Braided Sectors, Anyonic Statistics, and Low-Dimensional Nets

Low-dimensional localization changes exchange topology. In a chiral net, or for cone-localized charges in $2+1$ dimensions, one exchange path cannot generally be deformed into its reverse without crossing a charge. Locality therefore produces a unitary braiding rather than a symmetric flip. Abelian anyons carry phases; non-Abelian sectors carry higher-dimensional braid-group representations. Fusion, braiding, twist, and conjugation are related but none is determined by fusion rules alone.

Required background. Sector Selection, Localization, and Transportability supplies localization and transportability; Endomorphisms, Intertwiners, and Tensor Products supplies the tensor category; Conjugates, Statistics Operators, and Statistical Dimension supplies conjugates and exchange operators.

Helpful background. Anyons as Quasiparticles gives the condensed-matter interpretation; Linking, Braiding, and Framing explains topological exchange data; Fermionic Subsystems, Parity, and Graded Tensor Products contrasts braiding with a fermionic sign.

Oriented exchange from localization

Let $\rho$ and $\sigma$ be transportable localized endomorphisms. Move them with intertwiners $U\in(\rho,\widetilde\rho)$ and $V\in(\sigma,\widetilde\sigma)$ into causally disjoint regions with a chosen left/right or clockwise/counterclockwise order. One convenient exchange operator is

$$\varepsilon(\rho,\sigma) =V^*\sigma(U^*)U\rho(V) \in(\rho\sigma,\sigma\rho).$$

Locality proves independence from auxiliary transports within the same homotopy class and gives naturality and the hexagon identities. Consequently adjacent exchanges satisfy

$$b_i b_{i+1}b_i=b_{i+1}b_i b_{i+1}, \qquad b_i b_j=b_j b_i\quad (|i-j|\ge2),$$

the braid-group relations. What fails is $b_i^2=1$. The general superselection construction and positive-statistics structure are developed in Fredenhagen, Rehren, and Schroer 1989, §§2–4, pp. 204–221.

For simple sectors, the twist $\theta_\rho$ is the phase associated with a $2\pi$ rotation or framed self-exchange. Double braiding, $M_{\rho,\sigma}=\varepsilon(\sigma,\rho)\varepsilon(\rho,\sigma)$, is invariant under reversing conventions in a way a single exchange phase may not be. The balancing relation ties $M$ to twists on fusion channels. These data make orientation visible.

An integral U(1) current example

Consider the rational even-lattice extension of the chiral $U(1)$ current associated with $L=\sqrt{2k}\,\mathbb Z$, $k\in\mathbb N$. Its irreducible sectors are labeled by $n\in L^*/L\cong\mathbb Z_{2k}$. Their fusion and conformal weights are

$$n\otimes m=n+m\pmod{2k}, \qquad h_n=\frac{n^2}{4k}\pmod1.$$

Thus every sector has $d_n=1$, conjugation is $\bar n=-n$, and

$$\theta_n=e^{2\pi i h_n}=e^{\pi i n^2/(2k)}, \qquad M_{n,m}=e^{2\pi i(h_{n+m}-h_n-h_m)}=e^{\pi i nm/k}.$$

At $k=1$, the nontrivial class $n=1$ has twist $i$ and self-monodromy $-1$: it is neither an ordinary boson nor fermion despite having dimension one. Localized automorphisms of the $U(1)$ current and their braid relations are constructed in Buchholz, Mack, and Todorov 1990, §§2–4, pp. 359–373. The current-algebra and WZW setting is developed in Affine Current Algebras and WZW Models.

Domain, result, and nonconverse

The braided conclusion requires transportable localization in regions whose causal-complement configuration space retains an orientation class, plus locality and coherence of charge transporters. Finite statistics and conjugates yield a rigid braided category; complete rationality of a chiral net adds finitely many simples and nondegenerate braiding, producing a unitary modular tensor category.

Braiding does not follow from low dimension alone: the chosen representation class and localization criterion are essential. Conversely, a braid-group representation does not construct a local net. Fusion coefficients do not determine $F$-symbols, $R$-symbols, or twists; even equal fusion rings can support inequivalent braided structures. Modular data also need not classify the net.

The distinction between one exchange and monodromy is both mathematical and operational. A single-exchange phase can depend on transporter or framing conventions, while the double exchange compares a charge carried completely around another. In an Abelian sector it is a scalar of mutual statistics; in a non-Abelian sector it acts on a specified fusion space. A complete calculation must therefore name the fusion channel on which each braid operator acts.

Adversarial failure: erasing orientation

Suppose the $k=1$ sector above is assigned the ordinary sign $+1$ because it has $d=1$. Then a double exchange would be $+1$. The sector calculation gives $M_{1,1}=-1$. Assigning sign $-1$ fails as well, because its square is again $+1$. The contradiction is independent of single-exchange phase convention. It pinpoints the invalid step: quotienting the braid group by $b_i^2=1$ without a spacetime deformation that licenses that relation.

Independent checks

Verify the Yang–Baxter and hexagon relations for the proposed exchange. Check the balancing identity on every fusion channel. Reverse the orientation and confirm that braiding is inverted while fusion is unchanged. For a rational net, compute the Müger center: nondegeneracy requires every transparent simple object to be the tensor unit.

Exercises

1. Monodromy from weights. Derive $M_{n,m}=e^{\pi i nm/k}$ from the stated conformal weights.

Solution

$h_{n+m}-h_n-h_m=((n+m)^2-n^2-m^2)/(4k)=nm/(2k)$. Multiplication by $2\pi i$ gives the claimed phase.

2. Nondegeneracy. Show that a label $n\in\mathbb Z_{2k}$ transparent to every $m$ must be $n=0$.

Solution

Transparency requires $e^{\pi i nm/k}=1$ for all $m$. Taking $m=1$ gives $n=0$ modulo $2k$, so only the tensor unit is transparent.

3. Orientation reversal. What happens to $M_{n,m}$ when all braids are reversed?

Solution

Each braiding operator is inverted, hence the scalar monodromy becomes its inverse $M_{n,m}^{-1}=\overline{M_{n,m}}$. Fusion and dimensions are unaffected.

References

• Buchholz, Detlev, Gerhard Mack, and Ivan Todorov. “The Current Algebra on the Circle as a Germ of Local Field Theories.” Nuclear Physics B, Proceedings Supplements 5B (1988): 20–56. DOI.
• Buchholz, Detlev, Gerhard Mack, and Ivan Todorov. “Localized Automorphisms of the U(1)-Current Algebra on the Circle: An Instructive Example.” In The Algebraic Theory of Superselection Sectors, edited by Daniel Kastler, 356–378. Singapore: World Scientific, 1990. Book DOI.
• Fredenhagen, Klaus, Karl-Henning Rehren, and Bert Schroer. “Superselection Sectors with Braid Group Statistics and Exchange Algebras. I. General Theory.” Communications in Mathematical Physics 125 (1989): 201–226. DOI.