Brillouin Zones of Integer Lattices and Their Perturbations

Herbert Edelsbrunner, Alexey Garber, Mohadese Ghafari, Teresa Heiss, Morteza Saghafian, Mathijs WintraeckenarXiv 2022 · DOI · arXiv

TL;DR

This paper provides a rigorous geometric and combinatorial analysis of $k$-th Brillouin zones, focusing on integer lattices and their behavior under noise. The authors prove that these zones are topologically stable under small perturbations and establish key bounds on the number and size of the "chambers" that constitute them.

What Problem Does This Paper Address?

Brillouin zones are a fundamental concept in solid-state physics, originally introduced by Léon Brillouin to describe how electron waves propagate through crystal lattices. Geometrically, the $k$-th Brillouin zone of a point in a set is the region where that point is the $k$-th closest neighbor. While the first Brillouin zone is identical to the well-known Voronoi cell, higher-order zones ($k > 1$) form complex, "thickened sphere" shapes surrounding the central point.

Despite their importance in physics and materials science, many mathematical properties of higher-order Brillouin zones remained unproven, particularly for non-ideal sets. The authors address several gaps:

  1. Stability: Does a small "jiggle" in the positions of atoms (perturbation) drastically change the shape of the Brillouin zones?
  2. Complexity: How many individual geometric pieces (chambers) make up a zone?
  3. Scale: How do these chambers shrink as we look at points further away (increasing $k$)?

Methods

The researchers combine classical geometry with modern combinatorial techniques:

  • Order-$k$ Voronoi Tessellations: They leverage the fact that the $k$-th Brillouin zone is essentially the "overlay" of the $(k-1)$-st and $k$-th order Voronoi diagrams.
  • Lipschitz Continuity: To prove stability, they show that the function mapping a point to its $k$-th smallest distance is 1-Lipschitz, meaning small changes in input (point positions) lead to bounded changes in output (zone boundaries).
  • Geometric Inversion: To handle higher dimensions ($d > 2$), they use a sphere inversion trick to map the problem of counting chambers to the well-studied problem of counting "$k$-sets" in discrete geometry.
  • Computational Verification: The authors developed Python tools using rational arithmetic to precisely count chambers and measure areas in 2D lattices, ensuring that numerical errors didn't obscure the highly degenerate nature of perfect lattices.
Fig. 1: Left: the arrangement of bisectors defined by the point in the center and all other points in the integer lattice. Starting with the second, every fourth Brillouin zone is colored dark blue alternating with light blue. Middle: the 6-th Brillouin zone sandwiched between two circles centered at the point in the center. Right: the order-k Brillouin tessellation of the integer lattice obtained by overlaying the order-5 with the order-6 Voronoi tessellations or, equivalently, by drawing the 6-th Brillouin zones of all points in the integer lattice.

Datasets

The study does not use external datasets. Instead, the authors generated synthetic datasets consisting of:

  • The Integer Lattice ($\mathbb{Z}^d$): The "perfect" baseline.
  • Perturbed Lattices: Points from $\mathbb{Z}^d$ were moved by a random vector within a cube of side length $2\tau$. They categorized these as "weak" ($\tau=0.02$), "medium" ($\tau=0.10$), and "strong" ($\tau=0.50$) perturbations.

Key Insights and Ideas

  • Topological Robustness: The $k$-th Brillouin zone (for $k \ge 2$) is always homeomorphically equivalent to a sphere ($S^{d-1}$). More importantly, the authors proved that these zones are stable: as the magnitude of the perturbation goes to zero, the Hausdorff distance between the original and perturbed zone boundaries also goes to zero.
  • The "Chamber Shrinking" Paradox: In a perfect integer lattice, chambers must get smaller as $k$ increases. However, the authors discovered that even a tiny perturbation can prevent this. They proved that for any perturbation strength $\tau < 1/2$, there exists a way to move points such that a large chamber (with diameter at least $\tau$) persists for every $k$.
  • Dimensional Complexity: In 2D, the number of chambers is strictly linear ($O(k)$). In higher dimensions, the complexity is tied to the "k-set problem," one of the most famous open problems in discrete geometry, suggesting that Brillouin zones in 3D and beyond are significantly more complex than their 2D counterparts.

Key Results

  • Zone Width: For the integer lattice $\mathbb{Z}^d$, the width of any $k$-th Brillouin zone is bounded by $\sqrt{d}$, regardless of how large $k$ is.
  • Chamber Count (2D): For a 2D lattice, the number of chambers in the $k$-th zone is at most $6k-6$. Experimentally, the authors found that perturbations actually increase the number of chambers compared to the perfect lattice.
  • Chamber Diameter: For the perfect lattice, the maximum diameter of a chamber shrinks at a rate of roughly $O(k^{-1/d})$.
Figure 2: Graphs showing the minimum and maximum distances ($r_k$ and $R_k$) of the Brillouin zones from the origin. The solid lines represent the perfect lattice, while the dotted lines show the stability under strong perturbation.

Limitations and Future Work

  • Higher Dimensions: While the authors provided upper bounds for $d > 2$, they are not yet proven to be tight. They suspect the true complexity might be $O(k^{d-1})$.
  • Boundary Distortion: The authors observed a curious phenomenon where the "roughness" or distortion of the zone boundary seems to converge to a universal constant of $4/\pi$. They suggest this as a promising area for a formal proof.
  • Aperiodic Sets: The current work focuses on lattices. Extending these stability and complexity results to aperiodic sets (like Quasicrystals) or hyperuniform sets remains an open challenge.

Summary updated

Abstract

For a locally finite set, $A \subseteq \mathbb{R}^d$, the $k$-th Brillouin zone of $a \in A$ is the region of points $x \in \mathbb{R}^d$ for which $\|x-a\|$ is the $k$-th smallest among the Euclidean distances between $x$ and the points in $A$. If $A$ is a lattice, the $k$-th Brillouin zones of the points in $A$ are translates of each other, which tile space. Depending on the value of $k$, they express medium- or long-range order in the set. We study fundamental geometric and combinatorial properties of Brillouin zones, focusing on the integer lattice and its perturbations. Our results include the stability of a Brillouin zone under perturbations, a linear upper bound on the number of chambers in a zone for lattices in $\mathbb{R}^2$, and the convergence of the maximum volume of a chamber to zero for the integer lattice.
BibTeX
@misc{edelsbrunner2022brillouin,
  author = {Herbert Edelsbrunner and Alexey Garber and Mohadese Ghafari and Teresa Heiss and Morteza Saghafian and Mathijs Wintraecken},
  title = {{Brillouin Zones of Integer Lattices and Their Perturbations}},
  year = {2022},
  eprint = {2204.01077},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2204.01077},
}
Figure