Solitons, Instantons and Topology

Perturbation theory — Feynman diagrams expanded around the vacuum — misses an entire class of phenomena that are nonperturbative and topological: field configurations that cannot be continuously deformed to the vacuum. These come in two kinds: solitons (static, localized, particle-like lumps) and instantons (localized in time, mediating quantum tunneling). They explain the $\theta$-vacuum of QCD, the strong-CP problem, and magnetic monopoles.

Conventions: $\hslash =c=1$, ${\eta }_{\mu \nu }=\mathrm{diag}(+1,-1,-1,-1)$.

Topological charge

The common thread is a conserved topological charge that is not a Noether charge — it comes from the topology of the field configuration, not from a symmetry. Finite-energy configurations must approach the vacuum at spatial infinity, defining a map from the boundary of space into the vacuum manifold. When that map is topologically non-trivial (has non-zero winding), the configuration is stable for reasons no local dynamics can undo: it cannot be unwound without passing through infinite energy.

$$Q_{\text{top}}\in {\pi }_n(\text{vacuum manifold}),\frac{d{Q}_{\text{top}}}{dt}=0\text{(topologically)}.$$

The relevant homotopy group ${\pi }_n$ depends on the spatial dimension and the symmetry-breaking pattern. The mathematics of this classification — homotopy groups and which defect each protects — is developed in math/differential-geometry: homotopy and defects and solitons.

Solitons: static topological lumps

Solitons are static, finite-energy, spatially localized solutions of the classical field equations, stabilized by topology. Examples by dimension:

Their masses scale as $M\sim \Lambda /g^2$ (inverse coupling), so they are heavy at weak coupling and invisible to perturbation theory, which expands in positive powers of $g$. The 't Hooft–Polyakov monopole is especially significant: any grand unified theory that breaks a simple group down to include $U(1{)}_{\text{EM}}$ necessarily contains magnetic monopoles, and explains the quantization of electric charge via the Dirac condition.

Instantons: tunneling in imaginary time

Instantons are localized solutions of the Euclidean field equations — finite-action configurations in the Wick-rotated theory. They are not particles; they are tunneling events between topologically distinct vacua, and they contribute to the path integral with a weight

$$e^{-S_E}\sim e^{-8{\pi }^2/g^2},$$

an essential singularity in the coupling that no order of perturbation theory can reproduce — the mathematical signature of a genuinely nonperturbative effect. In Yang–Mills theory the instanton has unit topological charge, $Q=\frac{g^2}{32{\pi }^2}\int {ϵ}^{\mu \nu \rho \sigma }{F}_{\mu \nu }^a{F}_{\rho \sigma }^a=1$ — the same $ϵFF$ density that appears in the chiral anomaly.

The $\theta$-vacuum and strong CP

Because instantons connect vacua of different winding number, the true QCD vacuum is a superposition — the $\theta$-vacuum — labeled by an angle $\theta$. Its effect is a new term in the Lagrangian:

$${\mathcal{L}}_{\theta }=\theta \frac{g^2}{32{\pi }^2}{ϵ}^{\mu \nu \rho \sigma }{F}_{\mu \nu }^a{F}_{\rho \sigma }^a.$$

This term is a total derivative (so it never appears in perturbation theory) but is physical nonperturbatively, and it violates $CP$ (see C, P, T on fields). It would give the neutron an electric dipole moment; experiment bounds $\theta \lesssim 10^{-10}$. Why is $\theta$ so absurdly small? — the strong-CP problem, one of the two CP problems of the Standard Model. The leading proposed solution (Peccei–Quinn symmetry) predicts a new light pseudo-Goldstone boson, the axion, also a dark-matter candidate.

The $U(1{)}_A$ problem

Instantons also resolve a puzzle in the chiral anomaly: the axial $U(1{)}_A$ symmetry appears spontaneously broken, which would require a ninth light pseudo-Goldstone (a light ${\eta }^{\prime }$ meson) — but the ${\eta }^{\prime }$ is anomalously heavy. The resolution ('t Hooft): the anomaly plus instantons explicitly break $U(1{)}_A$, so there is no Goldstone boson and the ${\eta }^{\prime }$ gets a large mass. Anomaly and instanton are two faces of the same topological structure.

Summary

• Topological charge stabilizes configurations that cannot deform to the vacuum — not a Noether charge.
• Solitons (kinks, vortices, monopoles, skyrmions) are static topological lumps, heavy at weak coupling.
• Instantons are Euclidean tunneling events with weight $e^{-8{\pi }^2/g^2}$ — invisible to perturbation theory.
• Instantons give the $\theta$-vacuum, the strong-CP problem (axion), and resolve the $U(1{)}_A$ problem.

Where this leads

• The anomaly connection: the chiral anomaly.
• Nonperturbative computation of these effects: lattice field theory.
• The CP structure of the full theory: the Standard Model.

References

• Rajaraman, Solitons and Instantons.
• 't Hooft, Phys. Rev. D 14, 3432 (1976).
• Coleman, Aspects of Symmetry, Ch. 6–7.
• Weinberg, The Quantum Theory of Fields, Vol. 2, Ch. 23.