Triangular Lattice

Overview

The nearest-neighbor tight-binding model on the triangular lattice is the simplest frustrated lattice hopping problem. Each site has 6 nearest neighbours, the lattice has a single sublattice per unit cell, and the resulting single band exhibits a characteristic asymmetric dispersion that is the hallmark of geometric frustration.

\[H = -t \sum_{\langle i,j \rangle} \bigl(c^\dagger_i c_j + c^\dagger_j c_i\bigr)\]

Lattice properties: 1 sublattice per unit cell, 6 nearest neighbours, not bipartite (frustrated), no flat band.

Key physics: Because the triangular lattice is non-bipartite, there is no chiral (particle-hole) symmetry, and the spectrum is not symmetric about $E = 0$. The band ranges from $-6t$ (at $\Gamma$) to $+3t$ (at the $K$-points), an asymmetric window that is a direct consequence of frustration. The density of states exhibits a Van Hove singularity within the band.


Bloch Spectrum

Statement

Since the triangular lattice has one sublattice, the Bloch "Hamiltonian" is a scalar at each $\mathbf{k}$-point:

\[E_{mn} = -2t\left[\cos\theta_1 + \cos\theta_2 + \cos(\theta_2 - \theta_1)\right]\]

where $\theta_1 = 2\pi m/L_x$ and $\theta_2 = 2\pi n/L_y$, with $\mathbf{a}_1 = (1, 0)$, $\mathbf{a}_2 = (1/2, \sqrt{3}/2)$ the primitive vectors of Lattice2D's triangular topology.

The full spectrum has $L_x L_y$ eigenvalues, one per allowed momentum.

Band edges

Point$(m/L_x, n/L_y)$Energy
$\Gamma$$(0, 0)$$-6t$ (global minimum, unique)
$K$$(1/3, 2/3)$$+3t$ (global maximum)
$K'$$(2/3, 1/3)$$+3t$ (global maximum)

The band range $[-6t, +3t]$ has total width $9t$, compared to $[-4t, +4t]$ (width $8t$) for the square lattice. The asymmetry $|E_{\min}| = 6t \neq |E_{\max}| = 3t$ is a direct manifestation of the absence of bipartite symmetry.

Van Hove singularity

At certain energies within the band, the density of states $g(E) = \sum_{\mathbf{k}} \delta(E - E(\mathbf{k}))$ diverges logarithmically due to saddle points in the dispersion $E(\mathbf{k})$. These Van Hove singularities are responsible for electronic instabilities (magnetism, superconductivity) in triangular-lattice materials.

K-point commensurability

The $K$-point eigenvalue $E = +3t$ appears in the finite-size spectrum only when both $L_x$ and $L_y$ are divisible by 3, since the $K$-point momenta $(1/3, 2/3)$ and $(2/3, 1/3)$ must be commensurate with the discrete Brillouin zone grid. When this condition is met, exactly 2 eigenvalues sit at $+3t$.

References

  • G. H. Wannier, "Antiferromagnetism. The Triangular Ising Net", Phys. Rev. 79, 357 (1950) – triangular lattice frustration.
  • T. Koretsune, M. Ogata, "Electronic structures of triangular lattice models", J. Phys. Soc. Jpn. 76, 074706 (2007) – NN tight-binding spectrum and Van Hove singularity.

QAtlas API

# Sorted single-particle spectrum, 6×6 triangular PBC
λ = QAtlas.fetch(QAtlas.Triangular(), TightBindingSpectrum(); Lx=6, Ly=6, t=1.0)
# → 36 eigenvalues, ranging from -6.0 to +3.0

Verification

Test fileMethodWhat is checked
test_triangular_tight_binding.jlReal-space ED via Lattice2D$\lambda_{\text{real}} = \lambda_{\text{Bloch}}$ for $3 \times 3$ through $6 \times 6$
test_triangular_tight_binding.jlBand edges$E_{\min} = -6t$ (unique), $E_{\max} = +3t$ (when 3
test_triangular_tight_binding.jlFrustrationSpectrum is NOT symmetric about zero
test_triangular_tight_binding.jl$K$-point degeneracy$\geq 2$ eigenvalues at $+3t$ when both $L_x, L_y \equiv 0 \pmod{3}$
test_triangular_tight_binding.jlStructural$\text{tr}\,H = 0$
test_triangular_tight_binding.jl$t$ scaling$\lambda(t) = t \cdot \lambda(1)$
test_bloch_generic.jlGeneric Bloch builderScalar formula $=$ bloch_tb_spectrum

Connections

  • Lattice family: Part of the tight-binding model catalogue. The triangular lattice is frustrated (non-bipartite) like Kagome, but has only one sublattice and no flat band.
  • Bipartite contrast: The square lattice (also 1 sublattice) has a symmetric band $[-4t, +4t]$ because it is bipartite. The triangular band $[-6t, +3t]$ breaks this symmetry.
  • Magnetic frustration: The non-bipartite structure underlies the classical Ising antiferromagnet frustration problem on the triangular lattice (Wannier 1950).
  • Methods: Computed via the Bloch Hamiltonian method.