BPS Boundaries, Surface Defects, Interfaces, and Fusion
Supersymmetric boundaries, surface defects, and interfaces are coupled lower-dimensional quantum systems, not merely singular labels in a bulk theory. A complete specification includes the preserved subalgebra, bulk boundary conditions, localized fields and interactions, anomaly inflow, global form, counterterms, moduli, orientations, and junction rules. Fusion is then a short-distance operation that can produce sums, new localized modes, or noninvertible defects.
Required background. Use the complete BPS line specification and the general treatment of boundaries, interfaces, and walls.
Helpful background. Duality walls and fusion explain when an interface represents an invertible equivalence.
Preserved supersymmetry and the variational problem
Section titled “Preserved supersymmetry and the variational problem”Place a boundary at . Varying the bulk action gives
A boundary condition is admissible only if the total boundary variation vanishes after adding boundary fields and an action . Supersymmetry adds the requirement
for a specified subspace of supercharges. Bosonic Dirichlet or Neumann labels alone do not determine the compatible fermion projectors, scalar conditions, and boundary interactions Gaiotto and Witten 2009, §§2–3.
For an interface between theories and , the folding trick replaces by its orientation reversal and treats the interface as a boundary of . This is useful only if orientation-dependent Chern–Simons and anomaly signs are also reversed.
Localized degrees of freedom and anomaly inflow
Section titled “Localized degrees of freedom and anomaly inflow”Chiral boundary or defect modes can carry anomalies. In anomaly-polynomial notation, consistency requires
For an interface, the bulk contribution is the oriented difference
This condition includes gauge, flavor, gravitational, and higher-form anomalies relevant to the preserved symmetry. A local counterterm can move a contact term between bulk and defect, but cannot remove a genuine anomaly. Omitting a localized fermion can therefore make an apparently supersymmetric boundary condition inconsistent.
Boundary global symmetries may be gauged by the bulk field. Their global form and allowed bundles must match: equality of Lie algebras does not guarantee a well-defined coupling of line endpoints or monopole sectors.
A half-BPS surface-operator benchmark
Section titled “A half-BPS surface-operator benchmark”Near a codimension-two surface in four-dimensional gauge theory, use polar coordinates and in the transverse plane and choose a Levi subgroup . A tame half-BPS surface operator has singular data
where is the Hitchin Higgs one-form. For a defect of fixed Levi type, are -invariant; after choosing Lie-algebra representatives they lie in the center , with the residual normalizer and lattice identifications imposed. A two-dimensional theta parameter for the Abelian part of couples to the magnetic flux through the surface. Thus the defect is labeled by
The monodromy is , so is periodic under the cocharacter lattice, while lies in the dual torus. S-duality exchanges electric and magnetic parameters and replaces by the appropriate Langlands-dual global theory Gukov and Witten 2008, §§2–3.
Equivalent surface operators can also be realized by a two-dimensional supersymmetric theory coupled to the bulk gauge field. The singular and coupled-QFT descriptions agree only in a stated parameter chamber and can differ by localized massive sectors or contact terms.
Junctions and operator categories
Section titled “Junctions and operator categories”Defects support their own local and extended operators. A line ending on a boundary becomes a boundary-changing operator; a junction between interfaces is a codimension-two morphism. Composition therefore forms a category or higher category rather than a set of numbers.
For interfaces and , fusion is the short-distance limit
The limit can require new counterterms and can leave light modes trapped between the walls. In a semisimple protected sector one may find
but a continuum, extensions, or derived structure can replace the direct sum.
An interface is invertible only if there is another interface whose fusion yields the transparent defect, including all localized sectors and global backgrounds. A wall whose fusion produces a sum of symmetry defects is noninvertible even if it acts invertibly on a restricted set of local operators.
Protected defect observables
Section titled “Protected defect observables”Depending on the preserved supercharge, one can compute:
- hemisphere wavefunctions and gluing kernels;
- boundary or defect indices;
- defect sphere partition functions;
- junction OPE coefficients;
- actions on bulk line and local operators;
- anomaly coefficients and displacement-multiplet data.
These quantities probe different structures. Equality of hemisphere partition functions does not by itself identify full boundary operator algebras, while matching anomaly inflow does not fix dynamics.
A complete specification
Section titled “A complete specification”Record the bulk theory on each side, orientation, preserved superalgebra, global symmetries and bundles, boundary conditions for every bulk field, localized QFT, superpotential and gauge couplings, anomaly counterterms, continuous and discrete defect parameters, allowed line endpoints, and fusion convention. Then check:
- the variational principle and supersymmetry variations;
- gauge and global anomaly cancellation;
- charge quantization and global form;
- localized zero modes and moduli;
- orientation reversal;
- protected observables in a solvable limit;
- junction associativity and the proposed duality image.
Exercises
Section titled “Exercises”Why does an interface anomaly contain rather than ?
Solution
Folding reverses the orientation of theory . Anomaly inflow changes sign under orientation reversal, so the two bulk contributions arrive at the wall with opposite signs. Localized wall fields must cancel the difference.
References
Section titled “References”- Gaiotto, D., and E. Witten. “Supersymmetric Boundary Conditions in Super Yang–Mills Theory.” Journal of Statistical Physics 135 (2009): 789–855. DOI; Open PDF.
- Gukov, S., and E. Witten. “Gauge Theory, Ramification, and the Geometric Langlands Program.” Current Developments in Mathematics 2006 (2008): 35–180. DOI; Open PDF.