Teo, Fu and Kane (Phys. Rev. B 78, 045426 (2008)) analyzed the surface states of the first experimentally realized three-dimensional topological insulator, the alloy Bi1−x_{1-x}Sbx_x, and identified a set of unresolved issues: the sign of the mirror chirality η\eta of bismuth, its experimental determination through the mirror Chern number nMn_{\mathcal M}, the role of self-consistent surface potentials, a reparametrized tight-binding model, a topological classification incorporating the full point group, and robustness against symmetry-breaking disorder. Most of these questions have since been settled: spin-resolved ARPES determined η=−1\eta=-1 and nM=−1n_{\mathcal M}=-1, the point-group classification was completed by topological crystalline insulator theory and topological quantum chemistry, and the disorder question was answered through statistical-symmetry classifications. What remains is a coherent quantitative-modeling problem: construct a tight-binding parametrization of the Liu-Allen type, solvable in a semi-infinite geometry, that reproduces both the band energies and the correct mirror chirality η=−1\eta=-1, together with a surface treatment including self-consistent Hartree and exchange surface potentials and surface relaxation, in order to predict quantitatively the Bi1−x_{1-x}Sbx_x (111) surface-state energies, additional surface bands split off from the bulk continuum, and the placement of the mirror-protected Dirac crossings.

Background

A three-dimensional topological insulator is a bulk insulator whose band structure is characterized by four Z2\mathbb{Z}_2 invariants that are determined by time-reversal symmetry; the invariant ν0\nu_0 distinguishes the strong topological insulator, which must carry gapless, helically spin-polarized surface states. The semiconducting alloy Bi1−x_{1-x}Sbx_x (bismuth diluted with antimony, crystallizing in the rhombohedral A7 structure with space group R3ˉmR\bar{3}m) was the first such material identified: the parity-eigenvalue method of Fu and Kane (Phys. Rev. B 76, 045302 (2007)), which determines the Z2\mathbb{Z}_2 invariants of an inversion-symmetric crystal from the parity eigenvalues of the occupied bands at the time-reversal-invariant momenta, together with angle-resolved photoemission spectroscopy (ARPES, which maps the energy and momentum of emitted electrons and thereby the surface band structure) established it as a strong topological insulator with invariants (1;111)(1;111).

Teo, Fu and Kane (2008) studied its (111) surface states with a phenomenological tight-binding model built from the Liu-Allen parametrization of Bi and Sb (Y. Liu and R. E. Allen, Phys. Rev. B 52, 1566 (1995)), a sparse tight-binding Hamiltonian with spin-orbit coupling that can be solved in a semi-infinite geometry (a crystal truncated at a surface and extended infinitely into the bulk), so that surface states can be resolved near the small bulk gap. Beyond the Z2\mathbb{Z}_2 structure, they showed that spatial symmetries impose additional, model-independent constraints. For crystals with inversion symmetry, the parity eigenvalues of the occupied bands at the eight time-reversal-invariant momenta (TRIM, momenta equal to half a reciprocal lattice vector modulo a reciprocal lattice vector, invariant under time reversal) define parity invariants that fix the 'surface fermion parity', i.e. which surface TRIM are enclosed by an odd number of electron or hole Fermi-surface lines. Mirror symmetry (reflection in a plane, here the xx-zz plane combined with inversion) allows the Bloch Hamiltonian on the mirror-invariant plane to be block-diagonalized into states of opposite mirror eigenvalue; the difference of the Chern numbers of the two blocks, nM=(ni+−ni−)/2,n_{\mathcal M}=(n_i^+ - n_i^-)/2, defines an integer mirror Chern number, which is nonzero in the topological phase of Bi1−x_{1-x}Sbx_x and forces gapless surface states crossing the gap within the surface-projected mirror plane. Which surface band connects the conduction and valence bands across the gap is fixed by the sign of a Z2\mathbb{Z}_2 parameter η\eta in the k⋅p\mathbf k\cdot\mathbf p theory (an effective Hamiltonian expanded to low order about a high-symmetry point) of the L point of pure Bi (L is a high-symmetry point of the rhombohedral Brillouin zone where the valence and conduction bands of Bi nearly touch), which Teo, Fu and Kane called the mirror chirality: in their convention ΔnM=−η\Delta n_{\mathcal M}=-\eta across the gap closing, so the anomalous sign η=−1\eta=-1 means the Σ2\Sigma_2 surface band rather than the Σ1\Sigma_1 band (these are the two surface bands along the Γˉ\bar\Gamma–Mˉ\bar M mirror line, carrying mirror eigenvalues +i+i and −i-i respectively) connects the bands, and the hole pockets of the (111) surface enclose no Dirac point.

The paper found a conflict: the Liu-Allen tight-binding model predicts the conventional sign η=+1\eta=+1 (Dirac points inside the hole pockets), whereas the pseudopotential band calculation of Falicov and Golin and first-principles surface-state calculations of Bi(111) indicate η=−1\eta=-1; the indirect experimental evidence (spin-resolved ARPES on Bi) supported η=−1\eta=-1. Teo, Fu and Kane therefore posed these issues. The two components that later literature has left open are: the calculation 'makes no attempt to self consistently describe the potential near the surface' — bond relaxation, and the Hartree and exchange potential of the surface charge and dipole layer, whose 'effect... will be to modify the energies of the bands and perhaps to split off additional surface state bands from the continuum' — and 'it should be possible to come up with a new parameterization of the Liu Allen model which would have η=−1\eta=-1', since that model 'was designed only to get the energies of the bands right'.

The intervening literature has fixed the topological facts that motivate these two requests. Spin-resolved ARPES (photoemission with measurement of the electron spin, SARPES) on Bi1−x_{1-x}Sbx_x by Hsieh et al. (Science 323, 919 (2009)) measured the surface spin texture and Berry phase and determined the mirror Chern number nM=−1n_{\mathcal M}=-1; Nishide et al. (Phys. Rev. B 81, 041309(R) (2010)) mapped the spin-polarized surface bands between Γˉ\bar\Gamma and Mˉ\bar{M} and determined the mirror chirality to be −1-1, 'which agrees with the theoretical prediction based on first-principles calculations'. So the target sign for a faithful model is now known from experiment, not merely inferred. What remains missing is the quantitative apparatus: the existing first-principles-derived calculations for the alloy (Zhang et al., Phys. Rev. B 80, 085307 (2009)) use maximally localized Wannier functions (an exact tight-binding representation of a first-principles band structure) within a virtual-crystal approximation (VCA, replacing the mixed Bi/Sb sites by an average atom, implemented by tuning the spin-orbit coupling strength), and compute semi-infinite surface spectra, but do not isolate or quantify the self-consistent Hartree/exchange surface-dipole potential or the effect of surface relaxation; and the Liu-Allen parametrization itself is still used in its original form (e.g. in studies of Bi/BiSb heterojunctions by Asaka et al., Phys. Rev. B 106, 245303 (2022)). Meanwhile, for pure Bi(111), Koie, Yaguchi and Fuseya (Phys. Rev. B 111, L201303 (2025)) showed from density functional theory that substantial near-surface inter-bilayer expansion, fed into a relativistic empirical tight-binding model, can 'topologically block' the surface states — evidence that surface relaxation and surface potentials are not a small correction in this material family but can qualitatively mask the bulk-boundary correspondence that a surface model is supposed to exhibit.

Problem Statement

Construct a quantitatively accurate and topologically faithful description of the (111) surface states of the topological-insulator alloy Bi1−x_{1-x}Sbx_x, in the sense specified by Teo, Fu and Kane (2008), now that the mirror chirality is experimentally known to be η=−1\eta=-1:

(1) A tight-binding parametrization of the Liu-Allen type (sparse, spin-orbit-coupled, solvable in a semi-infinite geometry — a crystal truncated at a surface and extended infinitely into the bulk — by transfer-matrix or surface Green-function methods, the standard techniques for extracting the surface spectrum of such a half-space Hamiltonian) that reproduces not only the bulk band energies of Bi/Bi1−x_{1-x}Sbx_x but also the correct mirror-chirality sign η=−1\eta=-1, i.e. the mirror Chern number nM=−1n_{\mathcal M}=-1 of the topological phase and the corresponding band connectivity in which the Σ2\Sigma_2 surface band (the mirror-line surface band of mirror eigenvalue −i-i) connects the conduction and valence bands across the surface gap and the hole pockets enclose no Dirac point.

(2) A treatment of the surface region that includes the self-consistent surface potential — the Hartree and exchange-correlation potential generated by the surface charge distribution and dipole layer, together with the relaxation of the near-surface bilayers — so as to predict quantitatively, band-by-band against the measured spin-resolved ARPES dispersions, the surface-state energies, the splitting-off of additional surface-state bands from the bulk continuum (in particular the surface Kramers pair, a pair of states degenerate by time-reversal symmetry, observed near Mˉ\bar{M} in ARPES but absent from the non-self-consistent calculation), and the placement of the mirror-protected crossings relative to the Fermi energy.

Both components are required and serve one objective: a validated model of the Bi1−x_{1-x}Sbx_x(111) surface spectrum that (1) reproduces the mirror-topological structure established experimentally — the mirror chirality η=−1\eta=-1, the nM=−1n_{\mathcal M}=-1 band connectivity, and the absence of a Dirac point inside the hole pockets — and (2) predicts the surface-state energies, the split-off surface bands and the placement of the mirror-protected crossings in quantitative band-by-band agreement, within the energy resolution of the measurements, with the spin-resolved ARPES dispersions of the alloy reported at the measured composition (Nishide et al., x=0.12x=0.12-0.130.13; Hsieh et al.), the unoccupied part of the spectrum being compared against pump-probe spin-resolved photoemission (Fukushima et al., Ref. 22) or inverse photoemission.

References

  1. J. C. Y. Teo, L. Fu, and C. L. Kane, "Surface states and topological invariants in three-dimensional topological insulators: Application to Bi1-xSbx", Phys. Rev. B 78, 045426 (2008). DOI: 10.1103/PhysRevB.78.045426; arXiv:0804.2664; https://arxiv.org/abs/0804.2664
  2. Y. Liu and R. E. Allen, "Electronic structure of the semimetals Bi and Sb", Phys. Rev. B 52, 1566 (1995). DOI: 10.1103/PhysRevB.52.1566; https://doi.org/10.1103/PhysRevB.52.1566
  3. L. M. Falicov and S. Golin, "Electronic Band Structure of Arsenic. I. Pseudopotential Approach", Phys. Rev. 137, A871 (1965). DOI: 10.1103/PhysRev.137.A871; https://doi.org/10.1103/PhysRev.137.A871
  4. L. Fu and C. L. Kane, "Topological insulators with inversion symmetry", Phys. Rev. B 76, 045302 (2007). DOI: 10.1103/PhysRevB.76.045302; arXiv:cond-mat/0611341; https://arxiv.org/abs/cond-mat/0611341
  5. D. Hsieh, D. Qian, L. Wray, Y. Xia, Y. S. Hor, R. J. Cava, and M. Z. Hasan, "A topological Dirac insulator in a quantum spin Hall phase", Nature 452, 970 (2008). DOI: 10.1038/nature06843; https://doi.org/10.1038/nature06843
  6. D. Hsieh, Y. Xia, L. Wray, D. Qian, A. Pal, J. H. Dil, F. Meier, J. Osterwalder, G. Bihlmayer, C. L. Kane, Y. S. Hor, R. J. Cava, and M. Z. Hasan, "Observation of Unconventional Quantum Spin Textures in Topological Insulators", Science 323, 919 (2009). DOI: 10.1126/science.1167733; arXiv:0902.2617; https://arxiv.org/abs/0902.2617
  7. H.-J. Zhang, C.-X. Liu, X.-L. Qi, X.-Y. Deng, X. Dai, S.-C. Zhang, and Z. Fang, "Electronic structures and surface states of the topological insulator Bi1-xSbx", Phys. Rev. B 80, 085307 (2009). DOI: 10.1103/PhysRevB.80.085307; arXiv:0901.2762; https://arxiv.org/abs/0901.2762
  8. A. Nishide, A. A. Taskin, Y. Takeichi, T. Okuda, A. Kakizaki, T. Hirahara, K. Nakatsuji, F. Komori, Y. Ando, and I. Matsuda, "Direct mapping of the spin-filtered surface bands of a three-dimensional quantum spin Hall insulator", Phys. Rev. B 81, 041309(R) (2010). DOI: 10.1103/PhysRevB.81.041309; arXiv:0902.2251; https://arxiv.org/abs/0902.2251
  9. Y. Ohtsubo, L. Perfetti, M. O. Goerbig, P. Le Fevre, F. Bertran, and A. Taleb-Ibrahimi, "Non-trivial surface-band dispersion on Bi(111)", New J. Phys. 15, 033041 (2013). DOI: 10.1088/1367-2630/15/3/033041; https://doi.org/10.1088/1367-2630/15/3/033041
  10. L. Fu, "Topological Crystalline Insulators", Phys. Rev. Lett. 106, 106802 (2011). DOI: 10.1103/PhysRevLett.106.106802; arXiv:1010.1802; https://arxiv.org/abs/1010.1802
  11. K. Shiozaki and M. Sato, "Topology of crystalline insulators and superconductors", Phys. Rev. B 90, 165114 (2014). DOI: 10.1103/PhysRevB.90.165114; arXiv:1403.3331; https://arxiv.org/abs/1403.3331
  12. K. Shiozaki, M. Sato, and K. Gomi, "Topological Crystalline Materials - General Formulation, Module Structure, and Wallpaper Groups -", Phys. Rev. B 95, 235425 (2017). DOI: 10.1103/PhysRevB.95.235425; arXiv:1701.08725; https://arxiv.org/abs/1701.08725
  13. B. Bradlyn, L. Elcoro, J. Cano, M. G. Vergniory, Z. Wang, C. Felser, M. I. Aroyo, and B. A. Bernevig, "Topological quantum chemistry", Nature 547, 298 (2017). DOI: 10.1038/nature23268; arXiv:1703.02050; https://arxiv.org/abs/1703.02050
  14. M. G. Vergniory, L. Elcoro, C. Felser, N. Regnault, B. Andrei Bernevig, and Z. Wang, "A complete catalogue of high-quality topological materials", Nature 566, 480 (2019). DOI: 10.1038/s41586-019-0954-4; https://doi.org/10.1038/s41586-019-0954-4
  15. I. C. Fulga, B. van Heck, J. M. Edge, and A. R. Akhmerov, "Statistical topological insulators", Phys. Rev. B 89, 155424 (2014). DOI: 10.1103/PhysRevB.89.155424; arXiv:1212.6191; https://arxiv.org/abs/1212.6191
  16. B. de Leeuw, C. Kuppersbusch, V. Juricic, and L. Fritz, "Stability of the X-Y-phase of the two-dimensional C4 point group insulator", Phys. Rev. B 91, 235430 (2015). DOI: 10.1103/PhysRevB.91.235430; arXiv:1411.0255; https://arxiv.org/abs/1411.0255
  17. A. Y. Chaou, M. Moreno-Gonzalez, A. Altland, and P. W. Brouwer, "Disordered topological crystalline phases", Phys. Rev. B 112, 035167 (2025). DOI: 10.1103/fpc3-bsz6; arXiv:2412.01883; https://arxiv.org/abs/2412.01883
  18. F. Schindler, Z. Wang, M. G. Vergniory, A. M. Cook, A. Murani, S. Sengupta, A. Yu. Kasumov, R. Deblock, S. Jeon, I. Drozdov, H. Bouchiat, S. Gueron, A. Yazdani, B. A. Bernevig, and T. Neupert, "Higher-order topology in bismuth", Nature Physics 14, 918 (2018). DOI: 10.1038/s41567-018-0224-7; arXiv:1802.02585; https://arxiv.org/abs/1802.02585
  19. A. K. Nayak, J. Reiner, R. Queiroz, H. Fu, C. Shekhar, B. Yan, C. Felser, N. Avraham, and H. Beidenkopf, "Resolving the topological classification of bismuth with topological defects", Science Advances 5, eaax6996 (2019). DOI: 10.1126/sciadv.aax6996; arXiv:1903.00880; https://arxiv.org/abs/1903.00880
  20. C.-H. Hsu, X. Zhou, T.-R. Chang, Q. Ma, N. Gedik, A. Bansil, S.-Y. Xu, H. Lin, and L. Fu, "Topology on a new facet of bismuth", Proc. Natl. Acad. Sci. U.S.A. 116, 13255 (2019). DOI: 10.1073/pnas.1900527116; https://doi.org/10.1073/pnas.1900527116
  21. L. Aggarwal, P. Zhu, T. L. Hughes, and V. Madhavan, "Evidence for higher order topology in Bi and Bi0.92Sb0.08", Nature Communications 12, 4420 (2021). DOI: 10.1038/s41467-021-24683-8; arXiv:2107.00698; https://arxiv.org/abs/2107.00698
  22. Y. Fukushima, K. Kawaguchi, K. Kuroda, M. Ochi, M. Hirayama, R. Mori, H. Tanaka, A. Harasawa, T. Iimori, Z. Zhao, S. Tani, K. Yaji, S. Shin, F. Komori, Y. Kobayashi, and T. Kondo, "Spin-polarized saddle points in the topological surface states of elemental bismuth revealed by pump-probe spin- and angle-resolved photoemission spectroscopy", Phys. Rev. B 110, L041401 (2024). DOI: 10.1103/PhysRevB.110.L041401; arXiv:2303.17816; https://arxiv.org/abs/2303.17816
  23. Y. Asaka, T. Kikuchi, and Y. Fuseya, "Long-range permeation of wave function and superficial surface state due to strong quantum size effects in topological Bi/BiSb heterojunctions", Phys. Rev. B 106, 245303 (2022). DOI: 10.1103/PhysRevB.106.245303; arXiv:2211.16017; https://arxiv.org/abs/2211.16017
  24. K. Koie, R. Yaguchi, and Y. Fuseya, "Topological blocking at the Bi(111) surface due to surface relaxation", Phys. Rev. B 111, L201303 (2025). DOI: 10.1103/PhysRevB.111.L201303; arXiv:2503.16819; https://arxiv.org/abs/2503.16819