Symmetry Defect: Theory, Methods & Applications

• Symmetry defects are defined by local failures or intentional perturbations of symmetry, influencing quantum transitions, field operators, and geometric invariants.
• They yield distinctive scaling behaviors in critical systems, as seen in fidelity susceptibilities and crossover regimes in models like the quantum Ising chain.
• Advanced frameworks, including Symmetry TFT and non-invertible fusion rules, enable applications in gauge theories, topological phases, and quantum computing.

A symmetry defect is a defect, operator, or singular locus whose defining property is specified by a symmetry action, a symmetry-breaking perturbation, or a failure of symmetry-related data to remain locally constant. In current usage the term spans several distinct but related constructions: localized fields that break a global symmetry at a continuous quantum transition; codimension-one walls implementing discrete or continuous symmetries in quantum field theory; non-invertible, gauge, and duality defects encoded by Symmetry TFT; decorated defects carrying anomaly-protected states; and, in algebraic geometry, the symmetry defect hypersurface defined by the failure of a midpoint-counting function to be locally constant (Franchi et al., 2022, Müller et al., 2019, Kaidi et al., 2022, Cordova et al., 2019, Calvo et al., 6 Mar 2025, Janeczko et al., 2014).

1. Symmetry-breaking defects at criticality

In the one-dimensional quantum Ising models at their continuous quantum transition, symmetry-breaking defects arise from localized external fields coupled to the order-parameter operator. These defects do not alter the bulk critical power laws, but they induce notable critical crossovers in finite systems. For the Ising ring and chain, the bulk and boundary perturbations

$$\hat D_k=-\kappa\,\hat \sigma_k^{(1)},\qquad \hat B=-\zeta(\hat \sigma_1^{(1)}+\hat \sigma_L^{(1)})$$

are relevant in the renormalization-group sense, with scaling variables

$K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$

and RG dimensions

$$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$

In the Ising ring, tuning $\kappa$ interpolates between periodic boundary conditions and parallel fixed boundary conditions; in the scaling limit $\kappa\sim L^{-y_\kappa}$, observables exhibit a smooth RG crossover as functions of $K$ (Franchi et al., 2022).

The crossover is sharply characterized by the ground-state fidelity

$$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$

whose small-$\delta\kappa$ expansion defines the fidelity susceptibility $A_2$. Within the critical crossover regime,

$$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$

so that in the Ising chain $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$0 for a bulk defect and $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$1 for a boundary defect, whereas outside the critical defect regime $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$2. Exact diagonalization and DMRG up to $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$3 show data collapse on universal scaling curves as functions of $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$4 and $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$5, and the gap, magnetization, and correlation length conform to the same crossover scaling structure (Franchi et al., 2022).

A useful distinction emerging from adjacent literature is between imposed symmetry-breaking defects and defects generated during symmetry breaking. Under weak explicit symmetry breaking, defect formation across a continuous transition is governed by rare, large-deviation events, leading to

$K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$6

rather than pure Kibble-Zurek power-law scaling. In ion Coulomb crystals, by contrast, the linear-to-zigzag transition directly realizes symmetry breaking with topological kink formation, and the probability of a single defect in the DIKZM regime scales as $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$7 (Liu et al., 26 Jun 2026, Pyka et al., 2012).

2. Defects as symmetry operators, gauging, and orbifolding

In two-dimensional Yang-Mills theory, discrete symmetry defects are codimension-one domain walls associated with outer automorphisms of the gauge group $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$8. For each $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$9, the defect twists a $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$0-bundle with connection $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$1 to $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$2, so the path integral in the presence of a defect network is an integral over $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$3-twisted $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$4-bundles. The partition function can be computed exactly, and in the weak-coupling limit it computes the symplectic volume of the moduli space of flat twisted $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$5-bundles. Gauging the discrete symmetry by summing over defect networks produces an orbifold theory that is again two-dimensional Yang-Mills theory, now with gauge group $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$6 (Müller et al., 2019).

A more general condensation perspective identifies gauging itself with the insertion of a spacetime-filling defect. In that formulation, each gaugeable symmetry defines a $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$7-form symmetry, and the resulting gauge defects have a fusion product that is generally non-invertible. In two dimensions, the $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$8 example is explicit: $$y_\kappa=z-y_\varphi=1-\frac18=\frac78,\qquad y_\zeta=z-y_b=1-\frac12=\frac12.$$9 The same framework extends to non-invertible symmetries such as $\kappa$0, where simple gauge defects are organized by Morita equivalence classes of Frobenius algebras and their fusion rules exhibit nontrivial multiplicities (Vandermeulen, 2023).

For continuous global symmetries in bosonic field theories, the naive codimension-one operator

$\kappa$1

does not by itself reproduce the expected action in correlation functions. Starting at order $\kappa$2, contact terms appear, so the defect must be modified by a specific contact term. In the $\kappa$3 example with $\kappa$4, the corrected operator is

$\kappa$5

which is equivalent to coupling the theory to a singular background gauge field $\kappa$6 through the covariant derivative $\kappa$7. This identifies the continuous symmetry defect with a regulated background gauge-field insertion rather than a purely formal current exponential (Calvo et al., 6 Mar 2025).

3. Non-invertible symmetry defects and the Symmetry TFT

The Symmetry TFT program associates to any symmetry of a $\kappa$8-dimensional quantum field theory a $\kappa$9-dimensional topological field theory that captures generalized global symmetries and ’t Hooft anomalies. In this language, ordinary symmetry defects, duality defects, and higher duality defects are encoded as topological defects in the bulk SymTFT. For Kramers-Wannier-like dualities in $\kappa\sim L^{-y_\kappa}$0, the relevant symmetry category can be the Tambara-Yamagami category $\kappa\sim L^{-y_\kappa}$1, with invertible defects $\kappa\sim L^{-y_\kappa}$2 and a non-invertible defect $\kappa\sim L^{-y_\kappa}$3 obeying

$\kappa\sim L^{-y_\kappa}$4

In $\kappa\sim L^{-y_\kappa}$5, half-space gauging of a $\kappa\sim L^{-y_\kappa}$6-form $\kappa\sim L^{-y_\kappa}$7 symmetry produces codimension-one duality interfaces, and further EM gauging promotes them to genuinely non-invertible defects (Kaidi et al., 2022).

The SymTFT also supplies a computational notion of charge for general defects. For a codimension-$\kappa\sim L^{-y_\kappa}$8 defect $\kappa\sim L^{-y_\kappa}$9, the corresponding higher representations are in one-to-one correspondence with gapped boundary conditions for the dimensionally reduced SymTFT on

$K$0

This dimensional-reduction statement turns the classification of defect charges into a boundary-condition problem in $K$1, and it applies uniformly to point, line, and surface defects. The same framework yields a bulk-to-defect projection rule,

$K$2

which organizes the defect-local operator multiplets induced by a bulk charge $K$3 (Copetti, 2024).

A notable consequence is that anomaly constraints on defects are subtler than anomaly constraints on boundaries. The absence of a symmetric boundary condition in an anomalous bulk theory generalizes to defects of arbitrary codimension, but a symmetric defect can still exist when the anomaly trivializes upon reduction on the transverse sphere: $K$4 This is particularly relevant for surface charges of $K$5 duality symmetries and for the analysis of Gukov-Witten operators (Copetti, 2024).

4. Anomalies, decorated defects, and defect-localized spectra

Decorated $K$6 symmetry defects provide a precise bridge between bulk unitary $K$7 anomalies and lower-dimensional time-reversal anomalies. Coupling a $K$8-dimensional theory to a background $K$9 gauge field $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$0 creates a codimension-one symmetry defect on the Poincaré dual of $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$1. The defect worldvolume is naturally invariant under $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$2, and the bulk anomaly descends to a $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$3-anomaly on the defect. In $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$4 bosonic systems, the nontrivial bulk class in $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$5 leads to Kramers degeneracy in the symmetry-defect Hilbert space, with $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$6. In $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$7 bosonic systems, anomalies classified by $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$8 give symmetry defects that support anyons which can be either fermions or Kramers doublets (Cordova et al., 2019).

A closely related mechanism appears in three-dimensional symmetry protected topological phases coupled to a dynamical $$A(L,g,\kappa,\kappa+\delta\kappa)=\left|\langle \Psi_0(L,g,\kappa+\delta\kappa)\mid \Psi_0(L,g,\kappa)\rangle\right|,$$9 gauge field. There the vison loop excitation is a line defect that behaves as a “1d boundary” of the bulk SPT. For the $\delta\kappa$0 example, the vison loop supports a $\delta\kappa$1d $\delta\kappa$2 nonlinear sigma model with $\delta\kappa$3, and more generally one obtains lower-dimensional defect theories with a $\delta\kappa$4-term or a level-1 Wess-Zumino-Witten term. The resulting line defect is therefore guaranteed to have a gapless or degenerate spectrum protected by symmetry (Bi et al., 2013).

When a conformal defect breaks a bulk global symmetry $\delta\kappa$5 down to a subgroup $\delta\kappa$6, the broken-current conservation law acquires a defect-local contact term,

$\delta\kappa$7

where the $\delta\kappa$8 are exactly marginal defect operators. Their couplings parametrize a defect conformal manifold equal to the coset $\delta\kappa$9, and the Zamolodchikov metric is the two-point function of the $A_2$0. At a more global level, the symmetry breaking long exact sequence classifies domain walls, vortices, and hedgehogs by relating the anomaly of the defect core to the anomaly of the broken symmetry and to the residual family anomaly $A_2$1, whose nonvanishing obstructs the existence of a local symmetric defect (Drukker et al., 2022, Debray et al., 2023).

5. Geometric, structural, and computational meanings

In algebraic geometry, the phrase “symmetry defect” has a specific and independent meaning. For a smooth irreducible subvariety $A_2$2, one defines $A_2$3 as the number of unordered pairs $A_2$4 such that

$A_2$5

If $A_2$6 is in general position, then outside a certain closed algebraic hypersurface $A_2$7, the function $A_2$8 is constant and nonzero. The hypersurface $A_2$9 is the symmetry defect hypersurface: the locus where the symmetry count $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$0 fails to be locally constant. For generic members of an irreducible algebraic family of $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$1-dimensional smooth irreducible subvarieties in general position in $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$2, their symmetry defect hypersurfaces are homeomorphic, and their singular sets $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$3 are homeomorphic for every $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$4. In particular, generic plane curves in $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$5 of the same degree have symmetry defect curves with the same number of singular points (Janeczko et al., 2014).

In finite clusters and atomistic structures, symmetry breaking induced by defects can be quantified continuously by comparing a normalized electron-weighted atomic density $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$6 with its image under a symmetry operation $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$7. The Jensen-Shannon symmetry breaking measure is

$$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$8

with $$A_2(L,g,\kappa)\sim L^{2y_\kappa},\qquad A_2(L,g,\zeta)\sim L^{2y_\zeta},$$9. It vanishes if and only if $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$00, is non-negative, and is bounded, which allows different clusters and different symmetry operations to be compared on the same scale. The construction distinguishes surface-induced symmetry lowering, local atomic displacements, and collective motions such as octahedral tilts (Lan et al., 2024).

Electronic-structure workflows extend this diagnostic viewpoint to defect orbitals in semiconductors. ADAQ-SYM takes plane-wave DFT orbitals, determines the point group with AFLOW-SYM, computes symmetry operator expectation values, assigns irreducible representations through character projection, and applies selection rules to determine allowed optical transitions. In diamond and $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$01-SiC, this automated symmetry analysis reproduces the $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$02 and $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$03 assignments of the NV center, identifies inversion-constrained transitions for SiV defects, and explains the different zero phonon line polarization of the $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$04 and $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$05 divacancies in $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$06-SiC (Stenlund et al., 2023).

6. Engineered defect modes, robustness, and applications

In non-Hermitian electrical circuits, symmetry defects become experimentally tunable probes of the interplay between topology and $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$07-type symmetries. In a non-Hermitian SSH circuit with alternating gain and loss, edge states persist in the $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$08-symmetric, $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$09-broken, and anti-$K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$10-symmetric regimes, whereas the topological defect state exists only in the $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$11-symmetric and anti-$K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$12-symmetric regimes. It disappears in the $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$13-broken regime and reemerges at large gain/loss. The bulk classification is captured not by the Zak phase in all regimes, but by the $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$14 winding number

$K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$15

which remains quantized even when the Zak phase fails (Stegmaier et al., 2020).

In one-dimensional bi-component magnonic superlattices, periodically introduced defect stripes generate localized gap modes whose parity is controlled by defect geometry. For $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$16, the defect mode is antisymmetric; for $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$17, it is symmetric. Large defect separation drives a transition from dispersive to practically flat bands, with the defect-band bandwidth decaying exponentially with the separation parameter $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$18. The antisymmetric mode is comparatively insensitive to external field, while symmetric modes move toward band edges and can lose localization (Gallardo et al., 2018).

For topological quantum computing, symmetry defects are modeled by $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$19-crossed braided extensions of unitary modular tensor categories. The defect sectors $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$20 supplement the anyon sector $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$21, and the associated algebraic data produce projective unitary braid-group representations. This permits non-abelian statistics to emerge from abelian topological phases: in the toric-code example with $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$22 electromagnetic duality, the symmetry defects $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$23 have quantum dimension $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$24 and obey Ising-like fusion rules. More generally, symmetry defects can enlarge the gate set accessible to an anyonic device, including protocols that realize a logical $K=\kappa L^{y_\kappa},\qquad Z=\zeta L^{y_\zeta},$25-gate in bilayer Ising systems (Delaney et al., 2018).

Related rigorous analyses show that defects need not restore a broken symmetry. In two-dimensional crystals with isolated missing atoms, rotational symmetry remains spontaneously broken at low temperature, uniformly in system size, and local increments align with the standard lattice orientation in the low-temperature, large-volume, high-defect-cost limit. On infinite square grids, by contrast, missing-edge defects can either preserve or destroy the dimensional crossover depending on their topology: uniformly bounded or suitably non-bottlenecked defects preserve the two-dimensional Sobolev inequality, while bottleneck defects invalidate it (Heydenreich et al., 2013, Dovetta et al., 2021).