Microscopic and exactly-solvable realization of the symmetry-respecting topologically ordered surface phase of the three-dimensional electron topological insulator, and the minimal such surface order

A three-dimensional electron topological insulator (ETI) — a band insulator protected by charge conservation (U(1)) and time-reversal symmetry ( $T$ , with $T^2=(-1)^F$ on electrons) — was long believed to admit no symmetry-preserving gapped surface. Metlitski, Kane, and Fisher deduced such a surface phase: a superconducting surface driven through condensation of flux- $4\pi$ (or, via an intermediate 96-anyon phase, flux- $8\pi$ ) vortices enters an intrinsically topologically ordered surface state, Moore-Read $\times\,\mathrm{U}(1)_{-2}$ (24 anyon types, 12 modulo the electron), in which U(1) and $T$ are implemented in a way impossible in strictly two dimensions. Later literature has fixed the question of which closely related Walker-Wang bulk construction equals the noninteracting ETI (T-Pfaffian+ is the surface of the standard ETI; T-Pfaffian- corresponds to the ETI stacked with a bosonic SPT phase; the two variants carry the same 12-anyon order and differ only in the implementation of $T$ ), and now accepts the smaller 12-anyon T-Pfaffian order as the minimal candidate. What remains open is the original problem's realization core: no microscopic surface interactions or materials are known to energetically favor the requisite vortex condensation; no explicit exactly-solvable lattice construction of the full anomalous Moore-Read $\times\,\mathrm{U}(1)_{-2}$ surface phase with its precise symmetry action has been given; and no experimental probe has been shown to distinguish this termination from competing ones.

Background

An electron topological insulator (ETI) is a three-dimensional band insulator whose nontrivial topology is protected by charge conservation (a U(1) symmetry) and time-reversal symmetry $T$ with $T^2=(-1)^F$ on the electrons (symmetry class AII of the Altland-Zirnbauer classification). Its band structure gives an electromagnetic response characterized by a topological theta-angle $\theta=\pi$ , and its clean surface must host an odd number of massless Dirac cones: any surface mass term breaks $T$ or U(1). One way to gap the surface while keeping $T$ intact is superconductivity: in the Fu-Kane mechanism a proximity-coupled superconductor gaps the surface Dirac cone, preserving $T$ but breaking U(1) charge conservation, and vortices of the superconducting order parameter bind Majorana zero modes (zero-energy neutral fermionic states localized at vortex cores).

When interactions are allowed, a third route exists: a surface that preserves both symmetries but is gapped by intrinsic topological order — a two-dimensional phase with anyonic quasiparticles (superselection sectors with fractional statistics; the number of anyon types and the total quantum dimension $\mathcal{D}=\sqrt{\sum_a d_a^2}$ over the quantum dimensions $d_a$ are standard measures of the order's size). Such an order realizes the symmetries anomalously: it cannot occur in any strictly two-dimensional system with the same U(1) and $T$ , but it is allowed as the boundary of the three-dimensional ETI. Metlitski, Kane, and Fisher (Phys. Rev. B 92, 125111 (2015); arXiv:1306.3286) deduced a concrete such surface phase. From the superconducting surface it is reached through a direct second-order transition in the XY* universality class — an XY-like critical point at which the proliferating object is, from the superconducting side, a vortex of magnetic flux $4\pi$ , and, from the topologically ordered side, the condensation of a charge- $e/2$ boson; alternatively, condensing flux- $8\pi$ vortices first produces an intermediate 96-anyon phase (T96) from which further condensation yields the final order. The final phase is Moore-Read $\times\,\mathrm{U}(1)_{-2}$ : the Moore-Read sector (the non-Abelian topological order of the $\nu=1/2$ Pfaffian quantum Hall state) supports non-Abelian charge- $e/4$ anyons, while the Abelian sector $\mathrm{U}(1)_{-2}=\{1,\bar{s}\}$ is electrically neutral, its single nontrivial anyon $\bar{s}$ being an anti-semion (an Abelian anyon of exchange phase $-i$ ). It has 24 anyon types (12 modulo the electron) and total quantum dimension $\mathcal{D}=\sqrt{32}$ . A one-dimensional domain wall between this phase and the $T$ -breaking surface quantum Hall state with $\sigma_H=1/2$ carries a right-moving neutral Majorana mode, a right-moving charge mode, and a left-moving neutral mode. Concurrent works (Wang-Potter-Senthil; Bonderson-Nayak-Qi; Chen-Fidkowski-Vishwanath) obtained closely related results, including a rival 12-anyon (6 modulo the electron) termination — the T-Pfaffian order, a restriction of Ising $\times\,\mathrm{U}(1)_{-8}$ — constructed as the surface of an exactly-solvable three-dimensional Walker-Wang lattice model. A Walker-Wang model is a commuting-projector lattice Hamiltonian built on a braided tensor (fusion) category; when the category is non-modular (some anyons braid trivially with everything), the bulk confines and only a surface topological order survives — but which bulk phase the model realizes (the standard ETI, or the ETI stacked with a bosonic symmetry-protected topological (SPT) phase — an interacting phase with no intrinsic topological order that becomes trivial only when the symmetry is broken, such as eTmT, whose symmetry-preserving surface must stay gapless with mutually semionic electric $e$ and magnetic $m$ excitations) is not automatic. The interacting classification of three-dimensional electron insulators with U(1) and $T$ is $(\mathbb{Z}_2)^3$ (Wang-Potter-Senthil), with roots generated by the ETI and two interaction-only topological paramagnets. A 't Hooft anomaly is the obstruction that prevents a two-dimensional topological order with a given symmetry action from being realized in strictly two dimensions; anomaly indicators — functions of the anyon data, such as the modular S-matrix (the matrix of mutual braiding phases between anyon types), anyon electric charges, and Kramers parities (the eigenvalue of $T^2$ on a quasiparticle state) — were later developed to diagnose which candidate surface orders carry the ETI anomaly versus the ETI-plus-bosonic-SPT anomalies.

Problem Statement

The common objective is to establish the symmetry-respecting, intrinsically topologically ordered surface phase of the three-dimensional electron topological insulator — deduced at the field-theory level as the Moore-Read $\times\,\mathrm{U}(1)_{-2}$ order reached from the superconducting surface by flux- $4\pi$ (or, via the 96-anyon intermediate phase, flux- $8\pi$ ) vortex condensation, with its anomalous implementation of U(1) charge conservation and time-reversal symmetry $T$ — as a physically realizable phase, by answering the following linked questions left open by the source paper: (1) identify explicit microscopic surface interactions (or materials-realistic/engineered Hamiltonians) whose energetics favor the requisite vortex condensation over competing surface orders, and demonstrate in a controlled microscopic model that they realize the symmetry-preserving topologically ordered surface phase; (2) provide an explicit, fully worked-out exactly-solvable lattice construction realizing the complete Moore-Read $\times\,\mathrm{U}(1)_{-2}$ surface topological order together with the precise U(1) and $T$ symmetry action; the exactly-solvable Walker-Wang realization of the smaller 12-anyon T-Pfaffian+ order (the accepted minimal candidate) is already known and is not itself the open target of this sub-question; (3) determine with full rigor which symmetry-preserving gapped ETI surface order is minimal in the number of anyon types and the total quantum dimension — the constraint analysis of Cheng et al. (2026) identifies the T-Pfaffian order ( $\mathcal{D}^2=8$ , 6 anyons modulo the electron) as the unique minimal theory within their one-form-symmetry framework, but a framework-independent proof that no smaller order can carry the anomaly is still lacking — thereby settling the source paper's "smallest surface phase" question; and (4) specify experimental probes or transport signatures that could distinguish this termination from competing surface terminations in real ETI materials. Each of (1)-(4) is an independent sub-question that can be resolved on its own: a resolution of any one of them is a complete answer to that sub-question, and the four do not have to be settled jointly. Sub-questions concerning which closely related Walker-Wang bulk equals the noninteracting ETI (T-Pfaffian+ versus T-Pfaffian-, the latter being the ETI stacked with a bosonic SPT phase) have been settled by later literature and are not part of the open core.

References

Max A. Metlitski, C. L. Kane, Matthew P. A. Fisher, "Symmetry-respecting topologically ordered surface phase of three-dimensional electron topological insulators", Phys. Rev. B 92, 125111 (2015), DOI 10.1103/PhysRevB.92.125111, arXiv:1306.3286, https://arxiv.org/abs/1306.3286 (source paper)
Parsa Bonderson, Chetan Nayak, Xiao-Liang Qi, "A time-reversal invariant topological phase at the surface of a 3D topological insulator", J. Stat. Mech. P09016 (2013), DOI 10.1088/1742-5468/2013/09/P09016, arXiv:1306.3230, https://arxiv.org/abs/1306.3230
Xie Chen, Lukasz Fidkowski, Ashvin Vishwanath, "Symmetry enforced non-Abelian topological order at the surface of a topological insulator", Phys. Rev. B 89, 165132 (2014), DOI 10.1103/PhysRevB.89.165132, arXiv:1306.3250, https://arxiv.org/abs/1306.3250
Lukasz Fidkowski, Xie Chen, Ashvin Vishwanath, "Non-Abelian topological order on the surface of a 3D topological superconductor from an exactly solved model", Phys. Rev. X 3, 041016 (2013), DOI 10.1103/PhysRevX.3.041016, arXiv:1305.5851, https://arxiv.org/abs/1305.5851
Chong Wang, Andrew C. Potter, T. Senthil, "Gapped symmetry preserving surface state for the electron topological insulator", Phys. Rev. B 88, 115137 (2013), DOI 10.1103/PhysRevB.88.115137, arXiv:1306.3223, https://arxiv.org/abs/1306.3223
Chong Wang, Andrew C. Potter, T. Senthil, "Classification of interacting electronic topological insulators in three dimensions", Science 343, 629-631 (2014), DOI 10.1126/science.1243326, arXiv:1306.3238, https://arxiv.org/abs/1306.3238
C. W. von Keyserlingk, F. J. Burnell, "Walker-Wang models and axion electrodynamics", Phys. Rev. B 91, 045134 (2015), DOI 10.1103/PhysRevB.91.045134, arXiv:1405.2988, https://arxiv.org/abs/1405.2988
David F. Mross, Andrew Essin, Jason Alicea, "Composite Dirac liquids: Parent states for symmetric surface topological order", Phys. Rev. X 5, 011011 (2015), DOI 10.1103/PhysRevX.5.011011, arXiv:1410.4201, https://arxiv.org/abs/1410.4201
Chong Wang, T. Senthil, "Dual Dirac liquid on the surface of the electron topological insulator", Phys. Rev. X 5, 041031 (2015), DOI 10.1103/PhysRevX.5.041031, arXiv:1505.05141, https://arxiv.org/abs/1505.05141
Max A. Metlitski, "S-duality of u(1) gauge theory with theta=pi on non-orientable manifolds: Applications to topological insulators and superconductors", arXiv:1510.05663 (2015), DOI 10.48550/arXiv.1510.05663, https://arxiv.org/abs/1510.05663 (unpublished preprint, v1 only)
Matthew F. Lapa, Michael Levin, "Anomaly indicators for topological orders with U(1) and time-reversal symmetry", Phys. Rev. B 100, 165129 (2019), DOI 10.1103/PhysRevB.100.165129, arXiv:1905.00435, https://arxiv.org/abs/1905.00435
Jun Ho Son, Jason Alicea, "Commuting-projector Hamiltonians for two-dimensional topological insulators: Edge physics and many-body invariants", Phys. Rev. B 100, 155107 (2019), DOI 10.1103/PhysRevB.100.155107, arXiv:1906.11846, https://arxiv.org/abs/1906.11846
Qing-Rui Wang, Yang Qi, Chen Fang, Meng Cheng, Zheng-Cheng Gu, "Exactly solvable lattice models for interacting electronic insulators in two dimensions", Phys. Rev. B 108, L121104 (2023), DOI 10.1103/PhysRevB.108.L121104, arXiv:2112.15533, https://arxiv.org/abs/2112.15533
Jing-Ren Zhou, Zheng-Cheng Gu, "Exactly solvable models for fermionic symmetry-enriched topological phases and fermionic 't Hooft anomaly", Phys. Rev. X 16, 011019 (2026), DOI 10.1103/r6d4-b538, arXiv:2410.19126, https://arxiv.org/abs/2410.19126
Meng Cheng, Seth Musser, Amir Raz, Nathan Seiberg, T. Senthil, "Ordering the topological order in the fractional quantum Hall effect", Phys. Rev. B 113, 115103 (2026), DOI 10.1103/7gtq-y9d9, arXiv:2505.14767, https://arxiv.org/abs/2505.14767
Omid Tavakol, Thomas Scaffidi, "Pairing around a single Dirac point: A unifying view of Kohn-Luttinger superconductivity in Chern bands, quarter metals, and topological surface states", Phys. Rev. B 113, 144502 (2026), DOI 10.1103/1tg5-qhtf, arXiv:2508.13271, https://arxiv.org/abs/2508.13271