is a constant. When $\theta (x)$ varies, its gradient contributes to the equations of motion — hence the requirement for a dynamical axion field to produce physical effects.
• In topological insulators, $\theta$ is not a dynamical field but a fixed material parameter determined by the band topology. The quantization $\theta =0$ or $\pi$ (in units of $\pi$) is protected by time-reversal symmetry.
• Common confusion: the axion of axion physics (the Peccei-Quinn axion, a solution to the strong CP problem) is a dynamical pseudoscalar field. The $\theta$ parameter of QCD is a fixed coupling. The topological insulator $\theta$ is a material property. All three involve the same $F\tilde{F}$ coupling but in very different physical contexts.
• The BF theory connection: $\mathcal{L}=\frac{1}{4\pi }B\wedge dA$ (where $B$ is a 2-form and $A$ a 1-form) in 3+1 dimensions encodes the linking number of worldlines and worldsheets. Integrating out $B$ or $A$ gives complementary descriptions — one axionic, one topological.
• θ-walls are Chern–Simons levels (Witten 2026, eq. 25).
  • A film’s Chern–Simons function is ${\mathrm{CS}}_X(A)={\int }_{X}F\wedge F/8{\pi }^2$ over any 4-manifold $X$ it bounds. A region with $\theta =2\pi k$ bounded by $W$ is therefore equivalent to $2\pi k\mathrm{CS}(A)$ on $W$.
  • A θ-wall of height $\Delta \theta$ carries Hall conductance $(\Delta \theta /2\pi )e^2/h$.
  • At $\Delta \theta =\pi$ the level is $\frac{1}{2}$, which Witten’s integrality argument forbids for any standalone 2d insulator. The $\theta =\pi$ surface of a topological insulator must therefore be anomalous: gapless (a single Dirac cone), gapped only by a $T$-breaking perturbation that makes sense only as the boundary of the bulk, or carrying surface topological order.
  • Its half-quantized ${\sigma }_{xy}=e^2/2h$ is the anomaly-inflow signature of the bulk θ-term. See chern-simons-theory and quantum-hall-effect.
• Periodicity needs a spin structure. On a spin 4-manifold the intersection form is even, so $\int F\wedge F/8{\pi }^2\in \mathbb{Z}$ and $\theta \cong \theta +2\pi$. With purely bosonic matter and no spin structure the integral is only in $\frac{1}{2}\mathbb{Z}$, and $\theta \cong \theta +4\pi$ (Witten 2026, eqs. 23–24). In a bosonic system $\theta =2\pi$ is then a distinct phase, the bosonic topological insulator, where a monopole becomes a fermion: the “statistical Witten effect” (Metlitski–Kane–Fisher, PRB 88 (2013) 035131).
• Time reversal and Witten’s footnote 4. The footnote says the coefficient of $\vec{E}\cdot \vec{B}$ in a material’s effective action must vanish if $T$ holds. That is true for any local coupling. The loophole is that θ is an angle and $\theta =\pi \equiv -\pi$ is $T$-invariant, which is the topological-insulator story in one line.