The three-dimensional classical Z2 gauge-Higgs model (equivalently, the 2+1-dimensional toric code in a magnetic field with components hxh_x and hzh_z acting on both spin species) has a topologically ordered deconfined phase bounded by two Ising-type continuous transition lines and a first-order transition segment lying on the self-dual line. Tupitsyn, Kitaev, Prokof'ev, and Stamp (2010) left open whether all three lines meet at a single multicritical point or whether the first-order segment terminates away from the point where the two second-order lines touch the self-dual line, and posed the construction of an effective field theory for this multicritical region — where charge and vortex condensates with nontrivial mutual braiding statistics compete with a first-order line — as an open problem. Later work provided strong numerical evidence for a single, continuous, scale-invariant multicritical point in the isotropic classical model, but its local line-meeting topology has not been fixed beyond finite-size ambiguity, and its universality class remains actively contested: proposed descriptions include a multicritical XY (O(2)) Landau-Ginzburg-Wilson theory, a distinct fixed point without a known continuum Lagrangian, and, most recently, an SO(4)SO(4) Chern-Simons-Higgs theory (a topological gauge theory whose Chern-Simons term encodes anyonic mutual statistics). In the anisotropic quantum (parallel-field toric code) model, the local topology of the line meeting remains numerically unresolved. The problem is to determine unambiguously the local phase-diagram topology and the continuum field theory (universality class, operator content, critical exponents) of this topological multicritical point, in both the isotropic classical and anisotropic quantum realizations.

Background

The toric code is an exactly solvable spin-1/2 model on the bonds of a two-dimensional lattice, with Hamiltonian HTC=−Jx∑sAs−Jz∑pBpH_{TC}=-J_x\sum_s A_s-J_z\sum_p B_p, where AsA_s and BpB_p are products of Pauli operators on the bonds touching a site ss and encircling a plaquette pp. Its ground state is topologically ordered: on a torus it is fourfold degenerate, and its excitations are gapped quasiparticles — Z2Z_2 electric charges (site defects As=−1A_s=-1) and Z2Z_2 magnetic vortices (plaquette defects Bp=−1B_p=-1). Each species is individually bosonic, but braiding a charge around a vortex yields a phase −1-1; the two are therefore Abelian anyons (quasiparticles whose exchange statistics is neither bosonic nor fermionic) with nontrivial mutual statistics, and they cannot condense simultaneously. Applying fields −hx∑bσbx−hz∑bσbz-h_x\sum_b\sigma^x_b-h_z\sum_b\sigma^z_b destroys the topological phase: hzh_z proliferates virtual charge pairs, driving a Higgs (charge-condensation) transition, while by duality hxh_x drives vortex condensation and charge confinement. Fradkin and Shenker showed for lattice gauge theories with Higgs fields that the Higgs and confinement regimes of such Z2Z_2 systems are analytically connected, so only the topological phase is a genuine thermodynamic phase; the two condensation transitions are nonetheless distinct continuous transitions.

Each condensation transition, when the other anyon species stays massive, has 3D Ising critical exponents in a 'gauged' variant (called Ising* by later authors, since only the gauge-invariant, Z2Z_2-even operators of the Ising conformal field theory survive as local operators). Via a Suzuki-Trotter path-integral construction — a mapping that rewrites a dd-dimensional quantum model as a (d+1)(d+1)-dimensional classical statistical model — the quantum model maps onto an anisotropic classical three-dimensional Z2Z_2 gauge-Higgs model with couplings λbond\lambda_{bond} (matter/gauge stiffness) and λpl\lambda_{pl} (plaquette stiffness); the isotropic model H=−λbond∑bSb−λpl∑p∏j∈pSjH=-\lambda_{bond}\sum_b S_b-\lambda_{pl}\sum_p\prod_{j\in p}S_j (Wegner's 3D Ising gauge theory, i.e. Z2Z_2 spins on bonds interacting through plaquette products) is expected to have the same phase-diagram topology. This classical model is self-dual: it maps to itself under λbond,pl→−12ln⁡tanh⁡(λpl,bond)\lambda_{bond,pl}\to-\tfrac12\ln\tanh(\lambda_{pl,bond}), leaving invariant a self-dual line in the (λbond,λpl)(\lambda_{bond},\lambda_{pl}) plane. Duality fixes one Ising critical point from the exactly known 3D Ising transition (λbond=0.221655\lambda_{bond}=0.221655) and its dual partner.

Monte Carlo simulations by Tupitsyn, Kitaev, Prokof'ev, and Stamp on lattices up to 60360^3 found that the transition lines remain second order until merging into a first-order line that lies along the self-dual line for 0.22635(5)<λbond<0.2575(5)0.22635(5)<\lambda_{bond}<0.2575(5) (identified by bi-modal energy distributions), separating the charge-condensed and vortex-condensed regions in a liquid-gas-type transition. They ruled out a splitting of the first-order line suggested by earlier, much smaller simulations (Jongeward, Stack, and Jayaprakash, 1980), but could not distinguish three remaining topologies: (1) all three lines meet at a single multicritical point; (2) or (3) the first-order segment terminates before or after the point where the two second-order lines touch the self-dual line — the two Ising lines approach the self-dual line at an extremely small (possibly zero) angle, and finite-size effects obscure the topology. They also noted that the two Ising order parameters (charge and vortex condensate amplitudes) are classically incompatible because of the nontrivial braiding, so that an effective field theory for the multicritical region was unknown.

Problem Statement

Determine, unambiguously and at the level of universal behavior, the local structure of the topological multicritical point of the Z2 gauge-Higgs system — the point where the topologically ordered (deconfined) phase, the two Ising-type charge- and vortex-condensation transition lines, and the first-order transition segment lying on the self-dual line come together — in both the isotropic three-dimensional classical model and the anisotropic (quantum toric code in a parallel field) realization. Specifically: (1) establish whether the two second-order Ising-type lines and the self-dual first-order segment meet at a single multicritical point, or whether the first-order segment terminates at a point distinct from where the two second-order lines touch the self-dual line (the existing Monte Carlo data are consistent with the single-meeting-point picture but cannot distinguish the alternatives, since the Ising lines approach the self-dual line at an extremely small angle and finite-size effects can obscure the topology, and finite sizes cannot rigorously exclude an extremely weak first-order transition); (2) construct (or definitively identify) an effective continuum field theory for the multicritical region, where two classically incompatible Ising orders — the charge and vortex condensates, quasiparticles with nontrivial mutual braiding that cannot condense simultaneously — compete with a first-order line, including its universality class, relevant operator content, and critical exponents, thereby resolving the current disagreement between the proposed descriptions (multicritical XY/O(2) Landau-Ginzburg-Wilson theory; a distinct non-XY fixed point with no known Lagrangian; U(1)×U(1)U(1)\times U(1) mutual Chern-Simons frameworks; SO(4)2,−2SO(4)_{2,-2} Chern-Simons-Higgs theory (Chern-Simons theories being topological gauge theories whose statistical term directly encodes the anyonic mutual braiding)); and (3) settle the corresponding questions in the anisotropic quantum model — the toric code in a parallel magnetic field — where the phase diagram was only predicted from the isotropic classical model under an assumed universality, and where existing quantum Monte Carlo data explicitly could not resolve how the three transition lines merge. The three parts share one objective: a complete, verified description (topology plus universal critical behavior) of this paradigmatic unconventional multicritical point — the local phase-diagram topology fixed beyond finite-size ambiguity, and the multicritical behavior given by a controlled continuum description in both realizations.

References

  1. I. S. Tupitsyn, A. Kitaev, N. V. Prokof'ev, P. C. E. Stamp, 'Topological multicritical point in the phase diagram of the toric code model and three-dimensional lattice gauge Higgs model', Phys. Rev. B 82, 085114 (2010), DOI: 10.1103/PhysRevB.82.085114, https://doi.org/10.1103/PhysRevB.82.085114 (arXiv:0804.3175, https://arxiv.org/abs/0804.3175)
  2. E. Fradkin, S. Shenker, 'Phase diagrams of lattice gauge theories with Higgs fields', Phys. Rev. D 19, 3682 (1979), DOI: 10.1103/PhysRevD.19.3682, https://doi.org/10.1103/PhysRevD.19.3682
  3. G. A. Jongeward, J. D. Stack, C. Jayaprakash, 'Monte Carlo calculations on Z2 gauge-Higgs theories', Phys. Rev. D 21, 3360 (1980), DOI: 10.1103/PhysRevD.21.3360, https://doi.org/10.1103/PhysRevD.21.3360
  4. J. Vidal, S. Dusuel, K. P. Schmidt, 'Low-energy effective theory of the toric code model in a parallel magnetic field', Phys. Rev. B 79, 033109 (2009), DOI: 10.1103/PhysRevB.79.033109, https://doi.org/10.1103/PhysRevB.79.033109
  5. F. Wu, Y.-J. Deng, N. V. Prokof'ev, 'Phase diagram of the toric code model in a parallel magnetic field', Phys. Rev. B 85, 195104 (2012), DOI: 10.1103/PhysRevB.85.195104, https://doi.org/10.1103/PhysRevB.85.195104 (arXiv:1201.6409, https://arxiv.org/abs/1201.6409)
  6. A. M. Somoza, P. Serna, A. Nahum, 'Self-Dual Criticality in Three-Dimensional Z2 Gauge Theory with Matter', Phys. Rev. X 11, 041008 (2021), DOI: 10.1103/PhysRevX.11.041008, https://doi.org/10.1103/PhysRevX.11.041008 (arXiv:2012.15845, https://arxiv.org/abs/2012.15845)
  7. C. Bonati, A. Pelissetto, E. Vicari, 'Multicritical point of the three-dimensional Z2 gauge Higgs model', Phys. Rev. B 105, 165138 (2022), DOI: 10.1103/PhysRevB.105.165138, https://doi.org/10.1103/PhysRevB.105.165138 (arXiv:2112.01824, https://arxiv.org/abs/2112.01824)
  8. L. Oppenheim, M. Koch-Janusz, S. Gazit, Z. Ringel, 'Machine learning the operator content of the critical self-dual Ising-Higgs lattice gauge theory', Phys. Rev. Research 6, 043322 (2024), DOI: 10.1103/PhysRevResearch.6.043322, https://doi.org/10.1103/PhysRevResearch.6.043322 (arXiv:2311.17994, https://arxiv.org/abs/2311.17994)
  9. C. Bonati, A. Pelissetto, E. Vicari, 'Comment on "Machine Learning the Operator Content of the Critical Self-Dual Ising-Higgs Gauge Model"', arXiv:2401.10563 (2024), https://arxiv.org/abs/2401.10563
  10. Z. D. Shi, A. Chatterjee, 'Analytic framework for self-dual criticality in ZkZ_k gauge theory with matter', Phys. Rev. B 112, L081111 (2025), DOI: 10.1103/9qrw-p5zn, https://doi.org/10.1103/9qrw-p5zn (arXiv:2407.07941, https://arxiv.org/abs/2407.07941)
  11. W. Ji, R. A. Lanzetta, Z. Zhou, C. Wang, 'Self-Dual Higgs Transitions: Toric Code and Beyond', Phys. Rev. Lett. 137, 086505 (2026), DOI: 10.1103/flfk-m8lj, https://doi.org/10.1103/flfk-m8lj (arXiv:2601.20945, https://arxiv.org/abs/2601.20945)
  12. C. Bonati, A. Pelissetto, E. Vicari, 'Three-dimensional Abelian and non-Abelian gauge Higgs theories', Physics Reports 1133, 1-92 (2025), DOI: 10.1016/j.physrep.2025.05.004, https://doi.org/10.1016/j.physrep.2025.05.004 (arXiv:2410.05823, https://arxiv.org/abs/2410.05823)
  13. P. Serna, A. M. Somoza, A. Nahum, 'Worldsheet patching, 1-form symmetries, and "Landau-star" phase transitions', Phys. Rev. B 110, 115102 (2024), DOI: 10.1103/PhysRevB.110.115102, https://doi.org/10.1103/PhysRevB.110.115102 (arXiv:2403.04025, https://arxiv.org/abs/2403.04025)
  14. S. M. Linsel, L. Pollet, F. Grusdt, 'Independent e- and m-Anyon Confinement in the Parallel Field Toric Code on Non-Square Lattices', PRX Quantum 7, 010332 (2026), DOI: 10.1103/gtth-cclr, https://doi.org/10.1103/gtth-cclr (arXiv:2504.03512, https://arxiv.org/abs/2504.03512)
  15. U. Borla, A. De, S. Gazit, 'Odd toric code in a tilted field: Higgs-confinement multicriticality, spontaneous self-duality symmetry breaking, and valence bond solids', Phys. Rev. B 113, 245134 (2026), DOI: 10.1103/wbvd-z9s4, https://doi.org/10.1103/wbvd-z9s4 (arXiv:2507.16523, https://arxiv.org/abs/2507.16523)