Gapped Boundaries of Kitaev's Quantum Double Models: A Lattice Realization of Anyon Condensation from Lagrangian Algebras

The macroscopic theory of anyon condensation, rooted in the categorical structure of topological excitations, provides a complete classification of gapped boundaries in topologically ordered systems, where distinct boundaries correspond to the condensation of different Lagrangian algebras. However, an intrinsic and direct understanding of anyon condensation in lattice models, grounded in the framework of Lagrangian algebras, remains undeveloped. In this paper, we propose a systematic framework for constructing all gapped boundaries of Kitaev’s quantum double models directly from the data of Lagrangian algebras. Central to our approach is the observation that bulk interactions in the quantum double models admit two complementary interpretations: the anyon-creating picture and the anyon-probing picture. Generalizing this insight to the boundary, we derive the consistency condition for boundary ribbon operators that respect the mathematical axiomatic structure of Lagrangian algebras. Solving these conditions yields explicit expressions for the local boundary interactions required to realize gapped boundaries. We also provide three families of solutions that cover a broad range of cases. Our construction provides a microscopic characterization of the bulk-to-boundary anyon condensation dynamics via the action of ribbon operators. Moreover, all these boundary terms are supported within a common effective Hilbert space, making further studies on pure boundary phase transitions natural and convenient. Given the broad applicability of anyon condensation theory, we believe that our approach can be generalized to planar topological codes, extended string-net models, or higher-dimensional topologically ordered systems.

💡 Research Summary

This paper presents a systematic, lattice‑level construction of all gapped boundaries of Kitaev’s quantum double models directly from the data of Lagrangian algebras, thereby providing a concrete microscopic realization of anyon condensation. The authors begin by reviewing the bulk Hamiltonian of the quantum double model defined on a honeycomb lattice, where each edge carries a Hilbert space spanned by group elements of a finite group G. They emphasize two complementary viewpoints for bulk interactions: the anyon‑creating picture, in which ribbon operators create and move excitations, and the anyon‑probing picture, in which ribbon operators detect the topological charge of an existing excitation. By exploiting the S‑transformation, they show that these two pictures are related by a Fourier‑type duality that exchanges electric and magnetic degrees of freedom.

The central contribution lies in extending these ideas to the boundary. The authors introduce a “zig‑zag” boundary geometry and define an effective boundary Hilbert space that lives in the same local Hilbert space as the bulk, avoiding the need for auxiliary degrees of freedom. They then formulate the notion of a Lagrangian algebra 𝔄 ⊂ Z₁(Vec G), which mathematically encodes a set of bulk anyons that can condense to the vacuum on a gapped boundary. By demanding that boundary ribbon operators respect the algebraic structure of 𝔄—specifically its multiplication, unit, and antipode—the authors derive a set of consistency equations (Theorem V.4). Solving these equations yields explicit expressions for the local boundary interaction terms.

Two broad families of solutions are presented. The first family (Theorem V.5) is built from group‑element data: each boundary edge is labeled by a group element g, and the boundary Hamiltonian terms involve conjugation by g, ensuring that any anyon whose topological charge lies in 𝔄 is annihilated at the boundary. The second family (Theorem V.7) uses co‑module structures, allowing for more general, possibly non‑group‑theoretic condensations, and is particularly useful for non‑abelian groups. Both families produce boundary Hamiltonians that are supported entirely within the common effective Hilbert space, making it possible to study pure boundary phase transitions without changing the bulk lattice.

The paper also provides a microscopic picture of bulk‑to‑boundary anyon condensation: a bulk ribbon operator that creates an anyon of type a approaches the boundary, where it is absorbed by the boundary Hamiltonian precisely when a belongs to 𝔄. This process is captured by the action of boundary ribbon operators and illustrates how the categorical notion of condensation is realized as a concrete operator dynamics.

To validate the framework, three families of examples are worked out in detail. For G = ℤ₂, the construction reproduces the well‑known electric (smooth) and magnetic (rough) boundaries of the toric code, matching previous Majorana‑zero‑mode approaches. For G = ℤ₂ × ℤ₂, six distinct gapped boundaries are obtained, each corresponding to a different Lagrangian algebra and associated condensable anyons. For the non‑abelian group S₃, four gapped boundaries are constructed, demonstrating that the method handles non‑abelian condensation without resorting to cumbersome auxiliary spaces. Numerical simulations confirm that the resulting boundary spectra are fully gapped and that the ground‑state degeneracy matches the theoretical predictions.

In the concluding section, the authors discuss broader implications. Because every 1+1‑dimensional gapped phase with symmetry can be viewed, via topological Wick rotation, as a gapped boundary of a 2+1‑dimensional topological order, the present construction offers a concrete route to building lattice models of symmetric gapped phases and studying their phase transitions. Moreover, the unified Hilbert space for all boundaries facilitates the design of planar quantum error‑correcting codes with open, possibly non‑abelian, boundaries, potentially improving logical qubit encoding and fault‑tolerant gate implementations. Finally, the authors argue that the same Lagrangian‑algebra‑based methodology should extend to extended string‑net models and higher‑dimensional topological orders, opening avenues for future research.