Kane and Lubensky (2013) showed that two-dimensional isostatic (Maxwell) lattices possess topologically protected zero-frequency boundary modes governed by a topological polarization vector and an index theorem, and posed four directions for future study: three-dimensional lattice models, topologically protected point zeros of det Q(k) (Dirac/Weyl analogues), an index-theoretic treatment of bulk-gapless and singular-boundary situations, and connections to frustrated magnetism. Later literature has already realized Weyl points in two dimensions and Weyl lines in three-dimensional pyrochlore generalizations, and has constructed three-dimensional topologically polarized phases. The open core is: a general index theorem and phase classification for Maxwell lattices whose bulk spectra contain topologically protected gapless loci (Weyl points in 2D, Weyl lines in 3D), including boundaries whose surface Brillouin zones are intersected by the projections of these loci, together with an explicit formulation of the correspondence between topological Maxwell lattices and frustrated magnetic systems.

Background

A frame, or mechanical lattice, is a network of mass points connected by central-force springs. Maxwell and Calladine counting relates the number of zero-frequency floppy modes N0N_0 (motions that stretch no springs) and states of self-stress NssN_{ss} (tension patterns with no net force on any site) to the number of sites NsN_s and bonds NbN_b in dd dimensions: N0−Nss=dNs−NbN_0 - N_{ss} = dN_s - N_b. A lattice is isostatic, or a Maxwell lattice, when every site has coordination z=2dz = 2d so that constraints and degrees of freedom balance exactly, placing the system on the verge of mechanical instability.

For a periodic lattice, the dynamical matrix has a chiral (sublattice) symmetry: it can be written as D=QTQD = Q^T Q, where the equilibrium matrix Q(k)Q(k) (the transpose of the compatibility matrix relating bond extensions to site displacements) is a square matrix at each Bloch wavevector kk in an isostatic lattice, so that det⁡Q(k)\det Q(k) is well defined. The number of bulk zero modes per unit cell is controlled by zeros of det⁡Q(k)\det Q(k). Kane and Lubensky showed that the winding numbers nin_i of the phase of det⁡Q(k)\det Q(k) around the independent cycles of the Brillouin zone (the torus of inequivalent Bloch wavevectors) define a topological polarization RT=∑iniTi\mathbf{R}_T = \sum_i n_i \mathbf{T}_i, where Ti\mathbf{T}_i are lattice vectors of a dual lattice; the number of topologically protected zero modes on a surface with outward normal G\mathbf{G} is νT=G⋅RT/2π\nu_T = \mathbf{G}\cdot\mathbf{R}_T/2\pi, the mechanical analogue of the edge-state count of the Su-Schrieffer-Heeger (SSH) model, the one-dimensional chain with alternating bond strengths whose end-state count is fixed by a winding number. They further proved an index theorem, adapted from the Callias index theorem (an index formula for Dirac operators on noncompact spaces carrying a position-dependent mass term), that fixes the signed count νS=N0(S)−Nss(S)\nu_S = N_0(S) - N_{ss}(S) of floppy modes minus states of self-stress in any subsystem SS of a larger lattice, and verified it on a two-dimensional deformed kagome model (the corner-sharing-triangle lattice) exhibiting distinct topological phases separated by gapless transitions.

That index theorem, however, assumes the bulk spectrum is gapped, apart from the acoustic zero at k=0k = 0 which is non-topological (it can be removed by a weak translational-symmetry-breaking perturbation). The paper noted that topologically protected gapless points can in principle occur - point zeros around which the phase of det⁡Q(k)\det Q(k) advances by 2π2\pi, forming the mechanical analogue of a Dirac semimetal (a gapless electronic phase with isolated conical band touchings) - but that these do not occur in the model studied. Since det⁡Q(k)=0\det Q(k) = 0 is a complex equation, such zeros are generically isolated points in a two-dimensional Brillouin zone and lines (now called Weyl lines) in three dimensions, each carrying an integer winding charge. In its closing paragraph the paper posed the future directions considered here: three-dimensional lattice models; lattices supporting point singularities in det⁡Q(k)\det Q(k); a fuller index-theoretic treatment of bulk-gapless situations; and connections with theories of frustrated magnetism, referring to Lawler's emergent-gauge description of highly frustrated magnets, in which the competition between spin interactions is described by an emergent gauge field subject to a Maxwell-like constraint counting.

Subsequent work closed the two most concrete of these directions. Rocklin, Gin-ge Chen, Falk, Vitelli and Lubensky realized the conjectured point zeros: deformed two-dimensional square Maxwell lattices host pairs of oppositely charged Weyl points carrying zero-energy bulk modes, whose existence forces a wavenumber-dependent count of topological states at free surfaces and domain walls (interfaces between regions of different topological polarization). Stenull, Kane and Lubensky constructed three-dimensional generalized pyrochlore lattices (three-dimensional networks of corner-sharing tetrahedra) whose bulk phonon spectra contain Weyl lines, defined the three-dimensional topological charge RT=∑imiTi\mathbf{R}_T = \sum_i m_i \mathbf{T}_i from winding numbers mim_i, and showed that zero surface modes flip from one edge to the opposite as a function of surface wavenumber across the projections of the Weyl lines onto the surface Brillouin zone, the space of Bloch wavevectors parallel to a given boundary. Tang and co-workers demonstrated, in theory and experiment, three-dimensional isostatic phases that are fully polarized because they are free of the Weyl lines that otherwise disrupt polarization in three dimensions, and related 2025-2026 works built omnidirectionally polarized three-dimensional metamaterials and topologically polarized Maxwell beams, the latter requiring a modified index precisely because the spectrum lacks a complete gap. Paulose, Gin-ge Chen and Vitelli generalized the counting to dislocations (topological lattice defects whose Burgers vector shifts the counting rules in a gauge-like way). Susstrunk and Huber gave a symmetry-based classification of topological phonons in generic linear mechanical metamaterials, but for ordinary elastic media rather than for Maxwell lattices with topological polarization and Weyl lines. What has not been achieved is a general and unified index theorem covering lattices whose bulk spectra contain topologically protected gapless loci and boundaries intersecting their projections, a classification of three-dimensional topological mechanical Maxwell-lattice phases, and an explicit theory linking topological Maxwell lattices to frustrated magnets.

Problem Statement

Complete the topological-mechanics program that Kane and Lubensky posed for isostatic (Maxwell) lattices, in the part that remains open after the realization of Weyl points and Weyl lines. Specifically:

(1) Establish a general index theorem for Maxwell lattices whose bulk spectra contain topologically protected gapless loci - isolated Weyl points of det⁡Q(k)\det Q(k) in two dimensions and charged Weyl lines in three dimensions - that determines the signed count of zero modes minus states of self-stress, νS=N0(S)−Nss(S)\nu_S = N_0(S) - N_{ss}(S), for a subsystem SS bounded by an arbitrary surface or domain wall, including the case in which the boundary's surface Brillouin zone (the projection of the bulk Brillouin zone onto wavevectors parallel to the boundary) is divided by the projections of the singular loci. The theorem should reduce to the known gapped-phase result (surface count G⋅RT/2π\mathbf{G}\cdot\mathbf{R}_T/2\pi) and to the existing model-specific wavenumber-dependent counts for deformed square lattices and generalized pyrochlores, and should specify precisely when and how the count changes across a projected Weyl locus.

(2) Construct the corresponding classification of three-dimensional topological mechanical Maxwell-lattice phases: enumerate the distinct phases (with and without Weyl lines, fully polarized and unpolarized, and phases distinguished by additional symmetry protection such as mirror symmetry or higher-order corner modes, i.e. corner-localized zero modes characteristic of higher-order topological phases), with explicit invariants generalizing the topological polarization RT\mathbf{R}_T to three dimensions, and determine which phase boundaries require gapless bulk spectra and which can be traversed through gapped intermediates.

(3) Formulate the correspondence between topological Maxwell lattices and frustrated magnets explicitly: identify spin-lattice-coupled frustrated models (for example kagome-lattice Mott insulators, materials in which electron motion is suppressed by Coulomb repulsion so that localized spins sit on a corner-sharing-triangle geometry whose exchange interactions cannot all be satisfied simultaneously) whose ground states or classical limits realize Maxwell lattices with nonzero topological polarization, characterize the resulting boundary floppy modes and their magnetic control, and determine whether the mechanical topological invariants can be engineered or switched magnetically.

These three components together address the originating paper's stated future directions for the bulk-gapless and three-dimensional regime. The three components are independent, and each can be pursued on its own: component (1) asks for a general index theorem for Maxwell lattices whose bulk spectra contain topologically protected gapless loci, one that reduces to the known gapped-phase result and to the existing model-specific wavenumber-dependent counts; component (2) asks for the classification of the full class of three-dimensional topological mechanical Maxwell lattices, under the symmetry protections named above, enumerated by the explicit invariants generalizing RT\mathbf{R}_T; and component (3) asks for the correspondence to be realized through an explicit spin-lattice-coupled frustrated model family, with the correspondence argued from that model family.

References

  1. C. L. Kane and T. C. Lubensky, "Topological boundary modes in isostatic lattices", Nature Physics 10, 39-45 (2013; published online 8 December 2013). DOI: 10.1038/nphys2835, https://doi.org/10.1038/nphys2835; arXiv:1308.0554, https://arxiv.org/abs/1308.0554
  2. D. Zeb Rocklin, Bryan Gin-ge Chen, Martin Falk, Vincenzo Vitelli and T. C. Lubensky, "Mechanical Weyl Modes in Topological Maxwell Lattices", Physical Review Letters 116, 135503 (2016). DOI: 10.1103/PhysRevLett.116.135503, https://doi.org/10.1103/PhysRevLett.116.135503; arXiv:1510.04970, https://arxiv.org/abs/1510.04970
  3. Olaf Stenull, C. L. Kane and T. C. Lubensky, "Topological Phonons and Weyl Lines in Three Dimensions", Physical Review Letters 117, 068001 (2016). DOI: 10.1103/PhysRevLett.117.068001, https://doi.org/10.1103/PhysRevLett.117.068001; arXiv:1606.00301, https://arxiv.org/abs/1606.00301
  4. Jayson Paulose, Bryan Gin-ge Chen and Vincenzo Vitelli, "Topological modes bound to dislocations in mechanical metamaterials", Nature Physics 11, 153-156 (2015; published online 19 January 2015). DOI: 10.1038/nphys3185, https://doi.org/10.1038/nphys3185; arXiv:1406.3323, https://arxiv.org/abs/1406.3323
  5. Zheng Tang, Fangyuan Ma, Feng Li, Yu-Gui Yao and Di Zhou, "Fully Polarized Topological Isostatic Metamaterials in Three Dimensions", Physical Review Letters 133, 106101 (2024). DOI: 10.1103/PhysRevLett.133.106101, https://doi.org/10.1103/PhysRevLett.133.106101; arXiv:2409.02607, https://arxiv.org/abs/2409.02607
  6. Shaoyuan Zhang, Xuejian Gong, Fangyuan Ma, Zheng Tang, Ying Wu, Di Zhou, Feng Li and Yu-Gui Yao, "3D topologically polarized elastic metamaterials enable asymmetric energy isolation at finite frequencies", Science Advances 12, eaec6144 (2026). DOI: 10.1126/sciadv.aec6144, https://doi.org/10.1126/sciadv.aec6144; preprint arXiv:2607.16588, https://arxiv.org/abs/2607.16588 (preprint title: "3D Topologically Polarized Elastic Metamaterials Enable Asymmetric Energy Isolation at Low Frequencies")
  7. Yi Chen, James P. McInerney, Paul N. Krause, Jonathan L. G. Schneider, Martin Wegener and Xiaoming Mao, "Observation of Floppy Flexural Modes in a 3D Polarized Maxwell Beam", Physical Review Letters 134, 086101 (2025). DOI: 10.1103/PhysRevLett.134.086101, https://doi.org/10.1103/PhysRevLett.134.086101
  8. Guido Baardink, Anton Souslov, Jayson Paulose and Vincenzo Vitelli, "Localizing softness and stress along loops in 3D topological metamaterials", Proceedings of the National Academy of Sciences 115, 489-494 (2018). DOI: 10.1073/pnas.1713826115, https://doi.org/10.1073/pnas.1713826115
  9. Sai Sanjit Ganti, Ting-Wei Liu and Fabio Semperlotti, "Weyl points and topological surface states in a three-dimensional sandwich-type elastic lattice", New Journal of Physics 22, 083001 (2020). DOI: 10.1088/1367-2630/ab9e31, https://doi.org/10.1088/1367-2630/ab9e31
  10. Hong-Hao Song, Pengwei Zhao and Gang v. Chen, "Magnetopological mechanics in Maxwell lattice frustrated Mott insulators", arXiv:2602.12168 (2026). arXiv:2602.12168, https://arxiv.org/abs/2602.12168
  11. Xiaoming Mao and Tom C. Lubensky, "Maxwell Lattices and Topological Mechanics", Annual Review of Condensed Matter Physics 9, 413-433 (2018). DOI: 10.1146/annurev-conmatphys-033117-054235, https://doi.org/10.1146/annurev-conmatphys-033117-054235
  12. Lucien Jezequel and Pierre Delplace, "Mode-shell correspondence, a unifying phase space theory in topological physics - Part I: Chiral number of zero-modes", SciPost Physics 17, 060 (2024). DOI: 10.21468/SciPostPhys.17.2.060, https://doi.org/10.21468/SciPostPhys.17.2.060; preprint arXiv:2310.05656, https://arxiv.org/abs/2310.05656 (2023)
  13. Roman Susstrunk and Sebastian D. Huber, "Classification of topological phonons in linear mechanical metamaterials", Proceedings of the National Academy of Sciences 113 (33) (2016). DOI: 10.1073/pnas.1605462113, https://doi.org/10.1073/pnas.1605462113